Skip all navigation and go to page content

Chapter 3. Science and Engineering Labor Force

S&E Workers in the Economy

This section profiles how the S&E labor force is distributed across employment sectors in the U.S. economy. It shows that members of the S&E labor force work in all sectors, including for-profit businesses, nonprofit organizations, educational institutions, and government. The section begins with a brief description of patterns and trends in the proportions of the S&E labor force in these different employment sectors, and in the characteristics of organizations that employ S&E workers. The section looks at employment patterns in sectors and industries that have unusually high concentrations of S&E workers and variations among employers of different sizes. It then closes with a brief presentation of data on geographical areas with major concentrations of S&E workers. This includes data both on areas where workers in S&E occupations constitute a large percentage of the labor force and areas where large numbers of workers in these occupations are geographically concentrated.

The section then analyzes S&E employment in the different economic sectors. In the business/industry sector, it describes differences between for-profit and nonprofit organizations and in the proportion of S&E workers by industrial sector. The section also examines self-employed workers with S&E degrees and in S&E occupations. Throughout the section, the analysis distinguishes between employment sectors for individuals with S&E degrees and for those working in S&E occupations.

A brief analysis of the education sector, including all levels of education at both public and private institutions, and the government sector follows. In light of specialized scientific missions and the scope of scientific activities supported by the U.S. government, this section focuses on federal employment.

The S&E labor force is often seen as a major contributor to innovation. The section concludes, therefore, with data on various activities associated with innovation, such as performing R&D, patenting, and enhancing knowledge and skills through work-related training. This includes a description of data on job changes among S&E workers, which enable them to apply work-related learning in new contexts and may thereby spur innovation.

Characteristics of Employers of Scientists and Engineers

Employment Sector

In general, the labor market is divided into workers in the public sector and those in the private sector. This classification works awkwardly for analysis of the S&E labor force. Because educational institutions are significant employers of scientists and engineers in the United States, these institutions are better treated as a distinct sector, which spans public and private institutions and includes 4- and 2-year colleges and universities and precollege institutions. Employees in the business/industry sector work in for-profit businesses and nonprofit organizations, as well as being self-employed. The government sector includes local, state, and federal employees.

The S&E workforce includes both those working in S&E occupations and those trained in S&E fields. In 2008, approximately 70% of individuals trained or working in S&E worked in the business/industry sector, 12% in the government sector, and 18% in the education sector. This distribution has stayed relatively stable since the early 1990s (see figure 3-12), with some minor shifts. Although the overall percentage of scientists and engineers working in educational institutions has stayed at approximately 18% of overall employment, the relative proportion working in 4-year institutions versus other educational institutions has changed from about 50/50 in 1993 to 40/60 in 2008. Compared with 1993, a smaller proportion of scientists and engineers are working in the federal government in 2008 (6.4% versus 4.5%). The largest change has been within the nonprofit sector. In 1993, the proportion working in this sector was 5.8%; by 2008, it was 10.4%, an 80% increase.

The different sectors in which scientists and engineers are employed are shown in table 3-6. The sector distributions of scientists and engineers by highest degree in S&E versus any degree in S&E are very similar, and mirror the distributions found among all employed S&Es. Workers in different broad occupational categories are concentrated in different employment sectors. Four-year educational institutions, for example, employ a higher percentage of workers in S&E occupations than other institutions in the education sector. A larger proportion of S&E-related workers are employed in nonprofit organizations, compared to those in S&E or non-S&E occupations.

Employer Size

Employer size can affect the breadth and depth of S&E employment concentration. Educational institutions and government entities that employ scientists and engineers are, primarily, larger employers. A large majority of these organizations have 100 or more employees (88% in the education sector, 91% in the government sector). Scientists and engineers working in the business/industry sector are more broadly distributed across firms of many sizes.

S&E degree holders who work in for-profit businesses are distributed particularly broadly. Moreover, within the business/industry sector, workers at different degree levels are distributed similarly across firms of different sizes (figure 3-13). Companies with fewer than 100 employees, for example, employ 36% of S&E highest degree holders who work in the business/industry sector, ranging from 32% of master's degree holders to 38% of doctorate holders. S&E doctorate holders in this sector, however, are concentrated at very small and very large firms. Some 23% work at the smallest firms (under 10 employees), but the proportion of them at firms with fewer than 500 employees is similar to that among S&E highest degree holders generally. At the other end of the spectrum, close to 20% of doctorate holders work at firms of 25,000 or more employees.

The distribution of employees in the business/industry sector in S&E occupations, however, shows a different pattern. Among this group, there is a greater concentration of employment in firms with more than 5,000 employees (44%), compared to those in smaller firms of 100 employees or fewer (25%).

S&E Occupation Density by Type of Industry

Industries vary in their proportions of S&E workers (table 3-6). The OES survey provides detailed estimates for employment by type of industry, although it excludes the self-employed and those employed in recent startups. OES classifies the government sector within the broad category "government," and educational institutions within the broad category of "educational services." In the for-profit sector, the industry with the highest percentage of S&E workers was "professional, scientific, and technical services" with 29%, followed by information with 16% (figure 3-14). The government (federal, state, and local) had 6% and the educational services sector had 5% of total employment in S&E occupations in 2010.

In 2010, slightly more than 1 million workers in S&E jobs were employed in industries whose S&E employment component was less than the national average of 4.4% (table 3-7). These industries employ 75% of all workers and 21% of all workers in S&E occupations. Examples include local government (at 3.0%, with 165,960 S&E jobs), hospitals (at 1.5%, with 77,890 S&E jobs), and plastic parts manufacturers (at 2.6%, with 13,000 S&E jobs).

Industries with higher proportions of individuals in S&E occupations tend to pay higher average salaries to both their S&E and non-S&E workers (table 3-7). The average salary of workers in non-S&E occupations employed in industries where more than 40% of workers are in S&E occupations is nearly double the average salary of workers in non-S&E occupations in industries with below-average proportions of workers in S&E occupations ($79,540 versus $29,970).

S&E Workers by Metropolitan Area

The availability of highly skilled workers can affect an area's economic competitiveness and its ability to attract business investment. The federal government uses standard definitions to describe geographical regions in the United States for comparative purposes. It designates very large metropolitan areas, sometimes dividing them into smaller metropolitan divisions that can also be substantial in size (Office of Management and Budget 2009).

Two measures indicate availability of workers in S&E occupations: (1) the number of these workers in a metropolitan area or division and (2) the proportion of the entire metropolitan workforce in S&E occupations. For both measures, estimates are affected by the geographic scope of a metropolitan area, which can vary significantly. Thus, comparisons between areas can be strongly affected by how much territory outside the urban core is included in the metropolitan area.

Table 3-8 presents the total number and proportion of workers in STEM and S&E occupations in the very large metropolitan areas with multiple metropolitan subdivisions. Metropolitan divisions with the largest estimated proportion of the workforce employed in S&E occupations are shown in table 3-9; those with the largest estimated number of workers employed in S&E occupations are listed in table 3-10. The metropolitan areas with the highest estimated proportion of S&E employment are mainly smaller and perhaps less economically diverse. However, some large areas, such as Washington, D.C.; Seattle; Boston; and San Jose, also appear on the list of metropolitan areas with the greatest intensity of S&E employment. Differences between estimates for different areas are not necessarily statistically significant. More detailed information on all metropolitan areas can be found in appendix table 3-4.

S&E Workers by Employment Sector

Education Sector

Overall, the education sector employs 18% of scientists and engineers and 16% of those in S&E occupations (table 3-6). Depending on the population, however, the proportion working within different parts of the education sector varies. For example, for workers with an S&E doctorate, 4-year colleges and universities are the most important employer (appendix table 3-5). However, only a minority (41%) of S&E doctorate holders work in this sector, and not all of these are tenured or tenure-track faculty. This figure also includes individuals holding postdoc and other temporary positions, working in various other S&E teaching and research jobs, performing administrative functions, and employed in a wide variety of non-S&E occupations. (See chapter 5 for additional details on academic employment of science, engineering, and health (SEH) doctorates.)

Within the education sector, the portion of the workforce in S&E occupations is concentrated in 4-year institutions (81%). In contrast, most education sector workers in S&E-related or non-S&E occupations are found in precollege or other institutions (63% and 68%, respectively). These workers are primarily teachers in these types of institutions.

Business/Industry Sector

For-profit businesses. For-profit businesses employ the greatest number of individuals with S&E degrees (figure 3-12). In 2008, they employed 59% of all individuals whose highest degree is in S&E and 35% of S&E doctorate holders (appendix table 3-5). By occupation, they employ 53% of those working in S&E occupations.

Nonprofit organizations. Nonprofit organizations have shown substantial growth in the percentage of scientists and engineers that they employ (see figure 3-12). However, this is primarily driven by those working in S&E-related occupations (which include health-related jobs); 18.4% of the workers in S&E-related occupations work in nonprofit organizations (table 3-6). Among those in S&E occupations, the proportion is much smaller—4.4%.

Self-employment. More than 3.6 million individuals with S&E degrees or working in S&E occupations were self-employed in 2008—18.8% of all scientists and engineers in the United States (table 3-11; NSF/NCSES 2008). This SESTAT estimate of self-employment is much higher than others that have been published elsewhere because it includes those self-employed individuals who work in incorporated businesses. In contrast, most reports of federal data on self-employment are limited to individuals whose businesses are unincorporated.

Although only about one-third of all self-employed workers in the United States work in incorporated businesses (Census Bureau 2009), about two-thirds of self-employed scientists and engineers in the broad SESTAT population work in such businesses (table 3-11). The rate of incorporated self-employment is much higher for individuals with S&E degrees (12%), with S&E highest degrees (11%), or working in S&E occupations (8%) than for the U.S. workforce as a whole, where the comparable rate is 3% (Census Bureau 2009).

Scientists and engineers working in S&E-related or non-S&E occupations reported higher levels of self-employment (20% and 22%, respectively) than those working in S&E occupations. Some 16% of social scientists indicated that they are self-employed, but unlike the general pattern of higher incorporated self-employment exhibited among scientists and engineers in general, this group reported higher rates of unincorporated self-employment. This is largely driven by psychologists, 30% of whom are self-employed, mostly in unincorporated businesses (NSF/NCSES 2008). Many scientists and engineers who are self-employed are working in small businesses. Some 81% of self-employed individuals in unincorporated businesses and 46% of self-employed people in incorporated businesses are working in businesses with 10 or fewer employees. Some proportion of these scientists and engineers are likely to be working as independent professionals, rather than in small businesses.

The proportion of self-employed workers generally decreases by level of degree and increases with age (figure 3-15). Across all ages, 18% of S&E bachelor's degree holders are self-employed, but the proportion falls to 12% for S&E doctorate holders. However, self-employment increases with age at all degree levels. By ages 60–64, self-employment reaches about 35% for bachelor's degree, 27% for master's degree, and 21% for doctorate holders.

Government Sector

Federal government. The United States' federal government is a major employer of scientists and engineers. However, its employees are largely limited to those with U.S. citizenship.[6] According to data from the U.S. Office of Personnel Management, the federal government employed approximately 235,000 persons in S&E occupations in 2009. Many of these workers were in occupations that, nationwide, include relatively large concentrations of foreign-born persons, some of whom are not U.S. citizens, rendering them ineligible for many federal jobs. Among federal employees in S&E occupations, 60% were in science occupations and 40% were in engineering occupations.

The five federal agencies with the largest proportions of scientists and engineers among their workforce are those with strong scientific missions: the National Aeronautics and Space Administration (NASA), Nuclear Regulatory Commission (NRC), Environmental Protection Agency (EPA), National Science Foundation (NSF), and Department of Energy. The Department of Defense employed the largest number of scientists and engineers, with 43% of the federal S&E workforce (NSF/NCSES 2012b, forthcoming).[7]

Overall, scientists and engineers represent approximately 11.5% of the entire federal workforce. Among federal executives in the Senior Executive Service (SES),[8] 22% are scientists and engineers.

State and local government. Data from the 2010 OES survey show that there are approximately 7.89 million employees of state and local governments in the United States. In 2008, SESTAT estimated 1.48 million scientists and engineers working in this sector. Approximately 8% of individuals with highest degrees in S&E work in this sector; 7% of those with S&E occupations also work there (appendix table 3-5). Within S&E occupations, a larger proportion of biological and physical scientists work in state and local governments (11.3% and 10.5%, respectively), relative to other S&E occupations.

Scientists and Engineers and Innovation-Related Activities

Who Performs R&D?

Because R&D creates new knowledge and new types of goods and services that can fuel economic growth, individuals with S&E expertise who use their knowledge in R&D attract special interest. Using SESTAT data, this section reports two broad indicators of R&D work. One involves whether performing R&D is a major work activity constituting at least 10% of the worker's job. The other is whether workers report R&D as a primary or secondary work activity—an activity ranking first or second in work hours from a list of 14 choices.

In 2008, just over 14.1 million employed individuals had one or more S&E degrees (NSF/NCSES 2008). Overall, 31% of S&E degree holders report R&D as a major work activity in their principal jobs. The majority of them have bachelor's (52%) or master's (32%) degrees, while individuals with doctorates, who constitute only 6% of all individuals with S&E degrees, represent 12% of individuals who report R&D as a major work activity.

R&D as a work activity varies among S&E degree holders depending on the field of their highest degree. Figure 3-16 shows the proportion of S&E degree holders who report R&D as their primary or secondary work activity, by their highest degree level and field (which may not be in S&E). Among S&E fields, the highest degree holders in engineering reported the highest aggregate R&D activity rate (51%), while those in the social sciences reported the lowest rate (22%).

In all fields, doctorate holders report higher R&D activity rates than those at lower levels of educational attainment. Engineering doctorate holders report the highest R&D rates, with other doctorate holders in natural and mathematical sciences fields having slightly lower rates. Social sciences and health doctorates report the lowest R&D rates (figure 3-16). This pattern of differences among fields is similar to that found among all degree holders.

Doctorate holders in all fields engaged in declining amounts of R&D activity over the course of their careers (figure 3-17). The decline may reflect movement into management or other career interests. It may also reflect increased opportunity for more experienced scientists to perform functions involving the interpretation and use of, as opposed to the creation of, scientific knowledge.

Many S&E degree holders subsequently earn degrees in other fields, such as medicine, law, or business. Figure 3-16 includes individuals who have at least one S&E degree, but then may have earned other degrees in S&E-related and non-S&E fields. These individuals report substantial R&D activity rates less often than workers whose highest degrees are in S&E fields. Nonetheless, the proportions who report R&D as their primary or secondary activity—18% for those whose highest degree is in an S&E-related field and 21% for those whose degree is in a non-S&E field—are still substantial and are similar to those for people with their highest degree in the social sciences.

R&D activity spans a broad range of occupations. Table 3-12 shows the occupational distribution of S&E degree holders who spend at least 10% of their time on R&D or report R&D as a major work activity. Among the former, 39% are in non-S&E occupations (lawyers or non-S&E managers, for example). Twenty-seven percent of those for whom R&D is a major work activity are in non-S&E occupations.

R&D Employment in the Business/Industry Sector

A large proportion (78%) of scientists and engineers who work in the business/industry sector report spending at least 10% of their work hours on R&D activities; this proportion is 80% for those employed in the for-profit sector (NSF/NCSES 2008). The 2009 Business R&D and Innovation Survey, which includes only U.S.-located companies that fund or perform R&D, allows for further examination of R&D employment in this sector.

The proportion of R&D employment relative to total employment, or R&D employment intensity, is one indicator of a company's involvement in R&D activity. Companies located in the United States that performed or funded research and development domestically or overseas employed an estimated 27.1 million workers worldwide in 2009 (NSF/NCSES 2012a, forthcoming). The domestic employment of these companies totaled 17.8 million workers, including 1.4 million domestic R&D employees. Thus, domestic R&D employment accounted for 8% of companies' total domestic employment (table 3-13).

Smaller companies reported higher proportions of domestic R&D employment than did larger companies, with companies of 250 or more reporting 10% or fewer of their domestic employees as R&D employees, and small companies reporting rates higher than 10% (table 3-13). The greatest proportion of R&D employment (27.0%) is among companies of 5–24 employees, whereas the smallest proportion (5.1%) is among very large companies of 25,000 or more.

R&D employment is found in both manufacturing and nonmanufacturing industries, but at different rates. R&D employment intensity is 8.6% in manufacturing industries and 7.3% in nonmanufacturing industries (figure 3-18). Examination of this indicator across industries shows the highest R&D employment intensity rates in scientific R&D services (36%), communications equipment (30%), software publishers (28%), semiconductor and other electronics equipment (27%), and pharmaceuticals and medicines (20%).

Patenting Activity of Scientists and Engineers

The U.S. Patent and Trademark Office grants patents to inventions that are new, useful, and not obvious. Patenting is a limited but useful indicator of the inventive activity of scientists and engineers.

In its 2003 SESTAT surveys of the S&E workforce, NSF asked scientists and engineers to report on their recent patenting activities. Among those who had ever worked, 2.6% reported that from fall 1998 to fall 2003 they had been named as an inventor on a U.S. patent application (NSB 2010). Patenting activity rates were highest among those employed in the business/industry sector.

The patent office does not grant all patent applications, and not all granted patents produce useful commercial products or processes. NSF estimates that in the 5-year period for which data were collected, U.S. scientists and engineers filed 1.8 million patent applications. The patent office granted some 1 million patents (although applicants may have applied for some of these at an earlier period).

Of those patents granted between 1998 and 2003, about 54% resulted in a commercialized product, process, or license during the same period. Scientists and engineers employed in the business/industry sector reported the highest commercialization success rate (58%), much higher than the education (43%) and government (13%) sectors.[9] The overall commercialization rate varies by degree level, at 60%–65% for bachelor's and master's degree holders but 38% for doctorate holders (many of whom work in education, which has a low commercialization rate relative to other sectors).

In 2003, the patent activity rate of doctorate holders was 15.7%, compared with 0.7% among those whose highest degree was at the bachelor's level.[10] However, there are far fewer doctoral-level scientists and engineers, so they accounted for only about a quarter of all survey respondents named on a U.S. patent application. Bachelor's and master's degree holders accounted for 41% and 31%, respectively, of all patenting activity reported in the survey.

More recent data from 2008 on a subset of scientists and engineers—U.S.-trained science, engineering, and health (SEH) doctorates—show that the patent activity rate of this set of employed doctorate holders from 2003 to 2008 was 16.2% (table 3-14). The highest patenting activity rates were among doctorate holders in engineering (38.6%) and physical sciences (25.0%). Doctorate holders in these two fields also report the highest average number of applications per person (5.9 in both fields) and the highest average number granted (3.6 and 3.4, respectively). Doctorate holders in engineering and computer/information sciences report the highest average number commercialized (1.5 in both fields).

Work-Related Training

In addition to formal education, scientists and engineers very often engage in work-related training. Such training can contribute to innovation in the economy by enhancing skills and knowledge within the S&E labor force. According to SESTAT, about three-fifths of scientists and engineers participated in work-related training in 2008. Among those who were employed, the rate was approximately 67%; for the unemployed, it was 32% (table 3-15). Among employed scientists and engineers, those in S&E-related occupations (health-related occupations, S&E managers, S&E precollege teachers, and S&E technicians and technologists) had the highest participation rate (79%).

Most who took training did so to improve skills or knowledge in their current occupational field (53%) (appendix table 3-6). Others did so for licensure/certification in their current occupational field (24%) or because it was required or expected by their employer (14%). Relative to those who were employed or not in the labor force, those who were unemployed more often reported that they engaged in work-related training to facilitate a change to a different occupational field. Not surprisingly, those who were not in the labor force more often reported that they engaged in this activity for leisure or personal interest. Women participated in work-related training at a higher rate than men: 61% compared with 55% of men (appendix table 3-7). This difference exists regardless of labor force status or highest degree level.


[6] Only U.S. citizens and nationals may be appointed in the competitive civil service; however, federal agencies may employ certain noncitizens who meet specific employability requirements in the excepted service or the Senior Executive Service.
[7] This list does not include the National Institutes of Health, which is a part of the Department of Health and Human Services (DHHS). The proportion of all federal scientists and engineers working at DHHS is 5%.
[8] SES includes occupations of senior managerial, supervisory, and policy positions in the executive branch of the federal government who generally serve as the link between political appointees and the rest of the federal workforce.
[9] The commercialization success rate is the ratio of patents commercialized to patents granted.
[10] The patent activity rate is the proportion who report having been named as an inventor on a patent application in the previous 5 years.