Sex Segregation At Work Research Paper

Customarily, ‘segregation’ denotes physical separation, as in school segregation by race. However, the term ‘sex segregation’ refers to physical, functional, or nominal differentiation in the work that women and men do. ‘Men’s work’ does not simply differ from ‘women’s work,’ however; predominantly male jobs tend to be better and more highly valued. This occurs both in vertical segregation which assigns men higher-level jobs than women in the same occupation, and in horizontal segregation in which the sexes pursue different occupations.

1. Sex Segregation And Sex Inequality

Segregation—whatever its form—is a key engine of inequality. The domains reserved for members of dominant and subordinate groups are not just different, they are unequal, with the more desirable domains reserved for members of dominant groups. Because economic, social, and psychological rewards are distributed through people’s jobs, segregation both facilitates and legitimates unequal treatment. Sex segregation reinforces sex stereotypes, thereby justifying the sexual division of labor it entails, and it reduces equal-status cross-sex contacts that could precipitate challenges to differentiation. In sum, workplace segregation benefits men and harms women (Cotter et al. 1997).

1.1 Consequences Of Sex Segregation

Job, establishment, and occupational segregation account for most of the differences in the rewards that men and women garner from employment. The most important of these differences is earnings. Estimates of the importance of occupational segregation for the pay gap between the sexes vary widely, depending on the analytic approach. An estimated one-third of the earnings gap results from occupational segregation in the USA and abroad (Anker 1998). Establishment level data for the USA, Norway, and Sweden by Petersen and his colleagues attribute about three-quarters of the gap to occupational segregation (Petersen and Morgan 1995). This approach may still underestimate the effect of occupational segregation on the pay gap: the amount of sex segregation in metropolitan labor markets affects how much segregation affects women’s and men’s earnings with all women losing pay and all men gaining in highly segregated markets, and no pay gap if occupational segregation were very low (Cotter et al. 1997, p. 725).

Men dominate the more lucrative jobs partly because employers reserve such jobs for men and partly because female work is devalued (England 1992). The under-valuation of female activities not only shrinks women’s pay, it also reduces their economic independence, familial power, and sense of entitlement.

Sex segregation also creates disparities in the sexes’ authority, mobility opportunities, working conditions, and chance to acquire skills. Segregation disproportionately relegates women to dead-end jobs or jobs on short career ladder, thus reducing their aspirations and opportunity for mobility ladders (Baron et al. 1986). This vertical segregation creates a ‘glass ceiling’ that concentrates women in the lower level positions and reduces their authority.

2. Theories Of Segregation

The most common theories of sex segregation emphasize the preferences of workers and employers. Hypothetically, gender-role socialization or a sexual division of domestic labor that encourages men to pick jobs that maximize earnings and women to select jobs that facilitate child-rearing lead to sex-differentiated preferences (England 1992). Men’s vested interests also hypothetically prompt them to exclude women from ‘men’s jobs.’ The universality of the assumed sexspecific preferences limits the value of preference explanations: because preferences theoretically vary between, not within the sexes, the sexes should be completely segregated. Empirical evidence casts doubt on preference theories: youthful occupational preferences only loosely related to the sex composition of adults’ occupation (see Jacobs 1989, 1999). Moreover, the sexes similarly value high pay, autonomy, prestige, and advancement opportunities, which should minimize between-sex differences.

Explanations focusing on employers’ preferences stemming from sex biases and attempts to minimize employment costs through statistical discrimination are limited by their emphasis on hard-to-measure motives. Although it stands to reason that employers influence job-level segregation because they assign workers to jobs, it is difficult to learn why. However, by examining the effects of training and turnover costs and skill on women’s exclusion and sex segregation, Bielby and Baron (1986) showed that employers discriminated statistically against women by imputing to individuals stereotypical characters of their sex, but contrary to neoclassical economic theory, this practice was neither efficient nor rational.

In sum, standard theoretical approaches to segregation predict universally high segregation rather than explaining variation in its level. Although these theories have stimulated considerable research, it is difficult to directly test them, so the importance of women’s and employers’ ‘choices’ remains a matter of debate. More fruitful explanations (summarized later) try to explain variation in segregation across aggregates or over time.

3. Measuring Segregation

Segregation is a property of aggregates, not individuals. Its measurement depends on how it is conceptualized. The most common conceptualization— the sexes’ different representation across work categories—is measured by the index of dissimilarity, D, whose formula is Σ f_i – m_i/2 ( f_i and m_i represent the percentages of female and male workers in each work entity). D is the percentage of either female or male workers who would have to change to a work category in which they are underrepresented for the sexes’ distributions across categories to be identical. The size of D depends both on the extent of segregation and the relative sizes of the work entities (e.g., occupation) so it is unsuitable for comparing segregation across populations with work entities of different relative sizes. A size-standardized variant of D which holds entity size constant permits comparing levels of segregation across populations.

Segregation is also conceptualized as the concentration of the sexes in work entities dominated by one sex. This conception is relevant to the thesis that women are disproportionately crowded into relatively few occupations. A simple measure of concentration is the proportion of each sex in the n occupations that employ the most men or women. (For an index of concentration, see Jacobs (1999) who warns that concentration measures assume that female and male occupations are aggregated to the same degree.)

A third conception of segregation is each sex’s probability of having a coworker in their job or occupation of the same or the other sex, designated as P* (Jacobs 1999). The value of P* depends on both the extent of segregation and the sex composition of the population.

Occupations can be described in terms of their sex composition, and are sometimes characterized as ‘integrated’ or ‘segregated’ or ‘female- or male dominated’ based on whether one sex is disproportionally represented. For example, in the USA, billing clerk is a segregated occupation because 89 percent of its incumbents are female, whereas women are just 47 percent of the labor force. The individual-level experience of segregation is the sex composition of her or his job or occupation, so a billing clerk holds a predominantly female occupation.

In 1992, Charles proposed using log linear methods to study sex segregation. Because log linear methods adjust for both the varying sizes of work entities and the sexes’ respective shares of the labor force they are useful for studying why the extent of segregation varies cross-nationally and over time. They also allow researchers to estimate the impact of independent variables on a sex’s representation in specific occupations or occupational categories, net of the occupational structure (Charles 1992, p. 488, 1998).

All measures of segregation are sensitive to the level of aggregation of the work units of interest. If segregation exists within units—which it usually does—the more aggregated the units, the less segregation will be observed. Although fully disaggregated data, such as establishment-level data, are preferable, they are hard to come by. Cross-national data are particularly likely to be aggregated because researchers need to make work units comparable across countries and because the data some countries collect are broadly aggregated.

4. The Extent Of Segregation

Most research on segregation focuses on the sexes’ concentration in different occupations. (An occupation comprises jobs that involve similar activities within and across establishments.) In a ‘typical’ country in the early 1990s about 55 percent of workers held either ‘male’ or ‘female’ jobs (Anker 1998, p. 5). Although countries vary considerably in the extent to which the sexes are segregated occupationally, in the 1990s most developed nations had segregation indices between 50 and 70 (Anker 1998). In 1997, the segregation index for detailed occupations in the USA was 53.9 (Jacobs 1999).

The sexes are most segregated in the Middle East and Africa. In the early 1990s, segregation indices across detailed occupations were at least 70 for Tunisia, Kuwait, and Jordan, largely because of Muslim proscriptions against contact between the sexes. Most of the EU societies and ‘transition economies’ (former Soviet satellites) had indices in the mid-50s. Segregation was lowest in Asian Pacific nations, with indices ranging from 36 (China) to 50 (Japan). China’s low segregation level results from its large, sex-integrated agricultural sector.

Among the Western developed countries, Italian and US workers were least segregated, and Scandinavians most segregated. In these countries, occupational segregation is positively related to women’s labor force participation; indeed the same factors facilitate women’s labor force participation and encourage sex segregation—paid parental leave and part-time work which lead to statistical discrimination against women and the shift of customarily female domestic tasks into the market economy (Charles 1992). Also contributing to cross-national differences in segregation is the relative sizes of manufacturing and service sectors, since men tend to dominate the former and women the latter (Charles 1998). Variation in national commitment to gender ideology also explains cross-national differences in segregation (Charles 1992).

4.1 Segregation Across Jobs

Much higher levels of sex segregation exist in establishment-level studies of job segregation (a job is a position in an establishment whose incumbents perform particular tasks) because people in the same occupation work in different establishments or different jobs (Bielby and Baron 1984; Tomaskovic-Devey 1993). For example, male lawyers are more likely than female lawyers to work for law firms, and within law firms men dominate litigation. Organizations’ characteristics and personnel practices affect how segregated they are. Small and large establishments are the most segregated. Small establishments are segregated because they employ workers of only one sex; whereas large bureaucracies have segregative personnel practices, such as sex segregated job ladders (Bielby and Baron 1984).

5. Trends In Segregation

In the last quarter of the twentieth century segregation fell in most of the world, although women continued to dominate clerical and service occupations, and most customarily male production jobs remained inaccessible to women (Charles 1998). Segregation increased during the 1970s and 1980s in Asian/Pacific countries; remained constant in Southern and Central European countries; and declined in the USA, Western Europe, the Middle East and North Africa, and other developing countries (Anker 1998, p. 328). In most countries declines reflected increased occupational integration as well as shifts in the occupational structure in which heavily segregated occupations shrank.

In the USA, increased integration resulted primarily from women’s movement into customarily male occupations. The higher pay in customarily male jobs draw women. Large numbers of men enter customarily female occupations only when they become much more attractive, and such changes are rare. The sizestandardized index was stable between 1990 and 1997, indicating that after 1990 real occupational integration halted, although fewer persons worked in heavily segregated occupations (Jacobs 1999).

5.1 Explaining The Decline In Segregation

According to the theoretical framework summarized above, the more similar the sexes become in their preferences and skills, the less segregated they will be. According to cross-national research, the narrowing of the education gap between the sexes has contributed to integration. In the USA, for example, as fewer women majored in education and more majored in business, educational segregation declined and occupations integrated. The shrinking experience gap between the sexes may also reduce segregation if it relieves employers’ misgivings over hiring women for jobs that involve firm-specific skills.

However, employers’ hiring and job-assignment practices are the proximate determinants of the level of segregation. Many establishments employ women in customarily male jobs only when there are not enough qualified men. Thus, organizational and occupational growth both foster integration if they cause a shortfall of male labor. Shortages have also brought about integration when occupations deteriorated in their pay, autonomy, or working conditions (e.g., pharmacist, typesetter; Reskin and Roos 1990) and thus became less attractive to men.

Employers who face economic penalties for segregating and whose job assignments are monitored are more likely than are others to integrate jobs (Baron et al. 1991). Thus, anti-discrimination and affirmative action regulations have fostered integration by making it illegal for employers to exclude workers from jobs on the basis of their sex. These rules have made a difference largely through ‘class-action’ lawsuits against employers. Anti-discrimination laws are ineffective in reducing segregation when regulations are not enforced and when class-action lawsuits are not permitted.

Finally, according to development theory, modernization replaces ascription with achievement in filling social positions. However, modernization is associated with the growth of economic sectors that most societies label as ‘women’s work.’ Thus, the net effect of modernization, as measured by GNP, has been segregative because it expands female dominated clerical and sales occupations (Charles 1992).

5.2 Nominal Integration

What appears to be genuine integration may be a stage in a process of re-segregation in which women are replacing men as the predominant sex. In the USA, for example, insurance adjusting and typesetting re-segregated over a 20-year period when these occupations changed from male to female dominated. For individual women, the corresponding process is the ‘revolving door’ through which women in male occupations pass back into customarily female occupations (Jacobs 1989).

5.3 Explaining Stability

Because the amount of sex segregation has changed so little, scholars have devoted as much attention to explaining stability as to explaining change. Unchanging levels of segregation may result from stability in the causal variables or from processes whose opposing effects cancel out each other.

Gender ideology strongly favors stability. Many occupations are sex-labeled, representing a cultural consensus that they are appropriate for one sex but not the other (e.g., childcare worker, automobile mechanic). Although these labels are not deterministic (indeed they sometimes vary across societies), they predispose employers to prefer sex-typical workers and workers to prefer sex-typical jobs. Although state policies have altered gender ideology, it is usually a conservative force.

Work organizations also resist change in customary business practices. Organizations’ structures and practices reflect the cultural environment present when they were founded, and inertia preserves these structures and practices. Such practices include sex-segregated promotion ladders, credentials that are required for jobs, and the use of workers’ informal networks to fill jobs. For example, recruitment through networks perpetuates the sex composition of jobs because people’s acquaintances tend to be of their same sex. Statistical discrimination or stereotype-based job assignments perpetuate the sex composition of jobs because they prevent employers from discovering that their stereotypes are unfounded. Thus, barring external pressures to alter job assignments, job segregation is unlikely to change.

Of course, stable levels of segregation may be misleading if they are the result of counteracting forces that have opposing effects on the level of sex segregation. For example, in the 1980s Japan’s occupations became slightly more integrated, but this was offset by the increased number of workers employed in occupations that were either male or female dominated. The net effect of the growth of female dominated clerical and service sectors in most economically advanced countries will cancel out some shift toward greater integration within occupations.

Bibliography:

1. Anker R 1998 Gender and Jobs: Sex Segregation of Occupations in the World. International Labour Office, Geneva, Switzerland
2. Baron J N, Davis-Blake A, Bielby W T 1986 The structure of opportunity: How promotion ladders vary within and among organizations. Administrative Science Quarterly 31: 248–73
3. Baron J N, Mittman B S, Newman A E 1991 Targets of opportunity: Organizational and environmental determinants of gender integration within the California Civil Service 1979–1985. American Journal of Sociology 96: 1362–1401
4. Bielby W T, Baron J N 1984 A woman’s place is with other women. In: Reskin B F (ed.) Sex Segregation in the Workplace: Trends, Explanations, Remedies. National Academy Press, Washington, DC
5. Bielby W T, Baron J N 1986 Men and women at work: Sex segregation and statistical discrimination. American Journal of Sociology 91: 759–99
6. Charles M 1992 Cross-national variation in occupational sex segregation. American Sociological Review 57: 483–502
7. Charles M 1998 Structure, culture, and sex segregation in Europe. Research in Social Stratification and Mobility 16: 89–116
8. Cotter D A, DeFiore J A, Hermsen J M, Kowalewski B, Vanneman R 1997 All women benefit: The macro-level effects of occupational integration on earnings inequality. American Sociological Review 62: 714–34
9. England P 1992 Comparable worth. Theories and Evidence. Aldine de Gruyter, New York
10. Gross E 1968 Plus ca change … ? The sexual structure of occupations over time. Social Problems 16: 198–208
11. Hakim C 1992 Explaining trends in occupational segregation: The measurement, causes, and consequences of the sexual division of labor. European Sociological Review 8: 127–52
12. Jacobs J A 1989 Revolving Doors. Stanford University Press, Stanford, CA
13. Jacobs J A 1999 The sex segregation of occupations: Prospects for the 21st century. Forthcoming In: Powell G A (ed.) Handbook of Gender in Organizations. Sage, Newbury Park, CA
14. Petersen T, Morgan L A 1995 Separate and unequal: Occupation-establishment sex segregation and the gender wage gap. American Journal of Sociology 101: 329–65
15. Reskin B F 1993 Sex segregation in the workplace. Annual Review of Sociology 19: 241–70
16. Reskin B F, Roos P A 1990 Job Queues, Gender Queues: Explaining Women’s Inroads Into Customarily Male Occupations. Temple University Press, Philadelphia, PA
17. Tomaskovic-Devey D 1993 Gender and Racial Inequality at Work. Cornell University, Ithaca, NY