4. A Gendered Job Carousel

Employment Effects of Computer Automation

Corinna Schlombs

"Social and cultural ideas- not the technologies- shaped which kind of data processing work was done by women or by men, and under which conditions."

In the late 19th and early 20th century, clerical work became firmly divided by gender, informed by late Victorian ideals of women as housewives. Women were relegated to the low-pay, dead-end aspects of data processing, starting with punch-card machines and, after the introduction of electronic computing, to data entry and word processing positions. Prior to this, the only option available to women, if they were from an urban, middle-class family, was to be a housewife (of course, women of color, immigrant women and women from farming families had long been working in textiles mills, manufacturing, and other low-pay positions). It was clear that a new idea of what it meant to be a respectable young woman was needed. The federal government had begun to hire women to take on clerical work during the Civil War, marking the beginning not only of clerical work for women, but also of women's entry into male-dominated occupations during times of war. Women also did data processing for the federal government during times of peace, such as the 1890 US Census, the first to use punch-card machines for counting, sorting, computing, and tabulating.


NCR tag printers as "integral part" of information systems. While "unit automatically produces a wide variety" of labels, women's labor remained.

The Census Bureau initially ran a day shift of all women and a nightshift of all men, with work comprised of first punching census data onto cards and then sorting and counting the cards on specially designed machines. Women proved to be 50% faster than their male counterparts, and the nightshift was abandoned. There are many possible factors that could have gone into this decision besides the speed of the women. For instance, it was difficult to pass work between shifts, and, to the dismay of supervisors, men and women working on the same machines were sometimes found to be passing romantic notes. The census was never fully gender segregated though, and the 1890 Census was tallied by 527 women and 137 men. The machines used for this, developed by Herman Hollerith, were huge, loud, and dirty, making it somewhat odd that they became a part of women's work in the US. In Germany, by contrast, the same Hollerith machines were kept in their own room and department, so as not to bother the other employees, and were completely run by men.

Various other government offices tried a variety of systems of men and women working together, some employing women only for routine tasks, others giving them the full range of tasks at hand. The Patent Office kept a strict division of labor, with separate copying rooms for men and women. The Bureau of Statistics, by contrast, let women perform all tasks just like their male counterparts. Private corporations followed in the footsteps of the federal government and began employing women to do clerical work, but unlike the federal government, where women and men worked alongside each other respectably, corporations kept their data processing strictly gender segregated. Men and women used separate entrances, staircases, elevators, doorways, and worked in different departments.

Women working in punch-card operations were paid less than their male clerical counterparts, but more than some other female clerical workers (typists, switchboard operators, and file clerks for example.) The pay scale for punch-card operators was likely low because it was a routine and low-skill job, and also because it was a female-dominated occupation. At the time, women were considered to be a part of a family with a single male wage-earner, and a woman's wage was to be a living wage that would allow her to subsist on her own, just barely. The transition of clerical work from punched cards to electronic computers in the 1950s had significant impacts on the workforce, such as changing the working conditions. Many women lost their former data processing jobs to computers (a far higher proportion than men who were displaced by computers), and men began to outnumber women in new computing departments (a reversal from punched cards).

"As men took control of electronic computers, one electronic data processing task became a female domain: data entry."


Data entry on punch cards was gendered as female labor. In 1971, Univac's verifying punch combined keypunching and verifying of data.

In the punch-card division of MetLife, the five highest paying positions had all been held by men, and the six highest paying positions in the new computer division were also held by men, all of whom earned around $7,200. Women had and continued to occupy the mid-level jobs, mostly earning between $3700-5000. The low-wage punch-card workers, the vast majority of whom were women, were entirely eliminated in the new department. Women became responsible for transferring data from physical recordings (payroll time cards, checks for payroll, insurance payments, etc.) to a format that the computer could read, a highly monotonous task that companies demanded be done quickly, creating a stressful environment that caused high rates of turnover among the women working these jobs.

The technology for data entry changed over three decades and by the late 1960s, rather than having to punch information onto cards to be read by computers or other machines, women entered data on a workstation and it was stored in floppy disks. Although these technological changes made the work less physically taxing, data entry remained a routine yet stressful occupation, requiring high levels of concentration and accuracy under tight time pressure. It had been thought that computers would bring more accuracy to data entry since they were not prone to human mistakes, but the computer is only as good as the data fed into it by the operator. Thus, women were not only subjugated to the routine work of data entry but had the burden of suspicion placed on them by data processing executives in the event of an error.


Bar-coding systems, such as NCR's model 280 at Montgomery Ward, launched "point of sale" retailing.

Women continued to be paid less than men in the computing field, with companies and conservative politicians arguing against equal pay, warning the public of the threat of cheaper female labor displacing men's wages. Women were still meant to be wives and mothers, and so many companies barred women from working after marriage. They also argued that the necessary amenities, such as extra bathrooms for women, ought to be taken out of women's wages. Many women continued to work past the 3-5 years that companies such as MetLife expected of them, some even staying despite being married. The assumption that women were not seeking career advancement gave many executives an excuse to deny promotions or better pay. Unions also excluded these women until the 1930s, partly because they were clerks and partly because they were women.

As the use of electronic computers increased through to the 1980s, so did the number of women working data entry positions. The 1980s also saw the introduction of word processors, bringing with them a second wave of clerical automation and integrating tasks that had previously been fragmented, creating highly skilled positions for women with college degrees, although these positions still offered little in the way of career advancement.

4. A Gendered Job Carousel