The data in this section are from the NSFís Scientists and Engineers Statistical data system (SESTAT), which is a unified database primarily containing information on the employment, education, and demographic characteristics of individuals with S&E degrees in the United States. (See NSF 1999f.),
Estimates of the size of the U.S. S&E labor force can vary dramatically depending on what criteria are used to define a scientist or engineer. (See the sidebar, "Who Is a Scientist or Engineer?") Educational degree levels and fields, occupational categories, or a combination of these factors may all be taken into account. In 1997, more than 12.5 million people in the United States either held degrees in science or engineering or were working as scientists or engineers. (See appendix table 3-1.) The number of individuals holding a college degree in an S&E field in 1997 exceeded by a large margin the number of persons working in an S&E occupation because many S&E degree holders were not working in an S&E field. Numerous individuals were also working in S&E occupations who were educated in fields not considered science or engineering related.
There are many different definitions that can be use to classify a scientist or engineernone of which are perfect. For a more thorough discussion of these complexities, see SESTAT and NIOEM: Two Federal Databases Provide Complementary Information on the Science and Technology Labor Force (NSF 1999e) and "Counting the S&E WorkforceItís Not that Easy" (NSF 1999d). Different definitions are used at different places for different analytic purposes in this report, and even more are used in reports elsewhere. These are the three major definitions used in this report:
[*] In most collections of occupation data (SESTAT data mostly does not have this problem), the generic classification of post-secondary teacher also masks many university professors who should be included in most concepts of the S&E workforce.
 Those without U.S. S&E degrees are included in 1997 SESTAT data to the extent they were in the United States in 1990, 1993, 1995, and 1997 (in the case of individuals with foreign S&E degrees) or had at least a bachelorís degree in some field and were working in a S&E occupation in 1993, 1995, and 1997.
Including those either with science or engineering degrees or in science or engineering occupations, approximately 12.5 million scientists and engineers were residing in the United States as of April 1997. Only 84 percent (10.6 million) of these individuals, however, were in the workforce. (See appendix table 3-1.) The remainder were either unemployed, but seeking work (193,700), or were not in the labor force (1.75 million). Of the 10.6 million employed, the vast majority (10.1 million) held at least one college degree in a science or engineering field. About 30 percent (3.1 million) of the 10.1 million S&E degree holders in the workforce were also employed in S&E occupations. (See text table 3-1.)
Many of the Nationís scientists and engineers hold either multiple S&E degrees or have degrees in both S&E and non-S&E fields. Many S&E-educated workers also routinely find S&E-related employment in occupations not included in traditional S&E taxonomies. Of the 10.1 million S&E degree holders in the workforce in 1997, about three-fourths (7.7 million) reported that their highest degree was in an S&E field. (See appendix table 3-2.) Many of these individuals (4.9 million), however, were not principally employed in a traditional science or engineering occupation.
The likelihood of an S&E degree holder occupying an S&E job varies by field of degree. For example, about two-thirds (66 percent) of S&E degree holders whose highest degrees were in engineering fields were employed in an S&E job in 1997. However, most of the S&E degree holders who received their highest degrees in life science or social science fields (73 percent and 86 percent, respectively) were working in occupations outside the traditional S&E taxonomy, that is, "non-S&E occupations." (See appendix table 3-2.) About half of those with highest degrees in computer and mathematical sciences and physical sciences (51 percent and 46 percent, respectively) were also employed in a non-S&E occupation in 1997.
The fact that most S&E degree holders do not work in a strictly defined science or engineering occupation does not mean that they are not using their S&E training. Of the 4.9 million S&E degree holders working in non-S&E jobs in 1997, about 65 percent indicated they were working in jobs at least somewhat related to their highest S&E degree field. (See text table 3-2.) Over three-fourths of those with highest degrees in computer and mathematical sciences who were employed in non-S&E jobs were doing work related to their degrees, compared to 61 percent of those whose highest degrees were in social and physical sciences.
Out of all employed individuals whose highest degree was in S&E, 74.8 percent said that their jobs were related to the field of their highest degree, and 44.8 percent said their jobs were closely related to their field. This can be seen in appendix tables 3-3 and 3-4. The relatedness of a field of study to an individualís job differs in mostly predictable ways across level of degree, years since degree, and field of degree.
Figure 3-1 shows the percentage of employed S&E degree holders who say their jobs are closely related to their degrees by degree level and years since degree. For the period of one to five years after receiving their degree, 74.1 percent of S&E doctorates say their jobs are closely related to their field of degree, compared to 65.9 percent of those with masterís degrees and 41.1 percent of those with bachelorís degrees. This relative ordering of relatedness by level of degree holds across all years since receipt of degree. At every degree level, however, jobs generally become less closely related to field of degree as year since degree increases. There may be many reasons for thisindividuals change their career interests over time, they may gain skills in a different area while on the job, they may move into management responsibilities, or some of their original college training may become obsolete. Given all of these possibilities, the career cycle decline in the relevance of an S&E degree is fairly modest.
Differences in proportion for those who said their jobs were closely related to their field of degree are shown in Figure 3-2 for bachelorís degree holders by major groups of S&E disciplines. At one to five years after receipt of degree, the percentage of S&E bachelorís degree holders who said their jobs were closely related to field of degree ranged from 29 percent in the social sciences to 72 percent in computer science. Between the extremes of social sciences and computer sciences, most other S&E fields have similar percentages of recent graduates in closely related jobs53 percent for physical sciences, mathematical sciences, and engineering, and 45 percent for the life sciences.
A little over half of the 4.9 million S&E degree holders working outside S&E occupations in 1997 were employed in either management-administration occupations (29 percent), sales and marketing jobs (16 percent) or non-S&E related teaching positions (9 percent) in 1997. (See text table 3-3.) Almost 90 percent of those employed as non-S&E teachers said that their work was at least somewhat related to their S&E degree field, compared to 71 percent of managers-administrators and 47 percent of those employed in sales and marketing jobs.
About 82 percent of the 4.9 million S&E degree holders not working in S&E occupations in 1997 reported their highest degree as a bachelorís degree, while 15 percent listed a masterís degree and 3 percent a doctorate. Approximately three-fifths of bachelorís degree holders reported that their jobs were closely related to their highest degree field, compared to four-fifths of both doctorate and masterís S&E degree recipients.
Of the 7.7 million scientists and engineers in the workforce in 1997 whose highest degrees were in an S&E field, a little more than a third (2.84 million) were principally employed in S&E jobs. Additionally, there were 234,000 individuals with S&E degrees whose highest degrees were in a non-S&E field who were also employed in S&E occupations. There were also 294,600 college-educated individuals employed in S&E occupations that held no degrees in an S&E field.
Altogether, approximately 3.4 million individuals were employed in an S&E occupation in 1997. (See appendix table 3-5.) Engineers represented 41 percent (1.37 million) of the S&E positions, followed by computer and mathematical scientists with 31 percent (1.04 million) of the total. Physical scientists accounted for less than 10 percent of those working in S&E occupations in 1997. By subfield, electrical engineers made up about one-fourth (365,000) of all those employed as engineers, while biological scientists accounted for a little over one-half (182,000) of the employment in the life sciences. In the physical and social science occupations, chemists (120,000) and psychologists (182,000) were the largest occupational subfields, respectively.
Almost 57 percent of the individuals employed in S&E jobs reported their highest degree type as a bachelorís degree, while 29 percent listed a masterís degree and 14 percent a doctorate. Other first professional degrees were reported as the highest degree type by about 1 percent. Almost half of those with bachelorís degrees were employed as engineers. (See text table 3-4.) Another 35 percent of bachelorís degree holders had jobs as computer and mathematical scientists. These occupations were also the most prevalent among those with masterís degrees (39 percent and 31 percent, respectively). Most doctorate holders were employed as social scientists (27 percent), life scientists (25 percent) and physical scientists (18 percent). (See the sidebar, "How Important Is Temporary Work for Scientists and Engineers?") (See also the sidebar, "Data on Recent Ph.D. Recipients in Professional Society Data.")
One common form of flexible work arrangement in the general labor force is the temporary help firm. Although best known as a way for businesses to hire temporary clerical help, major temporary help firms have long included scientists, engineers, and technicians among the workers whom they offer to other businesses on a temporary basis. How important is temporary work as a source of employment for those with S&E degrees? The answer appears to be "not very" for most S&E degree holders. Figure 3-3 shows the percentage of S&E degree holders who in 1997 reported being employed by a temporary help or employment agency. The greatest use of temporary firms occurs for those within just one to five years since receipt of their degrees (1.7 percent) and for those with more than 35 years since receipt of their degrees (1.6 percent). Only about one-third of those with temporary agency jobs are employed in S&E occupations. Ph.D. recipients are less likely than those with other S&E degrees to work for a temporary agencyonly 0.4 percent even within one to five years since receipt of degree.
Of the approximately 3.5 million scientists and engineers in the labor force in 1997, only 1.5 percent (52,900) were unemployed. (See figure 3-4.) This compares with 4.9 percent for the U.S. labor force as a whole in 1997 and 2.0 percent for all professional specialty workers. The highest unemployment rates were for life scientists (2.2 percent) and the lowest for social scientists (1.0 percent). By degree level, 1.6 percent of the scientists and engineers whose highest degree was a bachelorís degree were unemployed, compared to 1.4 percent of those with masterís degrees or a doctorate. It should be remembered, however, that the unemployment rate is a poor indicator of labor market conditions for highly educated workersit does not measure how well their employment uses their training.
The private for-profit sector is by far the largest employer of S&E workers. In 1997, 73 percent of scientists and engineers with bachelorís degrees and 60 percent of those with masterís degrees were employed in a private, for-profit company. (See appendix table 3-6.) The academic sector was the largest sector of employment for those with doctorates (49 percent). Sectors employing smaller numbers of S&E workers include educational institutions other than four-year colleges and universities, nonprofit organizations, and state or local government agencies.
Among S&E occupations, there was a wide variation in the proportions of scientists and engineers employed in private for-profit industry. While nearly three-fourths of both computer and mathematical scientists and engineers were employed in this sector, only one-fourth of life scientists and one-fifth of social scientists were so employed in 1997. Educational institutions employed the largest proportion of life scientists (48 percent) and social scientists (45 percent).
In 1997 the median annual salary of bachelorís degree holders employed in S&E occupations was $52,000; for masterís recipients it was $59,000 and for doctorate holders $62,000. (See figure 3-5 and appendix table 3-7.) Engineers commanded the highest salaries at each degree level. The second highest salaries were earned by computer and mathematical scientists at the bachelorís and masterís levels, and physical scientists and computer and mathematical scientists at the doctorate level. The lowest median salaries were reported for social scientists at each degree level.
Median salaries for scientists and engineers were higher for those with more years since completion of their highest degree. For example, individuals who earned their bachelorís or masterís degrees five to nine years ago earned about $12,000 and $8,000 less, respectively, in 1997 than those who received these degrees 15Ė19 years ago. For doctorate holders, the difference between the two groups in terms of years since receipt of degree was $14,000. (See appendix table 3-8.)
Although individuals with an S&E education can use that knowledge in a great many other waysfor example, teaching, writing, evaluating, and testingthere is a special interest in those engaged in research and development (R&D). Figure 3-6 shows the distribution of individuals with S&E degrees who reported R&D as a major work activity by level of degree. Those with doctorates comprise only 5.6 percent of all with S&E degrees, but 13.0 percent of those reporting major R&D activities. Despite this, a majority of the S&E degree holders that report major R&D activities have only bachelorís degrees (55.5 percent). Another 28.5 percent have masterís degrees, and 2.9 percent have professional degrees (mostly in medicine). Figure 3-7 shows the distribution of individuals with S&E degrees who reported R&D as a major work activity by field of highest degree. Those whose highest degree is in engineering constitute more than one-third (34.9 percent) of those reporting major R&D work activities. Notably, 13.0 percent do not have their highest degree in an S&E field. In most cases, this is a person with an S&E bachelorís degree and a higher degree in a professional field, such as business, medicine, or law.
The involvement of S&E Ph.D. recipients in R&D as a major work activity is shown by field of degree and years since receipt of Ph.D. in figure 3-8. The highest R&D rates over the career cycle are found in the physical S&E. The lowest R&D rates are in the social sciences. While the percentage of employed Ph.D. recipients with R&D as a major work activity does decline with years since degree, it remains above 50 percent in most fields. A steeper decline might have been expected, which may reflect a normal career process of movement into management or into other career interests.