National Academies Press: OpenBook

Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives (1987)

Chapter: Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries

« Previous: New Technology and Office Tradition: The Not-So-Changing World of the Secretary
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 136
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 137
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 138
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 139
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 140
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 141
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 142
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 143
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 144
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 145
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 146
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 147
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 148
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 149
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 150
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 151
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 152
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 153
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 154
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 155
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 156
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 157
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 158
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 159
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 160
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 161
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 162
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 163
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 164
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 165
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 166
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 167
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 168
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 169
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 170
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 171
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 172
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 173
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 174
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 175
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 176
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 177
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 178
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 179
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 180
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 181
Suggested Citation:"Integrated Circuits/Segregated Labor: Women in Computer-Related Occupations and High-Tech Industries." National Research Council. 1987. Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives. Washington, DC: The National Academies Press. doi: 10.17226/951.
×
Page 182

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Integrated Circuits/Segregated Labor Women in Computer-Related Occupations and High-Tech Industries MYRA H. STROBER and CAROLYN L. ARNOLD . . We are just beginning to see the. repercussions in all of our lives of the technological feat of fitting the electronic wiring and switches of what was a room-sized computer in 1946 onto a less than fingernail-sized piece of silicon and metal by the end of the 1970s. This silicon chip is the core of a technological revolution, the result of many attempts over a century to produce a "com- puting machine" that is small, fast, and cheap. Now, as the chips and, thus, the computers they make possible get smaller, faster, and cheaper, their applications in both old and new products are spawning a new high-technology industry. We see changes in a multitude of workplaces and homes, the expansion of opportuni- ties in existing industries and occupations, and the creation of new The authors would like to thank Deborah Thresher for her excellent preliminary research on this topic. An earlier version of this paper was given at the annual meeting of the American Educational Research Association, New Orlean~, April 25, 1984, and benefited from comments by Russell Rumberger. We also received helpful comments from F`rancine Blau and Philip Kraft on the version presented at the National Academy of Sciences on February 28-March 1, 1985. 136

MYRA H. STROBER AND CAROLYN L. ARNOLD 137 industries and occupations that were not even imagined just a few years ago. The occupations most involved in the computer revolution are engineers, computer scientists/systems analysts, programmers, electronic technicians, computer operators, and data-entry work- ers; these occupations are expanding both within the computer industry and in other industries as well. In addition, the com- puter industry provides new opportunities for managers, clerical workers, and production workers. Traditionally women have been sharply segregated into different occupations from men and have been paid less than men (Gross, 1968; Lloyd and Neimi, 1979; Blau and Hendricks, 1979; O'Neill, 1983; Bielby and Baron, 1984; Strober, 1984; Neiman and Hartmann, 1981~. This study poses several questions. Are there better opportunities for gender in- tegration and earnings equity in these new occupations that are growing rapidly, are exhibiting labor shortages, and are suppos- edly not locked into past traditions and stereotypes? Are there better opportunities for gender integration and pay equity in high- technology (high-tech) industries? What does the growth of these occupations imply for women's employment? This paper is divided into four sections. The first discusses the details of the occupations we analyze. The second section uses published data as well as the 1/1000 Public Use Samples (P.U.S.) from the 1970 and 1980 U.S. Censuses to look at how women are faring in the six major computer-related occupations and in high- tech industries. In the third section, the 1980 published data and the 1980 P.U.S. are used to examine the relative earnings of men and women in three computer-related occupations in high-tech and non-high-tech industries. In the fourth section, we discuss our findings and their implications. In brief, we found that although high tech in general and computer occupations in particular are often seen as the great equalizers, especially for those with higher education, in fact, there is considerable gender segregation in both high-tech industries and computer-related occupations in all industries; there is also considerable male-female earnings differentiation. We suggest that one possible cause of the earnings differentiation is that men and women in computer occupations are not employed equally across

138 COMPUTER-RELATED OCCUPATIONS industries; women tend to be employed more frequently in the lower-paying end-user industries.) COMPUTER-RELATED OCCUPATIONS DESCRIPTIONS OF OCCUPATIONS The development of semiconductors, computers, and com- puter languages spawned several new occupations and expanded several others. There are six major computer-related occupation groups—engineers, computer specialists, engineering and science technicians, production workers, computer operators, and data- entry operators. For three of these groups (engineers, engineering and science technicians, and production workers), we have re- stricted the analysis to those employed in the computer industry. For the other three groups (computer specialists, computer oper- ators, and data-entry operators), we have examined employment in all industries. The following descriptions of the occupations in these groups are based on definitions in the Standard Occu- pational Classification Manual, 1980 (U.S. Department of Com- merce, 1980), California Employment Development Department publications (ABAG, 1981), and interviews with workers and em- ployment counselors. Engineers Engineers design hardware for computers, including the elec- tronic circuits. The largest group of engineers is electrical engi- neers, but mechanical and industrial engineers also work in the computer industry. Sometimes they incorporate software designs into the circuits. Engineering is the highest-status and highest- paid computer-related occupation, with engineers generally having at least a B.S. degree in engineering; many have advanced degrees. ~ We define End-user industries as those that use the products of the computer industry and make only minor changes to the products to accom- modate their needs, rather than making basic new developments in these products. The companies developing computers and/or their software are part of the computer industry. The companies in all other industries, which will use these computers and/or software, are part of end-user industries. It is true that even within the computer industry the administrative divi- sions of the companies are end-users of computers. However, census data do not permit us to make such fine distinctions within industries, and we are interested here mainly in any broad differences between industries.

MYRA N. STROBER AND CAROlYN L. ARNOLD Computer Specialists 139 As we look beyond designing hardware to designing software, the sets of instructions that tell the computer which operations to perform, we encounter the computer specialist occupations. While computer engineers tend to be employed largely by computer com- panies, computer specialists are employed in virtually every major industry group. These jobs involve a hierarchy of tasks that used to be done by one person with the title of computer programmer. When the first computer was unveiled in 1946 (it was room- sized because the circuits were made with glass vacuum tubes), the engineers who designed it thought that the main task of ar- ranging the circuits had been done, and that giving instructions to the computer to perform calculations would be a simple clerical task. So they hired people who usually do clerical tasks women. In this case the women were recent college graduates with math backgrounds. However, these women found that in order to get the computer to do calculations, they (the programmers) had to know all about the design of the circuits and the way those circuits worked in the computer; they had to tell the computer not only what to do, but how to do it. The simple operation of performing calculations (in this case for Navy shell trajectories) became a high-level task that involved a knowledge of logic, math- ematics, and electronic circuits. These women programmed the necessary calculations and went on to do others. However, those who watched the programming process began to realize that pro- gramming was a high-level, challenging, and creative occupation. As the occupation grew, it became largely male (Kraft, 1979~. Ironically, some programming today is akin to the type of cler- ical job that computer designers (mistakenly) thought it would be in the late 1940s. Over time, with the development of higher-level languages (closer to human languages)2 and more routine appli- cations, programming tasks that were previously highly skilled, highly paid, and concentrated among highly educated workers have been broken down into more routine tasks and distributed among less-skilled workers. Kraft (1977) has suggested that, as this ~deskilling" has occurred, it Is women who have moved into 2 Note that higher-level languages are closer to human languages and hence are easier to use in programming. Thus, paradoxically, ~higher"-level languages require lower skill and have lower prestige associated with their use.

140 COMP UTER-RELA TED O CCUPA TIONS the less-skilled jobs.3 Greenbaum (1979) found that the lowest- leve} programming jobs were disproportionately occupied by racial and ethnic minorities. This history of the developing hierarchy in computer program- ming is reflected in the designations given by the Bureau of the Census to the computer specialist occupations. In 1960 and 1970, computer specialists, including programmers, were included in the professional category. By 1980, the Census put the three-digit occupational category of computer scientists/systems analysts in the professional category and the three-digit category of computer programmers in the technical category. The following descriptions attempt to capture the current hi- erarchy and educational requirements among computer specialists. Some workers In these occupations do not have formal credentials, having been trained or self-taught on the job. These job titles and descriptions continue to change and overlap. Computer scientists and some systems analysts work with engineers to design the overall hardware and software systems and sometimes know just as much about the hardware, although their training is more concentrated in the logic and mathematical models of computer systems, rather than on electronic principles. They also develop new languages to be used by other programmers. Generally, they have an M.S. or Ph.D. in computer science (CS) or electrical engineering (EE) or both (CS/EE). Computer systems analysts conceptualize and plan how a busi- ness or industrial task, such as automating a payroll or an assembly line, will be solved by computerization. Systems analysts do not write the programs but make flow charts to show the subtasks that need to be done by people and computers and their sequence and . · ~ riming. Computer programmers are often promoted into systems ana- lysts positions because these positions represent higher-level skills, responsibility, and pay than do programming positions. If systems analysts were not previously programmers, their education is ei- ther in business or data processing. There is a debate in this field about whether systems analysts need programming skills or not. 3 Braverman (1974) originally identified and labeled this process as the "degradation of work.n It soon became known as `'deskilling.n

MYRA H. STROBER AND CAROLYN L. ARNOLD 141 Systems programmers maintain and modify operating sys- tems systems of programs that coordinate all the hardware in a particular computer so it will run according to certain high-level languages. They are also responsible for updating the high-level software on the system in their particular company. They gener- ally require a B.S., M.S., or Ph.D. in CS, EE, or math. Programmer/analysts update operating systems and write programs that tailor the computer's uses to each individual work- place. Although ready-made software is available for many pur- poses, most firms need programmers to modify or write programs that reflect their own computing needs. These programmers need to know both operating systems and high-level languages. Ed- ucation requirements are a B.A. in related subjects with some programming experience, a B.S. in CS, or an M.B.A. Software engineers, as some programmers are increasingly called, design and write programs in high-level languages specif- ically for certain computers. These programs, often called pack- ages, are sold with the computer to make it easier for nonprogram- mers to use. Packages can include such items as games, accounting programs, and instructional programs. Producing these programs requires the creativity to conceptualize and design new ways to use the computer. People writing these software packages require a good knowledge of the language used to program the software and good ideas about marketable packages. Acquiring knowledge of programming and having creative ideas are more important for job success than are educational degrees. Consequently, a range of people, from high school students to Ph.D.s, are designing and writing software. Programmers, sometimes called coders or applications pro- grammers, are mainly translators. They translate instructions for a certain application in one language into the programming language that their particular computer anti use to produce the same results. The job category itself encompasses a range of skill and creativity from routine coding of sections of an application program to a task more like a programmer/analyst, depending on their industry or firm. These jobs can be done with less than a B.S. in CS. However, if those with a B.S. degree are available, employers often prefer to hire them.

142 COMPUTER-RELATED OCCUPATIONS Engineering and Science Technicians A third major group of computer-related workers is engineer- ing and science technicians. This group is found in the com- puter industry and is mainly made up of engineering technicians trained in electronics, although there are also some science techni- cians working within the industry. The electronic technicians have enough specialized knowledge of electronics to be able to construct, test, and repair the circuitry and components of computers that the engineers design and to understand engineering specifications and problems. Although they do not do original design work, they operationalize designs, test them, and then advise engineers on possible modifications. They work both in research design and in production to test and troubleshoot both new and existing prod- ucts. Also included in the engineering technician category are drafters, who, using both manual and computer-assisted drafting tools, make drawings of the circuit boards and components that the engineers design. The standard degrees are the 2-year assm ciate of arts (A.A.) or associate of science (A.S.) for an engineering or science technician and an A.A. in drafting for a drafter. Computer Operators The fourth computer-related occupational group is computer operators. They are employed in all industries. Computer opera- tors run the external operation of the computer; ensure that the computer receives the programs and data; and coordinate disks, tapes, and printing connections to the computer, either manu- ally or by supervising automated systems. This occupation ranges from active high-level interactions with the programs to routinized supervision of automated systems. It is sometimes an entry-level job leading to low-level programming. Education needed is simple knowledge of the equipment from a short training course and/or from experience. The occupation is rapidly being deskilled as more of its functions become automated. Data-Entry Operators Data-entry operators, the fifth computer-related occupation, put information into a form that can be read by a computer. This information used to be keypunched onto cards but is now almost always put onto tapes or disks from terminals. The operators,

MYRA H. STROBER AND CAROLYN L. ARNOLD 143 who basically type numbers and letters into a terminal, require training in typing. They are also employed in all industries. Production Workers The sixth major computer-related occupation is production worker. While many of the production jobs are similar to those in other industries, the following jobs are unique to semiconductor and computer production. There are generally no specific edu- cational requirements for these jobs, although people with some knowledge of electronics are generally preferred. Often hazardous chemicals are used in the production process. Semiconductor processors put materials through chemical and mechanical processes to create semiconductor integrated circuits on chips. They work either manually or, as these tasks become mechanized, with processing machines. Semicond?tctor assemblers assemble chips into wired devices which become the complete integrated circuit. This includes bond- ing wires to circuits, a task which is done under a microscope, and cleaning the circuits with chemicals. Electronic assemblers assemble the integrated circuits and other electronic components into a frame that becomes the fin- ished product (e.g., a computer). Electronic assemblers can be promoted to electronic testers, who test chips, boards, and com- ponents as they go through assembly, or electronic inspectors, who examine the components for errors and specification requirements. A skilled occupation that is sometimes part of production and sometimes part of customer service is data-processing repair, which involves installing and repairing data-processing machinery in offices and on production lines. This job category includes a range of workers from electronic mechanics to assembly and wiring technicians. There are also, of course, managerial, professional/technical, sales, clerical, and service occupations within the computer indus- try. These jobs tend to be similar to such jobs in other industries. GENDER SEGREGATION COMPUTER-RELATED OCCUPATIONS In this section we begin the analysis of gender segregation by examining trends in total employment and women's employment

144 COMPUTER-RELATED OCCUPATIONS in computer-related occupations. Table 1 presents data both for occupations that are computer related regardless of their industry (computer specialists, computer operators, data-entry operators, and data-processing repairers) and for occupations that are com- puter related only when found in computer industries (engineers, engineering and science technicians, and most production work- ers). As Table ~ shows, between 1970 and 1980, employment in computer-relatec} occupations grew about 80 percent from about 1.5 million to 2.4 million. However, although the growth of these occupations is widely heralded, it is important to note that they represented only 2.0 percent of all employment in 1970j and 2.5 percent in 1980. (Among aD women workers, those in computer occupations represented 2.3 percent of employment in 1970 and 2.9 percent in 1980. The corresponding percentages among men workers were I.8 percent in 1970 and 2.2 percent in 1980.) This growth took place in the context of increasing participa- tion in the labor force by women. In 1970, women were 38 percent of the U.S. labor force; by 1980, they were 43 percent. However, in both those years, women's representation in computer-related occupations was either considerably below or considerably above their representation in the labor force as a whole, depending on the specific occupation. Despite the fact that the computer-related occupations are of relatively recent origin, they are already remarkably segregated by gender. In 1970, women were 2 percent of all engineers in the computer industry; in 1980, that figure had risen to only 5 percent. Thus, in the highest-paid, highest-prestige computer- related occupation, women are virtually absent. Among computer specialists in all industries, the situation is somewhat better, although women are still below their proportion in the overall work force. In 1970, women were 15 percent of all computer scientists/systems analysts. This occupation more than doubled from 1970 to 1980 (from 93,000 to 201,000), but by 1980, women were still only 22 percent. Among programmers, the proportion of women also increased, but they were also still underrepresented. The number of programmers almost doubled (from 16t,000 to 313,000), and the proportion of women grew from 23 percent in 1970 to 31 percent in 1980. Women were better represented among engineering and sci- ence technicians in the computer industry than among engineers, but were less well represented than among computer specialists.

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 1 Total and Women's Employment in Computer-Related Occupations, 1970 and 1980 145 Occupation 1970 Percent Number Women 1980 Number Percent Women Total employed in labor force Total employed in computer- related occupations Engineersa Electr~cal/electronic engineers Computer specialists Computer scientiste/eysteme analysts Computer programmers Engineering and science technicianea Electronic technicians Drafters Computer operators Data-entry operators Production workersa Operatives, fabricators, transporters, and laborers Assemblers Electronic assemblers Data-processing repairers— Percent of total employed labor force in computer-related occupations 76,553,599 38 1,497,683 90,626 2 47,004 2 254,537 20 93,200 15 161,337 23 58,292 11 31,454 11 16,963 7 117,222 29 272,570 90 680,299 46 519,221 58 158,191 74 24,137 3 1970 2.0 97,639,355 2,424,240 125,055 67,320 513,863 200,684 313,179 90,990 60,299 16,726 408,475 378,094 872,345 43 5 4 28 22 31 17 15 16 59 92 49 591,091 58 208,284 55,879 35,418 1980 2.5 72 77 88 NOTE: Computer-related occupations are defined in Appendix B. aEmployment data for these occupations are only for the computer industry defined as two three-digit industries: Electronic Computing Equipment (SIC Codes 1970:189; 1980:322) and Electrical Machinery, Equipment, and Supplies, not elsewhere classified, including semiconductors (SIC Codes 1970:208; 1980:342). Data for all other occupations are for all industries. bWorkere not in the computer industry; data-processing repairers in the computer industry are included in "Production workers" above: 6,707 in 1970 and 11,208 in 1980. Women also accounted for 3 and 8 percent, respectively in 1970 and 1980, for these repairers. SOURCE: 1970 data, Bureau of the Census (1972:Table 8); 1980 data, Bureau of the Census (1984:Tables 1 and 4). In 1970 in the computer industry, women were 11 percent of engi- neering and science technicians; in 1980 they represented 17 per- cent. They were similarly represented among electronic/electrical technicians t! percent in 1970 and 15 percent in 1980. As drafters, women did less well in 1970 than other technicians- only 7 percent were women but did better in 1980, when 16 percent were women.

146 COMPUTER-RELATED OCCUPATIONS Initially, there was no clear indication as to which gender would be assigned to the occupation of computer operator. In 1960, when there were only 2,000 computer operators, women held 65 percent of the jobs (Dicesare, 1975~. Between 1960 and 1970, the jobs in this occupation increased more than sixfold, to 117,000. More of these new jobs were filled by men than by women so that in 1970, women were only 29 percent of all computer operators. In the period from 1970 to 1980, however, while the occupation increased fourfold, to 408,000, more of the new additions to the occupation were women, so that in 1980, women were 59 percent of all computer operators. Like most clerical occupations, data entry is preponderantly female. In 1970 women were 90 percent of data-entry operators. Between 1970 and 1980 the occupation became even more seg- regated so that by 1980 women represented 92 percent of such operators. Of all production workers in the computer industry, women were only slightly over their representation in the labor force as a whole: 46 percent in 1970 and 49 percent in 1980. However, when we look more closely at the less-skilled production occupa- tions, women's representation is much higher. Of all the operators, fabricators, laborers, and transportation workers, a group which includes the semiconductor processors and assemblers and all other lower-level production workers in the computer industry, women represented 58 percent in both 1970 and 1980. Among assemblers, a subset of operatives, women were about 73 percent in 1970 and 1980. Of electronic assemblers, a group identified only in 1980, women were an even higher proportion: 77 percent.4 However, in the occupation "data-processing machine repairers," we find again the extraordinary gender segregation we often see in techni- cal occupations: women held 3 percent of these jobs in 1970 and 8 percent in 1980. In Table 2 we present the difference between the percentage of women in the total labor force and the percentage of women in each computer-related occupation listed in Table 1. This dif- ference represents a rough measure of occupational segregation. If we compare the data for 1970 and 1980 we find that except for 4 In Silicon Valley in 1981, 40 percent of women assemblers were ethnic minorities (Rogers and Larsen, 1984~. This production occupation is even more segregated abroad (Grossman, 1980~.

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 2 Underrepresentation of Women in Con~puter-Related Occupations 147 Index 1970 1980 Occupation Engineers 36 38 Electrical/electronic engineers 36 39 Computer specialists 18 15 Computer scientists/systems analysts 23 21 Computer programmers 15 12 Engineering and science techniciansa 27 26 Electronic technicians 27 28 Drafters 31 27 Computer operators 9 -16 Data-entry operators -52 -49 Production workersa -8 -6 Operatives, fabricators, transporters, and laborers -20 -15 Assemblers -36 -29 Electronic assemblers -- -34 Data-processing repairersb 35 35 NOTE: Data from Table 1. The index is defined as the difference between percent women in total labor force and percent women in each computer-related occupation. Computer-related occupations are defined in Appendix B. aSame as Table 1. bSame as Table 1. computer operators, the level of segregation was approximately the same for the 2 years. Table 3 presents total employment and women's employment for the computer industry and for computer-related occupations in all other industries. These totals show the number of people who are involved in developing, maintaining, or supporting the produc- tion of computers in the computer industry, plus those working in occupations that were created as a result of the computer revolu- tion. Between 1970 and 1980, combined employment in the com- puter industry and computer-related occupations grew about 60 percent—from about 1.8 million to 2.9 million. Still, they repre- sented only 2.4 percent of all employment in 1970 and 3.0 percent in 1980. (Among all women workers, those in the computer in- dustry and computer occupations represented 2.7 percent of em- ployment in 1970 and 3.4 percent in 1980. The corresponding percentages among men workers were 2.3 percent in 1970 and 2.8

148 - o v ._ h on _I ah o ._ Go 00 C' ~ OK ~ ~ O a; o~ _ _ _ ._ O oo 3 ~ C;, — _ ~ o ·— o~ ._ m~ ~ C) ~ 0 00 (t, o P<~ h O ~ O C) ~ O ~ _ h G} ,~ O o o >~·— h OS a' ~ O CO 00 _~ _1 ~ U: O O CS) 1 Ct:~ ~ ~ e~ oo ~ ~ ~ ~ ~4 1 e~ co U: ~ U: ~ ~ C~ U) ~ CO 00 ¢9 0 C5) co ~ ~ oO U: C5) C9 CO 00 ~4 Ch' C~ ~ CO 0O ~ ~ ~ ~ ~ CD ~ O ~ e~ eo ~ co - - - - - - - - - - - - - - - c5> 0 u, ~ ~ oo co ~ o: ~ ~ e~ ~ oo u, co ~ 00 00 ~ CC ~ ~ ~ u: ~ 00 ~ CD 03 co ~ co ~ ~ e~ ~ e~ a~ o: - - - - e~ ~ ~ CO ~ ~ CD ~ CD I u: CO O O a, ~ ~o: <90 ~ I ~ e~a, C53 ~ - cO u: ~ - ~D ~ CD CO ~ CO CO 00 ~ U ~ u~ CD a 00 CD ~ ~ ~ ~ 0O ~ ~ ~ {D ~ ~ a, ~ 00 0 0 0 ~ ~ C. CD ~ u: {D - - - - - - - - - - - - - - ~ o: <9 ~ ~ oO - 4 0 ~ ~ CD e~ co O ~ U) C~ ~ U: ~ oo CC ~ ~ ~ {D e~ 00 ~ C~ ~ CO CD ~ ~ e~ ~t1 =1 ~ o 3 `xs ~ ~, ° ~ ~ oo Q ~ ~ ~ ~ ·~ ~S ° 4U ~n I ~ h a, ', E o ° -> c -° ~ ~ ~— ~ =.E c~ cS ~ ~ ~ x 0 0 0 ~ ~ ~ o al ._ ~ O · _ ~ ~ v ·= h O 4, =1 CD oq ~ O ~ · - `lS C. , O ~ h ~ol a~ oo ~V h— ~ h O ~ 3 ~ ~ ~ ~ ~ O h _ ~.c O O O O VVV~

149 Io o of o en - ~ 5 ~ no =£ Cal o of ~~~ A} ~ ~ to ~ ~— To 3 .= ~ 4, so o ~ o ~ o lo. ~ o ~ ~ ~ ~ ~ ~ V i, ~ ~ ~ · 7_ ~ ~ . '¢~ ~ ~ ~ ~ ;~ ~ t i ~5 l .

150 COMPUTER-RELATED OCCUPATIONS percent in 1980.) Although there appear to be changes between 1970 and 1980 in the percentage of women in the occupations listed in Table 3, changes in occupational definitions during the Midyear period were such that the categories cannot be reliably compared. GENDER, RACE, AND ETHNIC DISTRIBUTION IN FOUR COMPUTER-RELATED OCCUPATIONS If we look more closely at the four computer-related occupa- tions that are present in all industries computer scientists/sys- tems analysts, computer programmers, and data-entry operators— we can see how women and men in the three largest racial and ethnic groups are represented. Table 4 shows that in the employed labor force as a whole, white men are 50 percent of the workers, white women are 36 percent, black men are 4.8 percent, black women are 4.8 percent, and other racial groups make up 8.4 per- cent of the workers. Men and women of Spanish origin, who can be of any race, represent 3 and 2 percent of the workers, respectively. In the occupations of computer scientist, systems analyst, and computer programmer, women of all groups and minority men are underrepresented compared to their representation in the labor force as a whole, while white men are overrepresented. Within each racial and ethnic group, men are better-represented than women. Among computer scientists/systems analysts, the highest paid of these four occupations, white men hold 71.3 percent of the jobs, which is much higher than their representation of 50 percent in the labor force. White women occupy 19.5 percent of the jobs, only half as high as their representation in the labor force. Black men are less represented in this occupation than in the labor force (3 percent versus 4.S percent). Black women's representation is even poorer, 1.7 percent in computer scientists/systems analysts versus 4.8 percent in the labor force. People of other races are only about half as well represented in this occupation as in the labor force. Like black men-, the representation of men of Spanish origin is just over half that of their presence in the labor force (1.8 versus 3 percent). Women of Spanish origin are virtually unrepresented; they are only 0.1 percent of computer scientists/systems analysts, though they are 2 percent of the labor force. Among computer programmers, the next lower paying occu- pation, white men are still overrepresented, and the other groups

151 ~ sat Do ,~ o te a a' 11 o s" o s" o - o - s" ._ ~o ._ o oo ~ ·— (:, U] ~ t) ·— C: ~o .,, ~ o ~ ._ o ~t ·— Q CC ·— (2, oo C) ·— C.) ~ O - :>. _ P" I E~Q o V m ~ ~ O ~ co ~ ~ a, 0 ~ ~ oo (D . · . . · . . . · . U. ~ O O ~ ~ O O _% U' s~ a~ — ~ S" 00 4= 0 0 ~ ~ 11 O = - V 0 ~ ~_ O -4 Ei o~ 11 O S" v _ - 00 0D al ~ 0 V C~2 U] ¢ ~ _% 0O O C9 _ S" ~ o 11 ,Q o (3 z; E~ C5) U, ~ ~ O · · · . . co ~ ~ CD u, co e~ u: ~ u: . . . · ~ t9 o: e~ U3 'o cO . . . . . ~ ~ cO o o oo oo ~ · . . . . o ~ ~ ~ oo u: cO ~ oo c~ ~ . . . . ~ ~o o . . . . ~ o oo o ~ o oo ~ ~ o . . . . ~ o oo o ~ o ooo o . . . . co e~ u~ 0 ~ o ~ ~ ~ ~ ~ 3 ~ ~ 3 ~ E~ ~ ~ 3 ~ b0E~ m o s" o L} ._ ._ o ._ ct CQ o OO o OC s" . ~s o s" ._ 03 O _ o ~5 - . . ~4 00 o o o o oD ~v _ O o° ~ m o 0o 6q v · . . . u, v o o :z ~

152 COMPUTER-RELATED OCCUPATIONS are still underrepresented, but somewhat less so. White men are overrepresented by 12 percentage points (62.2 percent versus 50 percent) and white women are underrepresented by 10 percentage points (26.5 percent versus 36 percent). Black men's represen- tation is still about 3 percent, but black women's representation has increased to 2.5 percent. However, both percentages are be- low their share of-the labor force. Other racial groups have also increased their representation, but it is still below their labor force representation. Men and women of Spanish origin have vir- tually the same representation among computer programmers as among computer scientists/systems analysts I.9 and 0.1 percent, respectively. This situation changes and in some cases reverses itself in the lower-paying occupation of computer operator. White men are now only one-third of the workers, 16 percentage points less than in the labor force, and white women are now half of the workers, 14 percentage points above their representation. Black men's representation is about the same as in the labor force, and black women's representation is higher among computer operators (6.9 percent) than in the labor force (4.8 percent). People of other racial groups are still underrepresented, however. Men of Spanish origin have a representation among computer operators below their labor force percentage, while the percentage of women of Spanish origin is above their percentage in the labor force. Thus, in this lower-paid computer occupation, white and black women and women of Spanish origin are overrepresented, and white men are underrepresented, while the percentage of black men and men of Spanish origin reflects their percentage in the labor force as a whole. When we look at the lowest-paid computer occupation of data- entry operator, a clerical occupation, we see that all women's representation is dramatically higher and all men's representation is dramatically lower than for the other occupations listed in Table 4. White women are 71 percent of data-entry operators, twice their labor force representation. Black women and women of Spanish origin are represented in this occupation with a frequency three times greater than their representation in the labor force 15 and 6 percent, respectively. Like gender, race and ethnicity of workers in computer occu- pations is associated with pay and status of the occupation. The higher status and pay of an occupation, the more white men are

MYRA H. STROBER AND CAROLYN L. ARNOLD 153 overrepresented and the more minority men and all women are underrepresented. In occupations with lower pay and status, the presence of white men drops to much below their percentage of the labor force, the percentage of women of all races becomes higher than their labor force percentage, and the percentage of minority men approaches their labor force representation. In an occupation that is clerical, women's representation doubles and triples above their labor force percentage, and men virtually disappear. HIGH-TECH INDUSTRIES This section looks at how women are faring not just in comput- er-related occupations but in the group of industries known as "high tech, of which the computer industry is one part. We are interested in whether the computer industry and other high-tech industries, because they are growing rapidly and are relatively new, are, therefore, perhaps less gender stereotyped and more hospitable to women than are non-high-tech industries. The Bureau of Labor Statistics recently reviewed all current definitions of high-tech industries and developed a range of def- initions based on three factors (Riche et al., 1983:51~: "~1) the utilization of scientific and technical workers, (2) expenditures for research and development, and (3) the nature of the product of the industry. Different combinations of these factors produced three groups of high-tech industries, from the least-inclusive definition with 6 three-digit industries, to the most-inclusive definition with 48 three-digit industries. In some respects, the most intuitively ap- pealing definitions of high-tech industries are the least inclusive (drugs; office, computing, and accounting machines; electronic components and accessories; miscellaneous electrical machinery; aircraft and parts; and guided missiles and space vehicles). How- ever, the P.U.S. 1/1000 sample of employment in these six indus- tries is too small for the kinds of analyses we wish to do. Thus, we use as our definition of h~gh-tech industries the middle group of 28 three-digit industries: 26-manufacturing industries (including computers and semiconductors) with a proportion of technologi- cally oriented workers equal to or greater than the average for all manufacturing industries and a ratio of R&D expenditure to sales close to or above the average for all industries, and 2 nonman- ufacturing industries that provide technical support to high-tech

154 COMPUTER-RELATED OCCUPATIONS TABLE 5 Number Employed in High-Tech Industries as Percentage of Employed Labor Force, 1970 and 1980 1970 1980 Number Percent of Number Percent of Employed All Employed Employed All Employed All workers 4,557,000 6.5 6,060,000 6.4 All women 1,254,000 5.0 2,026,000 5.1 All men 3,303,000 7.4 4,034,000 7.4 NOTE: See Appendix A for definition of high-tech industries. SOURCE: Calculated by authors from U.S. Census (1/1000 P.U.S. Tape, 1970 and 1980~. manufacturing industries (computer and data~processing services, and research and development laboratories) (Riche et al., 1983~. A list of the industries designated as high tech appears in Appendix A. Although employment in high-tech industries grew over the 1970-1980 decade from 4.6 million to 6.1 million, these industries employed only about 6.5 percent of all employed workers, about 5 percent of all women workers and about 7.4 percent of men workers. Table 5 presents these employment trends for high-tech industries. We find also that these industries are no less segregated by gender than are other industries and, in fact, may be even more so. Table 6 shows that in both 1970 and 1980, most men in high- tech industries were in management, professional/technical work, or production, while most women were in clerical work or produc- tion. Compared to women in non-high-tech industries, women in high-tech industries are less likely to be in managerial and profes- sional/technical jobs and are much more likely to be in production work. High-tech industries provide more low-status occupations to women, as a group, than do non-high-tech industries. By contrast, men in high-tech industries are more likely to be in managerial or professional/technical positions than are men in non-high-tech in- dustries.

MYRA H. STROBER AND CAROLYN L. ARNOLD 155 TABLE 6 Percentage of Men and Women Employed in High-Tech Industries and Non-High-Tech Industries, 1970 and 1980 High-Tech Industries Non-High-Tech Industries Occupation Meri Women Men ~~ Women 1970 Number (thousands) Total Managers Professional and technical Sales Clerical 8 Services 2 1 Production 48 Other (farm and transportation) Total 3,195 1,205 39,357 23,67ff ~rce~ta~e distribution 1Q 3 13 27 6 $ <1 36 12 7 8 55 37 2 <1 id 1980 Number (thousands) 5 16 10 33 20 14 4,034 2,206 50,263 37,720 Percentage distribution Managers 14 6 13 8 Professional and technical ?6 10 13 18 Sales 3 1 10 It Clerical 8 35 7 31 Services 3 2 9 18 Production 44 46 36 11 Other (farm and transportation) 3 1 12 2 NOTE: See Appendix A for definition of high tech and non-high tech. Occupational categories were adjusted for consistency between 1970 and 1980 by the authors; a detailed listing of the categories is available from the authors. SOURCE: Calculated by author from U.S. Census (1/1000 P.U.S. Tape, 1970 and 1980~.

156 COMPUTER-RELATED OCCUPATIONS ANALYSIS OF RELATIVE EARNINGS OF WOMEN AND MEN IN THREE COMPUTER-RELATED OCCUPATIONS UNCORRECTED EARNINGS DIFFERENTIALS To examine salary differences between women and men we analyze three computer-related occupations where a sufficient number of jobs (at least 20 percent) are held by the "m~nor- ity genders computer scientists/systems analysts, computer pro- grarnmers, and computer operators.5 We find that even when women are employed in the same occupations as men, they do not receive the same pay. We used two sources to calculate the gender differential in pay the published census reports, where the only available mea- sure was the mean annual earnings, and the census P.U.S. sam- ples, where we could calculate the median hourly earnings. With each source, for the years available, we calculated the ratio of the women's earnings to men's earnings for each of these three occupations in all industries combined. As each source has both advantages and disadvantages, we present data from both. The advantage of using the published census reports is that they are more accurate than the samples. The disadvantages are that only the mean annual earnings are available for 1980, and the mean is a poor estimate of average earnings since it is so influenced by a few high values. In addition, since the median was used as an estimate for most of the comparable occupations in 1970, the 2 years could not be compared, except in the combined category of computer specialist. Another disadvantage is that the use of annual earnings precludes controlling for part-time or part-year workers. The advantages of using the census samples are that we could calculate the median—the most accurate estimate for average earnings for each occupation and year and could control for part- time and part-year workers by calculating the hourly earnings.6 5 Although production workers also qualified under the criterion, we did not study them, because they are only "computer related if they are in the computer industry, and isolating that industry would have created too small a sample. In addition, we wanted to compare computer-related occupations across industries, and computer-related production workers by definition are not included in other industries. 6 Hourly earnings were calculated by dividing each person's yearly salary (from the year prior to the census year) by the number of weeks worked (in

MYRA H. STROBER AND CAROLYN L. ARNOLD 157 The disadvantage is that the estimates were subject to large stan- dard errors. Interestingly, the two sources of income data produced similar and consistent results. Table 7 contains the uncorrected gender differentials in pay, based on the published census data. In the first two columns are the numbers (in thousands) of total employees (men and women) in each occupation in 1970 and 1980. The second two columns show percentages of all employees who were women in each year. The next four columns show, for the years it was available, the mean annual earnings of men in that occupation and the ratio of women's mean annual earnings to men's earnings. This ratio is the gender differential in pay for each occupation, with no controls for education or experience. For the combined computer specialist occupations, the uncor- rected ratio of women's to men's mean annual earnings was 0.71 in 1970 and 0.72 in 1980. This constancy of the ratio is noteworthy, as employment more than doubled over the 10-year period and the proportion of women in these occupations increased by 40 percent. In 1980, the uncorrected ratio of women's to men's mean annual earnings was available for the three occupations. For both computer scientists/systems analysts and computer programmers, the ratios were 0.73. For computer operators the ratio was 0.65. Since these were not corrected for part-time or part-year workers, we would expect the ratio estimates that did have these corrections to be somewhat higher, and they were. Table ~ contains the gender differentials in pay based on the P.U.S. samples, corrected for part-time and part-year workers but not for education or experience. In the first two columns are the sample numbers of men and women employed in each occupation in 1970 and 1980. These numbers are the samples upon which the earnings estimates are based. The second four columns show, for each year, the median hourly earnings of men in the occupation and the ratio of women's median hourly earnings to men's earnings. For the occupation computer scientist/systems analyst, the ratio of women's to men's median hourly earnings was 0.75 in 1970 and 0.74 in 1980. Again, such constancy ~ noteworthy since the year prior to the census) and dividing that by the hours they worked in an average week. Because the data are based on the respondents' estimate of the average hours worked, the data are subject to possible errors.

158 h 4= o V ~4 5: r ~ Cal ._ o 3 0 so no, a ~ o ·— Ch _ ~ o ~ t- ¢ - DO ·~ hi. 0 43 ~ _ ._ 0 o m ~ ~ C) 0 be ._ - C~ .~ O ~3 ~ ,0 O 00 1-= ~ o ~ ~ _ ~ I;; ~ ELI 3 ~ o E ~ o ~ ~ he. CO CC to CD o o o . . . Cat ~ ~ . . . o Us ~ CO o CO o 0 ~ ~ en 0 a) o its Its o o to C53 o 0 us ~ Ale em ~ ~ o: ~ ~ Ct U: ~ C9 ~ OD ~n ~ ~ oo a~ ~ S ~ S ~ ~ ~ S ~ ~ ~ t) o 0 0 Q O (3 ;) V V oo e~ u O ~ a: e~ ~ e~ e~ CO ~ ~ Ut CD oo oo CO _ _ _ _ . CO o ~ oo ~ o ~ O u~ e~ - E~ . . oo - ~n o L. m - C: o 00 · _` 00 ,Q . . C~ C53 ~4 ~_ g ._ o C: S" S" o o o ._ ._ ~ C. c4 m X ._ ct ~: ~ t- Q) U. . . o OO a, v o . . v 0 ~h U] ~

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 8 Employment and Hourly Earnings for Men and Women in Three Computer-Related Occupations in All Industries, 1970 and 1980 159 Total Median Hourly Earnings Number Employed Men's F/M Earnings (sample size) Earnings Ratio Occupation 1970 1980 1970 1980 1970 1980 Computer scientists/ , systems analysts 105 199 $5.17 $10.19 0.75 0.74 Computer programmers 155 312 4.10 8.12 0.85 0.83 Computer operators 106 456 3.15 6.38 0.63 0.69 NOTE: See Appendix B for the definition of computer-related occupations. For 1980, computer operators included computer equipment operator supervisors. SOURCE: Calculated by authors from U.S. Census (1/1000 P.U.S. Tape, 1970 and 1980~. employment more than doubled over the decade. It is also inter- esting that correcting for part-time and part-year workers does not increase this differential to much above the estimate Mom the published data for computer specialists. Among programmers, there was still a gender differential after correcting for part-time and part-year workers, although the gap was smaller than that for computer scientists and systems analysts. The ratio of women's to men's median hourly earnings was 0.85 in 1970 and 0.83 in 1980 again, remarkably constant given the more than doubling of employment in the occupation. For computer operators there was a rise in the differential with the correction for part-time and part-year workers. In 1970, when women had only 29 percent of the jobs, the female/male ratio of median hourly earnings was 0.63. By 1980, the gender division of labor had reversed and women dominated the occupation, with 59 percent of all the jobs. Still, the earnings ratio was 0.69, quite close to what it had been 10 years earlier. Although the sources are based on different measures and samples, they both show that women earn substantially less than men in these three occupations and that these ratios do not change much over time. Clearly, earnings equity is not an automatic result of the existence or the growth of these occupations.

160 COMPUTER-RELATED OCCUPATIONS Despite the fact that women earn less than men in computer programming and computer science/systems analysis, professional computer-related occupations are financially attractive for women. Relative to what professional women earn in other occupations requiring similar years of educational attainment, computer pro- gramming and systems analysts positions enable women to earn at the top of the female earnings hierarchy. Table 9 shows that for women in 1981, computer systems analysts was the second highest paying occupation and computer programming the seventeenth highest. In both of these occupations, women earned more on av- erage than they did in secondary school teaching (Rytina, 1982), despite the fact that in 1980 the mean educational attainment was only 15.75 years for women systems analysts and only 14.88 years for women computer programmers. For women secondary school teachers, the mean educational attainment was 16.43 years in 1980 (according to P.U.S.~. EARNINGS REGRESSIONS Earnings differentials between women and men in the same occupations are in part a result of differences between women's and men's human capital and productivity and in part a result of wage discrimination flower payment to women even after human capital and productivity have been held constant). To what extent is wage discrimination against women present in the relatively new computer occupations? Does employment in a high-tech indus- try lessen wage discrimination within the computer occupations? In order to answer these questions, we ran OLS (ordinary least squares) regressions for the three computer-related occupations in 1970 and 1980 on the natural log of hourly earnings using the P.U.S. samples. The independent variables we used to proxy hu- man capital and productivity were determined by the availability of data. The census 1/1000 P.U.S. reports age, years of education, and gender but not years of work experience, type of degree, or field of college major. We used AGE and AGE2 as continuous variables. We divided years of education into six categories, re- flecting the fact that number of years of education affects one's position in the labor market in a discontinuous fashion. The cat- egories are: eight years or less; some high school; high school graduate; some college; college graduate; and more than a college

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 9 Occupations with Highest Median Weekly Earnings for Women Employed Full Time in Wage and Salary Work, 1981 Annual Averages 161 Occupational Titlea Female Earningsb Operations and systems researchers and analysts Computer systems analysts Lawyers Physicians, dentists, and related practitioners Social scientists Teachers, college and university Postal clerks Engineers Ticket, station and express agents School administrators, elementary and secondary Life and physical scientists Health administrators Public administration officials and administrators, not elsewhere classified Vocational and educational counselors Registered nurses Personnel and labor relations workers Computer programmers Editors and reporters Secondary schoolteachers Librarians $422 420 407 401 391 389 382 371 370 363 357 357 337 336 331 330 329 324 321 318 aOccupations listed are those in which female employment was 50,000 or more in 1981. Excludes earnings from self-employment. SOURCE: Rytina (1982), based on Current Population Surrey data. degree. The category "some colleges was the reference group and was omitted from the regression. These variables have some serious deficiencies as indicators of education and productivity. Although age is often used as a proxy for work experience, a component of human capital, age is less likely correlated with years in the labor force for women than it is for men, and is, therefore, a less than satisfactory proxy for experience. Level of education, although frequently used as a measure of human capital and productivity, says nothing about the quality or type of education. Because our controls for human capital and productivity are problematic, the gender variable is

162 COMPUTER-RELATED OCCUPATIONS a poor measure of discrimination.7 We use age and level of education as control variables only because we have no others; we have traded off poor human capital proxies for the relatively large samples of these three occupations which the Census provides. In addition to human capital variables, the regressions have two dummy variables: GENDER (equal to one for women) and HTECHIND (equal to one for those employed in a high-tech industry). Race and ethnicity could not be included as variables because the number of minorities in the P.U.S. sample was too small. The combined effects of the human capital variables and the two dummy variables, GENDER and HTECHIND, on salary were tested in an additive model.8 For computer scientists/systems analysts in 1970, age, gender, education, and industry type explained 20 percent of the variance in the log of hourly earnings (see Table 10~; however, only having a college degree and older age were significantly related to earnings. By 1980, only 10 percent of the variance was explained by these four factors. Age and a college degree still significantly increased earnings, but gender less significantly decreased earnings. Holding the other variables constant, having a B.A. or B.S. increased one's salary by 27 percent, while being a woman decreased it by 20 percent. The drop in explained variance suggests that by 1980 other unmeasured factors were beginning to affect salaries for this occupation. Being in a high-tech industry was not significantly related to earnings for computer scientists/systems analysts. For computer programmers, we saw a more dramatic trend. In 1970, age, education, gender, and industry explained almost none of the variance. However, by 1980, 20 percent of the variance was explained by these factors, and all but industry type were significantly related to earnings. Age and having a B.A./B.S. or more than a B.A./B.S. significantly increased earnings, while gen- der, somewhat less significantly, decreased earnings. The changes in the effects of age and education may reflect the formalization 7 See Strober and Reagan (1982) for a discussion of the problems of using regression techniques to measure discrimination. ~ We also ran regressions with the education and high-tech variables interacted with gender, testing a multiplicative model. No interactions were significant in 1980, the few interaction terms that were significant in 1970 could have been so by chance, and most residuals in the additive model were acceptable. Thus, we followed the usual statistical practice of reporting the more parsimonious model (Chatterjee and Price, 1977:78-85~.

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 10 Earnings Regressions for Computer Scientists/Systems Analysts, Computer Programmers, and Computer Operators, lg7() and 1980 163 Computer Scientists/Systems Analysts Variable Mean (S.D.) B 1970 1980 Variable Ln. of hourly earnings 1.7 (0.44) Education: up to 8 years 0.01 (0.10) 0.53 Some high school 0.02 (0.14) -0.23 High school 0.29 (0.45) -0.05 BA or BS 0.23 (0.42) 0.26** More than BA/BS 0.22 (0.42) 0.15 Age 34.69 (9.77) 0.07*** Age2 1,297.66 (797.95) _0.006** Gender (= 1 if female) 0.14 (0.35) -0.07 High-tech industry 0.29 (0.45) 0.13 Constant 0.02 2 105.0 Adjusted R .20 Ln. of hourly earnings 2.16 (0.63) Education: up to 8 years 0 (0) Some high school 0.01 (0.07) 0.21 High school 0.13 (0.33) 0.14 BA or BS ~ 0.31 (0.46) 0.27** More than BA/BS 0.27 (0.45) 0.15 Age 34.58 (8.39) 0.09* * Age2 1,265.98 (637.18) -0.001*. Gender (= 1 if female) 0.25 (0.43) -0.20* High-tech industry 0.44 (0.50) 0.06 Constant 0.10 2 199.0 Adjusted R .10 Computer Programmers Mean (S.D.) B 1970 Ln. of hourly earnings 1.30 (0.79) Education: up to 8 years 0.0 (0) Some high school 0.02 (0.14) 0.29 High school 0.23 (0.42) 0.09 BA or BS 0.37 (0.49) 0.07 More than BA/BS 0.07 (0.26) 0.04 Age2 29.65 (8.05) 0~04 Age 943.52 (568.57) -0.0002 Gender (= 1 if female) 0.32 (0.47) -0.16 High tech industry 0.36 (0.48) 0.05 Constant 0.50 2 155.0 Adjusted R -.01 1980 Ln. of hourly earnings 2.01 (0.57) Education: up to 8 years 0.0 (0) Some high school 0.02 (0.15) -0.16

164 TABLE 10 (continued) COMPUTER-RELATED OCCUPATIONS Variable Computer Programmers Mean [S.D.) B High school 0.17 (0.38) -0.006 BA or BS 0.31 (0.46) 0.22~** More than BA/BS 0.15 (0.36) 0.24*** Age2 32.28 (8.76) 0.10* ~ * Age 1,118.15 (656.98) -0.001*** Gender (= 1 if female) 0.29 (0.45) -0.12* High tech industry 0.32 (0.47) 0.04 Constant -0.13 2 312.0 Adjusted R .20 Variable Computer Operators Mean (S.D.) B 1970 Ln. of hourly earnings 1.16 (0.67) Education: up to 8 years 0.01 (0.10) -0.14 Some high school 0.07 (0.25) -0.12 High school 0.53 (0.50) 0.02 BA or BS 0.01 (0.10) 0.77 More than BA/BS 0.01 (0.10) 0.12 Age2 28.80 (9.40) 0.04 Age 917.07 (675.91) -0.0004 Gender (= 1 if female) 0.21 (0.41) -0.70*** High-tech industry 0.25 (0.43) -0.10 Constant 0.51 2 106.0 Adjusted R .16 1980 Ln. of hourly earnings 1.70 (0.58) Education: up to 8 years 0.01 (0.10) 0.01 Some high school 0.04 (0.19) -0.16 High school 0.48 (0.50) -0.01 BA orBS 0.11 (0.31) 0.04 More than BA/BS 0.02 (0.15) 0.09 Age2 32.08 (10.77) 0.06*** Age 1,145.03 (812.62) -0.001*** Gender (= 1 if female) 0.53 (0.50) -0.27*** High-tech industry 0.20 (0.40) 0.18*** Constant 0.59 2 456.0 Adjusted R .20 * p < .10. ** p ~ .05. *** p < .01. NOTE: See Appendix B for the definition of computer-related occupations. For 1980, computer operators included computer equipment operator supervisors. SOURCE: Calculated by authors from U.S. Census (1/1000 P.U.S. Tapes, samples for 1970 and 1980~.

MYRA H. STROBER AND CAROLYN L. ARNOLD 165 of qualifications for this occupation; whereas earlier, people with diverse education and experience were recruited an] trained into it, by 1980 there were more formalized career ladders and more institutional training in the field. The beginning of an effect based on gender suggests that some differences based on gender may also have been in the process of becoming institutionalized. For pro- grammers, being in a high-tech industry did not appear to affect salaries when other variables were held constant. It may be that for both computer programmers and computer scientists/systems analysts, it is employment in the computer industry specifically, rather than in high-tech industries in general, that has a positive effect on salary. Computer operators show a different pattern. For both 1970 and 1980, 16-20 percent of the variance was explained by age, education, gender, and industry. Being a woman decreased one's salary significantly in both years. In 1970, gender was the only significant variable in the regression; being female, all other vari- ables held constant resulted in a 50 percent decrease in salary.9 By 1980, higher age and working in a high-tech industry also contributed significantly to higher earnings, but education was still not significant. In 1980, the effect of gender on salary, while still highly significant, was much smaller (27 percent) than it had been in 1970. These results reflect that this occupation has few educational requirements, that training takes place largely on the job, and that jobs in high-tech industries pay more. However, it also points out that women with the same education and age are still paid less than men, and that this was true both when the occupation was predominantly male (in 1970) and when it was predominantly female (in 1980~. 9 When the log of earnings is the dependent variable, the coefficient (b) on an independent variable is approximately the percentage eject of a unit change in that variable on the dependent variable. That approximation is worse the larger the absolute size of b. The enact effect is em. The only coefficient on gender that was large enough for the exact effect to be appreciably different from the approximate effect was for computer operators in 1970: en 70~i = _.50.

166 COMPUTER-RELATED OCCUPATIONS A CLOSER LOOK AT EARNINGS AND EMPLOYMENT DIFFERENCES BY INDUSTRY How does the labor market operate to pay men and women differentially even when they are in the same occupation and are similar with respect to age and level of education? In her book on clerical employment, Blau (1977) reports that often women and men in the same occupation in the same city earn different salaries because they work for different firms women for low-wage firms and men for high-wage firms. It may be that, analogously, women and men in computer occupations earn different salaries in part because they work in different industries men in high-paying industries and women in lower-paying "end-user" industries. These differences may not have shown up in the reported re- gressions because the industry dummy variable divided industries into only two groups, high tech and non-high tech. Unfortunately, the P.U.S. sample sizes in each industry are too small to include major industry groups as variables in the reported regressions. However, Table 11, based on the census publication, Occupation by Industry, shows that within these three computer-related oc- cupations in 1980, women and men age not employed in equal proportions across major industry groups. In the first column of Table 1l, we find the estimated num- ber of total employees (men and women) in each occupation by industry. The second column contains the percentage of women in the occupation for each industry. The third column shows the men's mean annual earnings in the occupation by industry. In the fourth column we have calculated the ratio of the women's mean annual earnings to men's mean annual earnings for each indus- try. The first row of each occupation contains these data for all industries combined. The second row isolates these data for part of the computer industry, based on the census three-digit indus- try category, Electronic Computing Equipment Manufacturing." The remaining rows show the data in the major two-digit industry groups. Several industry groups (agriculture, forestry, and fish- ing; mining; construction; personal services; and entertainment and recreation) have been removed from the analysis altogether, because they employ so few persons In computer-related occupa- tions. As shown in Table Il. women are 22.3 percent of computer scientists/systems analysts in all industries, but only 14.9 percent

MYRA H. STROBER AND CAROLYN L. ARNOLD 167 of computer scientists in electronic computing equipment manu- facturing. Among the major industry groups listed in Table 11, the percentage of women employed ranges from 16.9 percent in mining to 35.6 percent in finance, insurance, and real estate. For all industries, the men's mean annual earnings for computer scien- tists/systems analysts is $23,405; earnings figures are not available for electronic computing equipment manufacturing. Among ma- jor industry groups, men's mean annual earnings range from a low of $20,296 in professional and related sciences to a high of $26,031 in mining. Note that in mining, where the percent women employed is lowest, men's earnings are highest; conversely, where the percent women employed is highest (in finance, insurance, and real estate) men's earnings are the second lowest. To examine the relationship between the percent women and men's mean annual earnings across major industry groups we computed a rank correlation coefficient. For computer scien- tists/systems analysts the rank correlation coefficient is -.95, significant at the 99 percent confidence level. However, because the sample sizes are relatively small, we also computed the rank correlation coefficient, taking into account the standard errors of the percentages and means.~° The corrected rank correlation coefficient is -.68, not significant at the 95 percent confidence level. The ratio of female/male (F/M) earnings for computer s entist/systerns analyst is 0.73 for Al industries and ranges from 0.69 in manufacturing to 0.82 in transportation, communications, and public utilities. The data do not suggest a relationship across industries between F/M earnings and percent women or between F/M and men's mean annual earnings, mainly because of the lack of wide variation in the F/M earnings ratio. it The standard errors for the percentages were calculated from Table B or the formula below the table in Appendix C in Occupation by Indwiry (Bureau of the Census, 1972, 1984~. This was multiplied by 1.2, the appropriate design effect factor for a cross-tabulation of industries and occupations, as shown in Table C of Appendix C. A 95 percent confidence interval was created by calculating two standard errors on either side of the estimate. The standard errors for means were calculated using the formula in Appendix C, page C-2, for standard errors of means. The variances needed in that formula were not provided by the Census, so a conservative guess was used of a standard deviation equal to 5,000 for all salary distributions. This was squared to produce the estimated variance. A 95 percent confidence interval was created by calculating two standard errors on either side of the estimate.

168 COMP UTER -RELA TED O CC SPA TI ONS TABLE 11 Employment and Earnings in Three Computer-Related Occupations for All Industries, Electronic Computing Equipment Manufacturing, and in Selected Major Industry Groups, 1980 Men's Mean F/M Total Percent Annual Earnings Occupation Employment Women Earnings Ratio Computer Scientists/ Systems Analysts All industries 200,684 22.3 $23,405 0.73 Electronic computing equipment manufacturing 22,129 14.9 n.a. n.a. Mining 2,159 16.9 26,031 0.77 Construction 1,500 22.3 22,661 0.70 Manufacturing 65,606 17.5 24,093 0.69 Transportation, communications, and public utilities 12,947 26.8 23,812 0.82 Wholesale trade 8,953 21.4 22,840 0.80 Retail trade 5,336 26.5 22,195 0.71 Finance, insurance, and real estate 18,051 35.6 21,381 0.78 Business services 46,384 19.9 23,288 0.74 Professional and related services 15,241 28.4 20,296 0.68 Public administration 23,415 24.4 24,991 0~75 Computer Programmers All industries 313,179 31.1 S17,967 0.73 Electronic computing 22,702 22.0 n.a. n.a. equipment manufacturing Mining 3,171 32.5 19,455 0.76 Construction 2,802 34.8 17,275 0.69 Manufacturing 93,010 26.5 19,037 0.75 Transportation, communications, and public utilities 22,537 34.9 19,704 0.78 Wholesale trade 11,477 32.1 18,064 0.67 Retail trade 10,052 33.6 16,400 0.73 Finance, insurance, and real estate 39,749 36.6 16,774 0.77 Business services 63,423 28.0 17,826 0.72 Professional and related services 35,352 34.3 12,353 0.83 Public administration 29,635 35.9 18,868 0.74 Computer Operators All industries Electronic computing equipment manufacturing 7,175 408,475 59.0 314,203 0.65 46.6 n.a. n.a.

MYRA H. STROBER AND CAROLYN L. ARNOLD TABLE 11 (continued) 169 Men's Mean Total Percent Annual Earnings Occupation Employment Women Earnings Ratio Mining 4,647 42.9 16,041 0.63 Construction 5,758 68.8 14,833 0.67 Manufacturing 98,886 55.3 16,079 0.63 Transportation, communications, and public utilities 35,852 56.7 17,067 0.69 Wholesale trade 30,876 73.2 14,213 0.64 Retail trade 26,140 67.4 13,026 0.62 Finance, insurance, and real estate 63,660 57.3 12,878 0.66 Business services 43,697 45.8 13,010 0.65 Professional and related services 54,466 67.2 10,024 0.80 Public administration 39,185 59.3 15,237 0.65 NOTE: Computer-related occupations are defined in Appendix B. The Electronic-Computing Equipment Manufacturing Industry is defined as Census Industry Code 322 in 1980. SOURCE: Bureau of the Census (1984:Tables 1 and 2~. Among computer programmers, women are 31.1 percent of employees in all industries but only 22.0 percent in electronic computing equipment manufacturing. As in the case of computer scientists/systems analysts, women are more likely to be employed in end-user industries. Across major industry groups, the percent women employed ranges from 26.5 in manufacturing (which in- cludes the manufacturing of electronic computing equipment) to 36.6 percent in finance, insurance, and real estate. Men's mean annual earnings are $17,967 for all industries; among major in- dustry groups, they range from a low of $12,353 in professional and related services to a high of $19,704 in transportation, com- munications, and public utilities. However, unlike the situation for computer scientists/systems analysts, neither the uncorrected nor the corrected (for standard error) rank correlation coefficient shows a significant relationship between percent women's and men's mean annual earnings. The ratio of F/M earnings for programmers does not vary across major industry groups. The ratio for all industries is 0.73;

170 COMPUTER-RELATED OCCUPATIONS for all but one major industry group, the F/M earnings ratio is within a few percentage points of the mean. However, in professional and related services the ratio is 0.83. It is interesting that in this industry men's earnings are the lowest for all of the industries listed. Among computer operators, women are 59.0 percent of those employed in all industries, but only 46.6 percent of those em- ployed in the manufacture of electronic computing equipment. As in the case of computer scientists/systems analysts and computer programmers, women computer operators are more likely to be employed in end-user industries than in the computer manufac- turing industry. Among major industry groups, the percentage of women employed ranged from a low of 42.9 percent in mining to a high of 73.2 in wholesale trade. Men's mean annual earnings were $14,203 for all industries and ranged from a low of $10,024 in professional and related services to a high of $17,067 in transportation, communications, and public utilities. There is no systematic relationship between men's mean annual earnings and percentage of women in the industry. The ratio of F/M earnings for computer operators was 0.65 for all industries. As In the case of the other two occupations discussed, the F/M earnings ratio did not vary much across in- dustries. The only "outliner was in the professional and related services industry where the ratio was 0.80. And, as in the case of computer programmers, this industry had both the highest F/M earnings ratio and the lowest men's mean annual earnings. Our data suggest that in computer occupations, women are more likely to be found in end-user industries than in the computer manufacturing industry itself. In some occupations, particularly computer scientists/systems analysts, women also may more likely be found in industries where men are lower paid. However, it is not possible at this point to test this hypothesis definitively. If we were to disaggregate industry groups further, the sample sizes wouIc! become even smaller and the standard errors would rise accordingly. ~ . 11 Kraft and Dubnoff found similar results in a 1982 survey of ``software specialists" in Boston. They found women concentrated in the "worst paying industries (Kraft and Dubnoff, 1983~.

MYRA H. STROBER AND CAROLYN L. ARNOLD CONCLUSIONS F INDINGS 171 1. Although computer-related occupations are of recent ori- gin, they are not gender-neutral. The computer field was sired by the fields of mathematics and engineering, and the newly born prestigious and technical jobs quickly took on the gender desig- nation of the parent fields. Computer engineering and electronic technical work employ very few women. On the other hand, data entry, which quickly took on the characteristics of clerical work, became a virtually exclusive female preserve. Production work, too, is preponderantly female (see Table 1~. Computer programmers were female when the occupation first emerged, but very shortly after the computer was introduced, men began to fill the emerging jobs. Although women have increased their representation in the jobs of both programmer and analyst, women remain less than a third of the incumbents of these occupations. While the occupation of computer operator did not seem irnrnediately to be gender typed, in that it was preponderantly female in 1960 and male In 1970, it was becoming preponderantly female again by 1980. 2. When the difference between the percentage of women in the total labor force and the percentage of women in each computer-related occupation is used as a measure of occupational segregation, the level of segregation was approximately the same in 1970 and 1980, except for computer operators (see Table 2~. 3. Among four computer-related occupations found in all industries—computer scientists/systems analysts, computer pro- grammers, computer operators, and data-entry operators the higher the status and pay of the occupation, the more white men were overrepresented, compared with their representation in the labor force as a whole, and the more minority men and women of all racial and ethnic groups were underrepresented. In occupa- tions with much lower pay and status, the presence of white men dropped to much below their percentage in the labor force, the percentage of women of all racial and ethnic groups became much higher than their percentage of the labor force, and the percentage of minority men approached their labor force representation (see Table 4~.

172 COMPUTER-RELATED OCCUPATIONS 4. Within high-tech industries, most men were in produc- tion, professional/technical, or managerial occupations, while most women were in production or clerical occupations. Women and men were equally likely to be in production occupations. However, men were more likely to be in managerial and pro- fessional/technical occupations in high-tech industries than in non-high-tech industries; women fared worse in these occupations in high-tech industries than in other industries (see Table 6~. 5. Within the occupations of computer scientists/systems an- alysts, computer programmers, and computer operators, women's mean annual earnings and women's median hourly earnings were less than those of men. In addition, the ratios of women's to men's earnings generally remained constant between 1970 and 1980 (see Tables 7 and 83. 6. Within the three occupations analyzed, women's hourly earnings were generally less than those of men, even after age, level of education, and high-tech versus non-high-tech industry were held constant (see Table 10~. 7. Women employed as systems analysts, programmers, and computer operators were more likely to be found in end-user industries than in the computer manufacturing industry itself. Within the three computer-related occupations, women were paid less than men no matter in what industry they were employed (see Table 11~. We found some evidence that in several occupations, particularly computer scientists/systems analysts, women may be more likely to be found in industries where men are lower paid. However, this hypothesis has not been tested definitively here and requires further investigation. These findings dispel the myth that high tech is automatically a great equalizer. High tech may produce integrated circuits, but it does not necessarily produce an integrated work force or eliminate the female/male earnings differential. DISCUSSION Although a detailed explanation of the implications of these findings is beyond the scope of this paper, it is likely that the explanations can be found in the discussions of gender segregation and earnings differentials commonly found in the literature (see Blau and Jusenius, 1976; Cain, 1976; Amsden, 1980; Sokoloff,

MYRA H. STROBER AND CAROLYN L. ARNOLD 173 1980; Strober, 1984; Reskin, 1984; and Reskin and Hartmann, 1986~. These theories focus on women's own behavior, on em- ployer discrimination, and on the interactions of labor markets and gender relations in society. However, to devise the various types of policies that are required to change existing patterns of gender segregation and earnings differentials for the computer industry and computer-related occupations reported here, more research on the dynamics of each aspect of gender segregation is needed. More detailed research on the differences in women's em- ployment between and within industries and between and within firms would identify the bottlenecks preventing the gender integra- tion of occupations. Attention needs to be given to the processes by which women are allocated and/or allocate themselves into the lower-paid occupations and industries. This involves investigating how employers structure and define occupations and career lad- ders and distribute skilled job applicants and workers in ways that result in gender-segregated occupations and industries. Research also needs to be done on the degree to which technologically trained women (and some men) self-select out of certain occupa- tions or industries because a certain definition or culture for the occupation or industry precludes respect for the participation of people with different work styles or cultures. Some explanations have been advanced for women's low rem resentation among the specialized computer-related occupations. For example, DeBoer (1984) argues that women are still more likely than men to exclude themselves from advanced math and science training: even when women in high school and college science perform at a higher level than their male classmates, they have a higher drop-out rate. He proposes that teachers in secondary and postsecondary education make special efforts to acknowledge the skills of talented women. Hacker's work (1981), however, suggests that merely encour- aging women may not be sufficient to change their educational decisions, since women's decisions to exclude themselves from technical fields may be related in part to a dislike of the fields' "culture. Hacker, based on research at a technical institute, argues that there is a "culture of engineering" that includes an extension of the profession's formal objectification and control of the natural world to an informal objectification of women. It may be that a distaste for being part of the "engineering cul- ture~ also leads technically trained women to exclude themselves

174 COMPUTER-RELATED OCCUPATIONS from certain sectors of the computer industry. If an engineering culture appears most strongly in those sectors and industries of the computer field that are at the technological forefront and most competitive technologically, then it may be that those sectors and industries are the least appealing to women. However, it may not be accidental that these sectors have the strongest engineering culture. In terms of Strober's theory (1984), men who work in these intellectually challenging and highly lucrative sectors may acquire the habits of the "engineering culture" in part precisely to keep women out. The work style and work pressures in the most technologically competitive sectors of the computer industry may also keep many women out. While firms in all industries must remain compet- itive with similar firms, the computer industry, a new industry with a steady stream of technological breakthroughs, has some unique pressures: to make and increase profits in a competitive nonoligopolistic environment, stay on the technological forefront, and stay ahead of not only young ant! old domestic companies but their Japanese counterparts as well. These financial and tech- nological pressures are intensified as each firm tries to survive and succeed before the industry "shakes down." There is much pressure on workers in the computer industry to maintain high levels of productivity, including overtime work and other forms of commitment to the success of the firm. Women who want to succeed have to put in long, hard hours of work, and this may be a barrier for women (and men) who are trying to balance their home and work lives. We have presented evidence of gender segregation of the high- tech industries and highly technical computer occupations. At the same time we have called for research-assisted strategies to end these observed patterns of occupational inequity. While we en- courage women to enter these computer-related fields, we need to disseminate the findings of studies such as this one. This will make women aware of the channeling that leads them into less-prestigious, lower-paying occupations or "end-user" indus- tries within high-tech fields and help them develop strategies to counteract this channeling. At the same time this research can be user] to assist in developing policies to make occupations and workplaces more welcoming to both genders and more compatible with satisfying personal and family lives.

MYRA N. STROBER AND CAROLYN L. ARNOLD REFERENCES 175 ABAG (Association of Bay Area Governments) 1981 Silicon Valley and beyond: high technology growth for the San E`rancisco Bay Area. Working Papers on the Region' Economy, 2. Berkeley, Calif. Amsden, Alice H. 1980 Introduction. Pp. 11-38 in Alice H. Amsden, ea., The Econornic~ of Womcn and Work. New York: St. Martin's Press. Bielby, William T., and James N. Baron 1984 A woman's place is with other women: sex segregation within organizations. Pp. 27-55 in Sex Segregation in the Workplacc: [Fend`, E~plana;lrions and Rcmcdie`. Committee on Women's Employment and Related Social Issues. Commission on Behavioral and Social Sciences and Education, National Research Council. Washington, D.C.: National Academy Press. Blau, E`rancine D. 1977 Equal Pay in the Once. Lexington, Mass.: D.C. Heath. Blau, E`rancine D., and Wallace E. Hendricks 1979 Occupational segregation by sex: trends and prospects. Me Journal of Human Resources 14:197-210. Blau, Francine D., and Carol Jusenius 1976 Economists' approaches to sex segregation in the labor market: an appraisal. Pp. 181-199 in Martha Blaxall and Barbara B. Reagan, eds., Womcn and the Workplacc: The Implications of Occupational Segregation. Chicago: University of Chicago Press. Braverman, Harry 1974 Labor and Monopoly Capital. New York: Monthly Review Press. Bureau of the Census 1971 Census of Population. 19~70. Alphabetical Index of Indwtric, and Occupations. U.S. Goverment Printing Office. Washington, D.C.: U.S. Department of Commerce. 1972 Cenaw of Population. 19~70. Occupation by Industry. Final Report PC(2~-76. U.S. Government Printing Office. Washington, D.C.: U.S. Department of Commerce. 1982 Ccr~w of Population: 1980. Classified Indcz of Industries and Occupa- tior~, Final edition. U.S. Government Printing Office. Washington, D.C.: U.S. Department of Commerce. 1983 Data U~cr News 18(June):8-9. 1984 Ccnsu, of Population and Housing: 198a Subject Report, Occupa- tion by Industry. PC80-2-7c. U.S. Government Printing Office. Washington, D.C.: U.S. Department of Commerce. Cain, Glen G. 1976 The challenge of segmented labor market theories to orthodox theory: a survey. Journal of Economic Literature 14(December):1215- 1257. Chatterjee, Samprit, and Bertram Price 1977 Rc~rca~ion by Example. New York: Wiley and Sons, DeBoer, George E. 1984 Factors Affecting the Scicnec Participation and Pcrformanec of Women

176 COMPUTER-RELATED OCCUPATIONS in High School arid College. Hamilton, N.Y.: Colgate University, Department of Education. Dicesare, Constance Bogh 1975 Changes in the occupational structure of U.S. jobs. Monthly Labor Review 98(March):24-34. Greenbaum, Joan M. 1979 In the Name of E~icier~cy: Management Theory and Shoppoor Practice in Data Processing Work. Philadelphia: Temple University Press. Gross, Edward. 1968 Plus ga change...? The sexual structure of occupations over time. Social Problems 16:198-208. Grossman, Rachael 1980 Women's place in the integrated circuit. Radical America 14(Janu- ary/February):29-49. Hacker, Sally L. 1981 The culture of engineering: woman, workplace, and machine. Womcn's Studies Intcrr~atiorzal Quarterly 4~33:341-353. Kraft, Philip 1977 Programmers arid Mar~agers. New York: Springer-Verlag. 1979 The industrialization of computer programming: from program- ming to software production. Pp. 1-17 in Andrew Zimbalist, ea., Case Studies in the Labor Process. New York: Monthly Review Press. Kraft, Philip, and Steven Dubnoff 1983 Software workers survey. Computer World XVII(November 14~:3- 13. Lloyd, Cynthia B., and Beth T. Neimi 1979 17`c Economics of Sex Diffcrcntiab. New York: Columbia University Press. O'Neill, June 1983 The Determinants and Wage Effects of Occupational Segregation. Project Report. Washington, D.C.: The Urban Institute. Reskin, Barbara F., ed. 1984 Scz Segregation ire the Workplace: lFcr~d`, Explanation`, Remedied. Committee on Women's Employment and Related Social Issues. Commission on Behavioral and Social Sciences and Education, National Research Council. Washington, D.C.: National Academy Press. Reskin, Barbara F., and Heidi I. Hartmann, eds. 1986 Womer-'s Work, McrIe Work. Sex Segregation on the Job. Committee on Women's Employment and Related Social Issues. Commis- sion on Behavioral and Social Sciences and Education, National Research Council. Washington, D.C.: National Academy Press. Riche, Richard W., Daniel E. Heckers, and John U. Burgan 1983 High technology today and tomorrow: a small slice of the em- ployment pie. Monthly Labor Review 106(November) :50-59. Rogers, Everett M., and Judith K. Larsen 1984 Silicon. Valley Fever. New York: Basic Books. Rytina, Nancy F. 1982 Earnings of men and women: a look at specific occupations. Monday Labor Review 105 (April) :2 5-31.

MYRA N. STROBER AND CAROLYN L. ARNOLD 177 Sokoloff, Natalie J. 1980 Between Money and Lone: The Dialectics of Womer-'~ Come and Market Work. New York: Praeger. Strober, Myra H. 1984 Toward a general theory of occupational sex segregation: the case of public school teaching. Pp. 144-156 in Barbara F. Reskin, ea., Scz Sc~regahon in the Workplace: lFcndi, Explorations, Remedies. Washington, D.C.: National Academy Press. Strober, Myra H., and Barbara B. Reagan 1982 Sense and nonsense in measuring employment discrimination in higher education. Unpublished paper. Treiman, Donald J., and Heidi I. Hartmann, eds. 1981 Wooer, Work, and Wages: Equal Pay for Jobs of Equal Value. Washington, D.C.: National Academy Press. U.S. Department of Commerce, Office of Federal Statistical Policy and Standards 1980 Standard Occupational Classification Marshal. APPENDIX A: INDUSTRIES WITHIN MAJOR INDUSTRY GROUPS, BY HIGH-TECH AND NON-HIGH-TECH CATEGORIES Industrial categories that comprise the high-tech sector defi- nition used here are based on those designated Group III in Riche et al., 1983. This is a moderately inclusive definition containing 27 three-digit SIC industries and 1 four-digit SIC industry. Cen- sus Industry Codes which most closely matched these SIC codes were used to designate high-tech industries in the 1970 and 1980 Census Public Use Sample tapes. The Census Codes each include from one to several SIC industries, and some Riche-designated high-tech SIC industries are grouped with non-high-tech SIC in- dustries, so it is not possible to perfectly match the Riche Group III list using census data. Below are listed the census industrial categories that were designated for this study as high tech and non-high tech. For the high-tech industries, 1970 and 1980 Census Industrial Codes are shown in parentheses, with the 1970 code first followed by the 1980 code (Bureau of the Census, 1971, 1982~. HIGH-TECHNOLOGY INDUSTRY GROUPS (HIGH TECH) DURABLE MANUFACTURING Ordnance (258;248~; engines and turbines (177;310~; office and accounting machines (188;321~; electronic computing equipment

178 COMPUTER-RELATED OCCUPATIONS (189;322~; radio, television, and communication equipment (207; 341~; electrical machinery, equipment, and supplies, n.e.c. (208; 342~; aircraft and parts (227;352~; scientific and controlling instru- ments (239;371~; optical and health services supplies (247;3723; photographic equipment and supplies (248;380~. NONDURABLE MANUFACTURING Industrial and miscellaneous chemicals (281 and 368;192~; plas- tics, synthetics, and resins (348 and 349;180~; drugs and medicines (357;181~; soaps and cosmetics (358;182~; paints, varnishes, and related products (359;190~; agricultural chemicals (367;191~; not specified chemicals and allied products (1970;369~; petroleum refining (377;200~; miscellaneous petroleum and coal products (1970;378~. BUSINESS AND REPAIR S ERVICES Commercial research, development, and testing labs (729;7303; computer programming or computer and data-processing services (739;740~. NON-HIGH-TECHNOLOGY INDUSTRY GROUPS (NON-HIGH-TECH) DURABLE MANUFACTURING Lumber, furniture, stone, clay, and glass products, other metal industries, cutlery, handtools, hardware, other machinery, house- hold appliances, transportation equipment, clocks, toys, sporting goods. NoNDuRABLE MANUFACTURING Food, tobacco, textile, apparel, paper, printing, rubber, leather products.

MYRA H. STROBER AND CAROLYN L. ARNOLD BUSINESS AND REPAIR S ERVICES 179 Advertising, buildings services, personnel supply, business man- agement and consulting, detective and protective services, busi- ness services, automotive services and repair, electrical repair, miscellaneous repair. AGRICULTURE, FORESTRY, AND FISHERIES Agricultural production, crops; agricultural production, livestock; agricultural services; horticultural services; forestry; fisheries; fish- ing, hunting, and trapping. MINING Metal mining; coal mining; crude petroleum- and natural gas extractions; nonmetallic mining and quarrying, except fuel. . CONSTRUCTION General building contractors; general contractors, except building; special trade contractors; nonspecified construction. TRANSPORTATION, COMMUNICATION, AND PUBLIC UTILITIES Rail, bus, taxi, truck services, warehouses, U.S. Postal Service, water and air transportation, pipelines, miscellaneous transporta- tion services, radio, television, telephone, telegraph, electricity, gas, steam, water supplies, and sanitary services. WHOLESALE TRADE All wholesale trade of durable and nondurable goods sale of high-tech products. RETAIL TRADE , including All ret se} outlets for clurable and nondurable goods, including sale of high-tech products.

180 COMPUTER-RELATED OCCUPATIONS F INANCE, INSURANCE, AND REAL ESTATE Banking, savings and loans, credit agencies, securities and invest- ment, insurance, real estate and real estate insurance and law offices. PROFESSIONAL AND OTHER SERVICES Offices of doctors, dentists, chiropractors, optometrists, and other health practitioners and services; hospitals, nursing, and personal care services; legal services; elementary and secondary schools; colleges and universities; business, trade, and vocational schools; libraries; educational services; job training and vocational rehabil- itation; child care services; residential care; social services; muse- ums; art galleries; zoos; religious and membership organizations; engineering, architectural, and surveying services; accounting, au- diting, and bookkeeping services; and noncommercial educational and scientific research. PUBLIC ADMINISTRATION Offices of chief executive and legislative bodies and their advisory and interdepartmental committees and commissions; government civil rights and civil service commissions; offices providing support services for government such as accounting, personnel, purchasing and supply; courts; police protection; correctional institutions; fire protection; government legal counsel; public finance; tax and monetary policy; administration of educational programs; public health, social, manpower, and income maintenance programs; veterans' affairs; environmental protection; housing and urban development programs; regulatory agencies; national security; and international agencies.

MYRA H. S TR OBER A ND CA R OL YN L . A RNOLD APPENDIX B: COMPUTER-RELATED OCCUPATIONS 181 We define computer-related occupations using the following detailed occupation categories and codes from the 1970 and 1980 U.S. censuses: Census Occupation Categories Census Occupation Codes 1970 1980 Engineers 006-023 044-059 Electrical/electronic engineer 012 055 Computer specialists 003-005 064,229 Computer scientist/systems analyst 004-005 064 Computer programmers 003 229 Engineers and science technicians 15~162 Elect ric al/elect ronic engineering technicians Drafters 153 152 213-217, 223-225 213 217- Computer operators (includes computer and peripheral equipment operators) 343 308,309 Data-entry operators 345 385 Production workersa (includes crafts, precision production, operatives, transportation, laborers, and farm occupations) 401-824 473-889 Electronic assemblers (a category in 1980 only) within precision production Data-processing repairers Within precision production 475 683 525

182 Operatives, fabricators, transportation, and laborersa Within production (excludes farm, crafts, and precision production. Not precisely the same occupations in each year.) Assemblers Within operatives or COMPUTER-RELATED OCCUPATIONS 601-785 602 703-889 785 a There were changes in occupational coding and categorizing between the 1970 and 1980 Censuses (Bureau of the Census, 1983), which affected the occupations included in large occupational categories such as "Production Workers" and `'Operatives, Fabricators, Transportation, and Laborers. The category of Production Workers as defined here is so large that the occupations included are identical except for three which are included in 1970 and not in 1980: decorators and window dressers (1970 code 425~; inspectors, n.e.c. (1970 code 4523; and conductors and motormen, urban rail transit (1970 code 704~. Since few of these workers would be in the computer industry, we do not feel that this affects the results. However, the category of "Operatives, Fabricators, Transportation, and Laborers does differ significantly between the years, and therefore the data presented for 1970 and 1980 are not strictly comparable.

Next: III. Technology and Trends in WomenWomen »
Computer Chips and Paper Clips: Technology and Women's Employment, Volume II: Case Studies and Policy Perspectives Get This Book
×
Buy Paperback | $125.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This companion to Volume I presents individually authored papers covering the history, economics, and sociology of women's work and the computer revolution. Topics include the implications for equal employment opportunity in light of new technologies; a case study of the insurance industry and of women in computer-related occupations; a study of temporary, part-time, and at-home employment; and education and retraining opportunities.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!