Gender differences in occupational employment

Monthly Labor Review

Volumn 120, Issue 4, 1997, Pages 15-24

  Wootton, Barbara H ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0037925804     PISSN: 00981818     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (34)

1. See, for example, Francine D. Blau and Marianne A. Ferber, The Economics of Women, Men, and Work (Englewood Cliffs, NJ, Prentice-Hall, 1986); and Barbara F. Reskin and Heidi I. Hartmann, Women's Work, Men's Work: Sex Segregation on the Job (Washington, National Academy Press, 1986).
2. See, for example, Francine D. Blau and Marianne A. Ferber, The Economics of Women, Men, and Work (Englewood Cliffs, NJ, Prentice- Hall, 1986); and Barbara F. Reskin and Heidi I. Hartmann, Women's Work, Men's Work: Sex Segregation on the Job (Washington, National Academy Press, 1986).
3. See David A. Cotter, Joann M. DeFiore, Joan M. Hermsen, Brenda M. Kowalewski, and Reeve Vanneman, "Occupational Gender Desegregation in the 1980s," Work and Occupations, February 1995, pp. 3-21.
4. Barbara F. Reskin and Patricia A. Roos, eds., Job Queues and Gender Queues (Philadelphia, Temple University Press, 1990), p. 16; and Jerry A. Jacobs, "Long-term trends in occupational segregation by sex," American Journal of Sociology, July 1989, pp. 160-73.
5. Barbara F. Reskin and Patricia A. Roos, eds., Job Queues and Gender Queues (Philadelphia, Temple University Press, 1990), p. 16; and Jerry A. Jacobs, "Long-term trends in occupational segregation by sex," American Journal of Sociology, July 1989, pp. 160-73.
6. See Cotter and others, "Occupational Gender Desegregation in the 1980s"; and Mary C. King, "Occupational segregation by race and sex, 1940-1988," Monthly Labor Review, April 1992, pp. 30-37.
7. See Cotter and others, "Occupational Gender Desegregation in the 1980s"; and Mary C. King, "Occupational segregation by race and sex, 1940-1988," Monthly Labor Review, April 1992, pp. 30-37.
8. The data used in this article are annual averages derived from the Current Population Survey (CPS), a monthly sample survey of about 50,000 households, conducted by the Bureau of the Census for the Bureau of Labor Statistics. The CPS has the benefit of providing the most current data available, as well as consistency in source over time. Some prior research has compared census data for one point in time to CPS data for another point in time (see, for example, King, "Occupational segregation by race and sex, 1940-1988"), which can be problematic because these sources can show different occupational distributions for the same year. See Suzanne M. Bianchi and Nancy F. Rytina, "The decline in occupational sex segregation during the 1970s: Census and CPS comparisons," Demography, 1986, vol. 23, pp. 79-86, note 2. Also, since 1983, relatively few changes have been made to the occupational classification system used in the CPS, providing comparability for nearly all of the detailed occupational categories over the period examined.
9. The data used in this article are annual averages derived from the Current Population Survey (CPS), a monthly sample survey of about 50,000 households, conducted by the Bureau of the Census for the Bureau of Labor Statistics. The CPS has the benefit of providing the most current data available, as well as consistency in source over time. Some prior research has compared census data for one point in time to CPS data for another point in time (see, for example, King, "Occupational segregation by race and sex, 1940-1988"), which can be problematic because these sources can show different occupational distributions for the same year. See Suzanne M. Bianchi and Nancy F. Rytina, "The decline in occupational sex segregation during the 1970s: Census and CPS comparisons," Demography, 1986, vol. 23, pp. 79-86, note 2. Also, since 1983, relatively few changes have been made to the occupational classification system used in the CPS, providing comparability for nearly all of the detailed occupational categories over the period examined.
10. Technological and organizational changes create new kinds of jobs and affect the structure of existing jobs. These changes are reflected in revisions to the occupational structures used to collect data, making detailed comparisons over long periods difficult. Thus, comparable CPS data for detailed occupational categories are not available prior to 1983; comparisons of broader occupational groups can be made back to 1972.
11. See, for example, Judith Fields and Edward N. Wolff, "The Decline of Sex Segregation and the Wage Gap, 1970-80," The Journal of Human Resources, Fall 1991, pp. 608-22 (finding that high employment growth within an occupation is a major factor associated with declines in gender segregation, particularly for male-dominated occupations).
12. i is the percentage of men employed in the same occupation i. See O.D. Duncan and B. Duncan, "A methodological analysis of segregation indexes," American Sociological Review, 1955, vol. 20, pp. 210-17.
13. The appeal of the index is both that it is widely used and that its meaning is fairly intuitive. A limitation of the index, and of other measures of occupational difference, is apparent when trying to compare levels of segregation over time. Changes to the occupational classification structure over time may result in an index that overstates or understates the degree of change. In this article, we focus the analysis on the 1985 to 1995 period, because there were only a handful of changes to the CPS classification structure during that time. Numerous changes to the occupational structure implemented in the CPS in 1983 make it problematic to extend the analysis back prior to that time. A different problem that exists for the analysis, regardless of the period, is that a number of occupational categories - "not elsewhere classified" (n.e.c.) and "miscellaneous" categories - consist of an unknown, heterogeneous mix of jobs. One option would be to exclude these groupings from the analysis because they are not "true" occupational categories; however, this would result in a reduction of the number of occupational categories examined from 501 to 438 and a loss of about 15 percent of total employment from the analysis. For this reason, and because the gender makeup of these categories typically is similar to that for specific related occupations, these categories were included in the index calculation.
14. The indexes for each year from 1985 through 1995 show a slow downward trend, with little change between 1991 and 1992. This puts to rest any apprehension that the introduction of 1990 changes to the census occupational structure into the 1992 CPS would alter trends in the data.
15. Bianchi and Rytina, "The decline in occupational sex segregation," table 2, p. 81. This figure is from census data. Using CPS data for 1972 and 1982, Bianchi and Rytina find a decline in the difference index of 7.4 percent.
16. Cotter and others, "Occupational Gender Desegregation," p. 13. This figure (6.5 points) is based on the full 1980 census and the 1990 census 5- percent Public Use Microdata Samples.
17. Difference indexes by educational attainment were calculated using less aggregated occupational categories than those used in the other calculations in this article because they were the most detailed data available. There are 45 of the less aggregated or "intermediate" occupational categories. Also, the occupational data by educational attainment are based on persons aged 25 and older - those most likely to have completed their education.
18. The difference index is sensitive to the level of occupational aggregation or number of occupations examined. In general, finer levels of definitions generate a higher index, unless the gender distribution within each of these finer categories is identical to the broader occupational grouping. Because the index by educational attainment was calculated using 45 intermediate-level occupations, the total for all educational levels in 1995 was somewhat lower (46) than the total index using the 501 most-detailed level occupations (53.5).
19. Difference indexes were calculated using detailed-level occupational categories.
20. These findings differ from the findings of Holden and Hansen with regard to the 1970s. They concluded that occupational differences level of education. Women aged 16 to 19 tend to hold a disproportionate employment share of salesworkers, apparel; cashiers; secretaries; receptionists; clerks; waitresses; food counter, fountain, and related occupations; child care workers; and early childhood teacher's assistants. Men aged 16 to 19, by contrast, tend to be disproportionately employed as cooks; janitors and cleaners; truckdrivers; farmworkers; groundskeepers; and various laborers and handlers.
21. See Francine D. Blau and Wallace F. Hendricks, "Occupational segregation by sex: trends and prospects," The Journal of Human Resources, 1979, vol. 14, no. 2, pp. 197-210, at pp. 199-200. To decompose the change over time, the 1995 difference index is standardized on the 1985 occupational structure. The occupational mix effect represents the change in the difference index that would have occurred had the gender composition of each occupation remained constant between 1985 and 1995. The gender composition effect represents the change in the difference index that would have resulted had the occupational employment structure remained constant over time.
22. See Bianchi and Rytina, "The decline in occupational sex segregation," table 2, p. 81. This study found the composition effect responsible for 75 percent of the decline in occupational segregation during the 1970s. Also see Cotter and others, "Occupational Gender Desegregation," table 3, p. 14, which attributed 71 percent of the decline during the 1980s to the composition effect.
23. See Bianchi and Rytina, "The decline in occupational sex segregation," table 2, p. 81. This study found the composition effect responsible for 75 percent of the decline in occupational segregation during the 1970s. Also see Cotter and others, "Occupational Gender Desegregation," table 3, p. 14, which attributed 71 percent of the decline during the 1980s to the composition effect.
24. Bianchi and Rytina, "The decline in occupational sex segregation"; and Cotter and others, "Occupational Gender Desegration."
25. Bianchi and Rytina, "The decline in occupational sex segregation"; and Cotter and others, "Occupational Gender Desegration."
26. See Reskin and Roos, Job Queues and Gender Queues, p. 16; and Blanchi and Rytina, "The decline in occupational sex segregation," p. 82.
27. See Reskin and Roos, Job Queues and Gender Queues, p. 16; and Blanchi and Rytina, "The decline in occupational sex segregation," p. 82.
28. Suzanne M. Bianchi and Nancy F. Rytina, "Occupational change, 1970-1980," Paper prepared for the Population Association of America, Minneapolis, May 1984 (cited in Reskin and Roos, Job Queues and Gender Queues, p. 6).
29. Suzanne M. Bianchi and Nancy F. Rytina, "Occupational change, 1970- 1980," Paper prepared for the Population Association of America, Minneapolis, May 1984 (cited in Reskin and Roos, Job Queues and Gender Queues, p. 6).
30. Rachel A. Rosenfeld and Kenneth I. Spenner, "Occupational sex segregation and women's early career job shifts," Work and Occupations, November 1992, pp. 424-49, at p. 429.
31. However, if one were to use a 70-percent concentration as the cutoff for "female" occupations, the proportion of women employed in such occupations would edge down only slightly from 57 percent in 1985 to 55 percent in 1995. Similarly, if a 60-percent cutoff were used, the share of women's employment in female-dominated occupations would edge down from 63 percent in 1985 to 61 percent in 1995. These differing results are not surprising because, as mentioned in the text, the share of women's employment increased most over the decade in occupations composed of 60 percent to 80 percent women.
32. Occupational titles that encompassed heterogeneous jobs (n.e.c., miscellaneous, and "all other" categories) were excluded from this analysis because such categories are not analytically meaningful. Occupations with a small sample base (estimates of less than 50,000) also were excluded because of their large associated sampling error.
33. See Reskin and Roos, Job Queues and Gender Queues, pp. 16-20.
34. Reskin and Roos, Job Queues and Gender Queues, p. 21.