Title : All textual matter Type : Nonacademic S&Es: Trends from the 1980 and 1990 Censuses NSF Org: SBE / SRS Date : March 10, 1995 File : s4190000 Nonacademic Scientists and Engineers: Trends From the 1980 and 1990 Censuses Mark C. Regets, Project Officer Division of Science Resources Studies Directorate for Social, Behavioral and Economic Sciences National Science Foundation NSF 95-306 Suggested Citation National Science Foundation, Nonacademic Scientists and Engineers: Trends From the 1980 and 1990 Censuses, NSF 95-306 (Arlington, VA, 1995). Availability of Publications Single copies are available free of charge from the Division of Science Resources Studies, National Science Foundation, Arlington, VA 22230. If you are a user of electronic mail and have access to Internet, you may order publications electronically. Internet users should send requests to pubs@nsf.gov. In your request, include the NSF publication number and title, your name, and a complete mailing address. Printed publications may also be ordered by fax (703-644-4278). Publications should be received within 3 weeks after receipt of request. See page iv for availability on the Science and Technology Information System (STIS). Telephonic Device for the Deaf (703) 306-0090 ------------------------------------------------ CONTENTS Section Page Data Issues.....................................................1 Highlights......................................................3 Table 1: General Characteristics........................3 Table 2: By Occupation..................................3 Table 3: By Region and State............................3 Table 4: Median Earnings................................3 Table 5: Ethnic Minorities and Women....................3 Table 6: Immigrant Scientists and Engineers.............3 List of Tables 1. General characteristics of nonacademic scientists and engineers: 1980 and 1990................................5 2. Nonacademic scientists and engineers, by census occupation: 1980 and 1990...............................7 3. Nonacademic scientists and engineers, by State: 1980 and 1990...........................................9 4. Nonacademic scientists and engineers, median earnings: 1979 and 1989..........................................11 5. Nonacademic scientists and engineers, by sex and ethnic minority: 1980 and 1990.........................13 6. Immigrant nonacademic scientists and engineers: percent of total, by educational level and age group: 1980 and 1990...................................15 ------------------------------------------------ STIS page ------------------------------------------------ DATA ISSUES This report presents a broad picture of the characteristics of scientists and engineers (S&Es) outside academia and how these characteristics changed between 1980 and 1990. Data for this report come primarily from National Science Foundation (NSF) tabulations of 1980 and 1990 Public Use Microdata Samples (PUMS) of the decennial census provided by the U.S. Bureau of the Census. Specifically, data for S&Es in 1980 and their wage and salary income in 1979 comes from the 1980 5-percent PUMS. Data for S&Es in 1990 and their wage and salary income in 1989 comes from a reweighted combined file of the 1990 1-percent and 5-percent PUMS. Data for wage and salary income for all workers come from the 1980 and 1990 1-percent PUMS. Decennial Census occupation codes were used to identify S&Es. All postsecondary teachers and individuals whose industry was identified as "colleges and universities" were excluded to maintain a focus on nonacademic S&Es. Technicians and technologist occupations, though closely related to science and engineering, were also excluded. The counts of S&Es in this report do not agree with other NSF reports and data on S&Es for a number of reasons. First, many NSF reports are based upon field of degree rather than occupation. On decennial census data, S&Es can only be identified through an occupation variable coded primarily through a write-in occupation question on the decennial long form. These variations in defining S&E result in obvious differences--the number of individuals with a Ph.D. in biology will be different from the number of individuals with Ph.D.'s whose occupation is biologist. Second, the decennial censuses may classify the occupation of many S&Es in ways that prevent their identification as S&Es. For example, "post-secondary teacher, subject not specified," "authors," "technical writers," various types of managers, and various types of technicians are all examples of occupation categories which may contain unidentified scientists and engineers. It is largely for this reason that this report makes no attempt to examine changes in the characteristics of academic S&Es. In NSF's own demographic surveys--the Survey of Doctoral Recipients, the National Survey of College Graduates, and the National Survey of Recent College Graduates-- special care is taken to reduce these problems in identifying those whose occupations are in science or engineering. Around one-third of those with science and engineering occupations on the 1980 and 1990 censuses appear to have less than a bachelor's degree. Other data sources typically use a minimum of a bachelor's degree in defining S&Es. Thus it is important to consider education level in most analytic uses of this data. Comparisons between the 1980 and 1990 Census that use education level are problematic due to differences in how education was measured in the two censuses. In 1980, census long-form recipients were asked about the number of years of education they had completed. In 1990, the education question was changed to ask about degree completion rather than years in school. However, it is possible to measure how close a comparison can be made between the two measures, because a group of the same individuals answered the two different versions of the education question on the 1991 and 1992 Current Population Survey (CPS). A tabulation of a match between the 1991 and 1992 CPS provided by David A. Jaeger of the Department of Economics and Population Studies Center of the University of Michigan1/ indicates that of individuals with 16 or more years of completed education measured by the 1980 education question, only 6.9 percent did not indicate a bachelor's degree or higher in their response to the 1990 question. By the same measures, 1.8 percent of those with less than 16 years do have a bachelor's degree. Overall, using 16 years or more of education as a measure overestimates the total number of bachelor's degrees by 1.2 percent. This report uses "bachelor's or higher" as an additional breakdown of 1980/1990 differences in characteristics but does not attempt finer comparisons by education where there is more disagreement between the 1980 and 1990 education measures. Accordingly, great care should be used in interpreting 1980/1990 changes reported by education. _____________________________ 1/ David J. Jaeger, Reconciling the Old and New Census Bureau Education Questions: A Recommendation for Researchers, Working Paper (Ann Arbor: University of Michigan Population Studies Center, June 1984). Earnings from wages and salaries are reported on each decennial census for the previous year; i.e., 1979 and 1989. In this report 1979 earnings are reported in 1989 dollars using the implicit price deflator for personal consumption expenditures. In calculating median earnings for this report, those with zero earnings were excluded. HIGHLIGHTS Table 1: General Characteristics Between 1980 and 1990 the population of nonacademic scientists and engineers became older; more educated; more likely to be attending school, more likely to be self-employed; more black, Hispanic, and Asian; more foreign born; more likely to report a disability; and much more female. The proportion of female S&Es increased from 12.6 to 22.2 percent from 1980 to 1990. Over three-quarters of nonacademic S&Es work in the private sector for a for-profit or non-profit company. However, this share drops to 57.3 percent at the Ph.D. level, where 24.7 percent work for government and 18.1 percent are in some form of self-employment. The median education level is a bachelor's degree, with 30.0 percent of nonacademic S&Es in 1990 having less than a bachelor's and only 5.3 percent with a doctorate. Table 2: By Occupation Overall the number of nonacademic S&Es increased by 64.4 percent, with a larger 74.0-percent increase in the number with a bachelor's degree or higher. Computer systems analysts and scientists; physical scientists not elsewhere classified, n.e.c.; psychologists; and social scientists, (n.e.c.), all more than doubled in number. The number of "operations and systems researchers and analysts" more than tripled-- increasing by 250.2 percent. There were decreases in the number of metallurgical and materials engineers; mining engineers; agricultural engineers; mathematical scientists, n.e.c.; and sociologists. Table 3: By Region and State All geographic regions and all but one State showed increases in the number of S&Es. The slowest growth in the number of S&Es, 41.1 percent, was in the East North Central region. The fastest growth occurred in the South Atlantic region-- 98.1 percent. Table 4: Median Earnings Median real earnings remained essentially stable for all major occupation groups. Earnings growth ranged from a 7.9-percent decline for social scientists to a 0.5-percent increase for engineers. Over this 10-year period there was a 1.3-percent decline in real median earnings for S&Es. This decline in earnings occurred despite the small increases in both age and education of the S&E population, and S&Es with bachelor's degrees or higher actually experienced a slightly greater decline. Younger S&Es (those under age 40) fared better than older S&Es in each occupation group. For those with a bachelor's degree or higher, the median earnings of younger S&Es increased by 1.2 percent, whereas median earnings declined by 5.1 percent for older S&Es. Despite this decline, S&Es continue to earn more than the national median for workers in all occupations. This is true even at the Ph.D. level, where the 1989 median salary for S&Es was $50,000, versus $43,000 for all Ph.D.'s. Table 5: Ethnic Minorities and Women Although representation of women and minorities varies greatly among occupations, every occupational group showed increases between 1980 and 1990 in the proportions of blacks, Hispanics, Asians, and women. Over all nonacademic S&E occupations and education levels, the proportion of blacks increased from 3.2 to 4.4 percent; Hispanics, from 2.2 to 3.1 percent; Asians, from 4.2 to 6.0 percent; and women, from 12.6 to 22.2 percent. Table 6: Immigrant Scientists and Engineers Overall, the proportion of foreign-born S&Es increased from 9.1 to 11.0 percent between 1980 and 1990, with increases occurring in all S&E occupational groups except social science. The importance of foreign-born S&Es to the total numbers of nonacademic S&Es increases with education level, rising to 22.6 percent of Ph.D.s in 1990. Both the greatest proportion and the greatest increase in the proportion of foreign-born S&Es occur for 45- to 54-year-olds, whose proportion increased from 8.5 to 15.5 percent. Among Ph.D.s, engineers had the greatest proportion of foreign born at 39.2 percent.