Printer Friendly

Occupational segregation by race and sex, 1940-88.

Occupational segregation by race and sex, 194048 Mary C. King

Changes in occupational distributions are an often overlooked indicator of the role of race and gender in the labor market. However, several recent studies suggest that occupational differentiation, rather than education or experience, drives the more commonly studied race and sex differences in earnings. Heidi Hartmann and Donald Treimann estimate that occupational differences alone accounted for between 35 and 40 percent of the current wage gap between men and women.[1] And Nadja Zalokar has shown that approximately half of the historical earnings difference between black and white women has been attributable to differential allocations among occupations and industries.[2]

Occupational distributions provide some insight into the future, because occupations are far more indicative than current wages of the potential for both advancement and unemployment. Thus, a focus on occupational distributions, rather than wages, allows a richer understanding of the mechanisms behind falling labor market barriers between blacks and whites, and between women and men.

The purpose of this report is to create a comprehensive, detailed picture of the evolution of occupational differentiation along race and gender lines since 1940, using a dissimilarity measure of occupational differentiation. The study focuses particularly on the largest change, the reduction of occupational differentiation between black and white women, which has become the subject of an ongoing controversy over the extent and source of black women's economic progress.3

This report expands upon an earlier study by Randy P. Albelda,[4] using a richer data source that lengthens the period under investigation from 24 to 48 years, creates a more accurate measure of occupational differentiation by working with a detailed occupational breakdown rather than the 29 occupations available to Albelda, and permits a more complex examination of differentiation within groups distinguished by age, region, economic sector, and educational level.

Because of its importance in interpreting the findings, the occupational history of black women is first briefly reviewed. Then, several methodological issues that stem from the use of the dissimilarity index are discussed. The results section presents measures of the level of occupational differentiation between black and white women, black and white men, white men and women, black men and women, black women and white men, and white women and black men from 1940 to 1988, broken down by selected demographic and economic characteristics.

This study relies upon the huge, nationally representative data base found in the 1-in-100 Public Use Micro-data Samples of the U.S. Census for 1940 and 1950, the 1-in-1,000 samples for 1960-80, and the somewhat smaller 1988 Current Population Survey Annual Income Extract. Work history of black women The most significant change in the demographic occupational structure since 1940 is the dramatic decrease in the level of differentiation between white women and black women. The dissimilarity index of occupational differentiation fell by half for black and white women between 1940 and 1988, as illustrated in chart 1, while slipping only about a fifth between black and white men, and a fifth between men and women of both races.

In 1940, domestic service and agricultural work accounted for three fourths of employment among black women, compared with only 13 percent of white working women. The concentration of nearly 60 percent of black working women in domestic work alone in 1940 had huge consequences for the earnings differential between black and white women. In 1945, median annual earnings of domestics were just 16 percent of those of all working women. In 1985, domestics were again earning 16 percent of the annual earnings of all women, but only 3 percent of black women held these jobs.5

Over the years since 1940, the occupational distribution of black women changed remarkably. Although black women still are overrepresented in service work and other low-paying jobs, their distribution among occupations now much more closely resembles that of white women, as shown in table 1. The most significant shifts have been the movement of black women out of domestic work and into clerical work, which now accounts for the largest numbers of both black and white working women.

What cannot be discerned from inspecting the allocation of black and white women among broad occupational categories is whether a new, but less obvious, pattern of occupational segregation has emerged. A more sophisticated measurement of occupational differentiation is needed to determine if the level of occupational segregation experienced by black women has indeed receded, or if the boundaries of black women's employment have only shifted, so that they are now concentrated in particular jobs within the larger occupational categories. An analysis of detailed occupations employing the dissimilarity index can address this issue.


Students of occupational segregation rely almost exclusively on a dissimilarity index, commonly referred to as the Duncan measure, following a 1955 article by Otis Dudley Duncan and Beverly Duncan reviewing the merits of this and other measures.[1] By way of example, the index for black and white women is:
 D = - [Sigma][(B.sub.i / B) - [W.sub.i]/ W)]

where [B.sub.i] is the number of black working women found in a particular occupation i, B is the total number of employed black women, and Wi and W represent the analogous numbers of white working women. In other words, if all black and white women were employed in two occupations, lawyers and file clerks, and 10 percent of black women were lawyers and 90 percent were file clerks, while half of white women were lawyers and half were file clerks, the dissimilarity measure would be:

I fl(. 10-. 50)1+1(.90-.50)11 =. 40


Values of the measure range from 0.0 to 1.0-0.0 representing perfect integration and 1.0 indicating complete segregation.

The biggest advantage of the dissimilarity measure, other than its widespread use, is its intuitive meaning. The index indicates that fraction of either population being compared that would have to change occupations for the two groups to be distributed identically among occupations. In the above hypothetical example, 40 percent of either black or white women would need to shift occupations in order to equalize the proportions of black and white women employed as lawyers and file clerks.

A major limitation of the dissimilarity index, and all measures of segregation, are the problems that arise when the measure is used to compare levels of segregation over time.' Although the changing relative size of groups being compared does not affect the measure, changing numbers of occupations generally does. Of course, the number of occupations in coding systems such as that employed in the decennial census has expanded in every decade. In 1940, the census used 226 job titles; the 1980 census included 503.

If the increase in number of job titles represents merely a growth in classifying detail-that is, finer definitions of existing occupations-then the index may generate higher measures of segregation without segregation having increased, unless each of the smaller occupations reflects the same race or gender balance as the original, more inclusive title. For example, if accountants were disproportionately male and bookkeepers disproportionately female, a dissimilarity measure that used a job title "accountants and bookkeepers" would uncover less occupational segregation than if two separate job titles were used.

While dissimilarity indexes based on the larger numbers of occupational descriptions in later census years may more accurately indicate the extent of differences in occupational employment, these measures will not be directly comparable to those calculated using the less refined census titles of the 1940's and 1950's. Dissimilarity indexes based on the entire census coding scheme underestimate the level of segregation in earlier years, and consequently may understate the decrease in segregation over time.

Researchers investigating occupational segregation have attempted to deal with the issue of increasing numbers of job titles in four ways. The first, used by Albelda, is to aggregate job titles into broad job categories. While these categories are roughly comparable over time, their breadth necessarily obscures a significant amount of differentiation.

The second method, employed by Edward Gross, is to make use of every available category for any year, treating all new job titles as legitimately distinct jobs.1 Growing numbers of job titles are implicitly assumed to be the logical outcome of an increasingly specialized labor force. To the extent that this assumption is mistaken, the results understate the extent of segregation in the beginning of the study period and the decrease in occupational segregation over time.

The third possible tactic is that employed by Francine D. Blau and Wallace E. Hendricks-the construction of a subset of occupational titles common to every sample.[9] The primary concern here is with the issue of representativeness. Because the occupations measured are only a subset of all occupations, the results may not accurately reflect the occupational differences found in the economy as a whole. Although the direction of any potential bias is not obvious, it seems reasonable to assume that integration should be easier in new jobs that have not yet been associated with any particular demographic group than in existing jobs that are already perceived as "white women's jobs" or "men's jobs." If so, indexes based only on occupations found in every period would again understate declines in segregation in the economy as a whole by excluding the very occupations likely to be most integrated.

A fourth approach has been taken by Gregory Williams, in an attempt to improve on Gross' measures.[10] Williams' idea is to divide occupations into four categories: (1) those found in each census year, (2) those new jobs created by technological change in any given decade, (3) new job titles that differentiated existing jobs that formerly had been classified together, and (4) jobs that earlier had been separated, but which were lumped together in later census coding systems. Williams based his analysis on the subset of jobs in categories 1, 3, and 4, aggregating job titles in groups 3 and 4 to the breadth of the most inclusive classification ever used.

This approach is appealing, because it seems to create the largest possible directly comparable subset of occupations. However, on closer examination, this method not only appears to be susceptible to the problems of representativeness resulting from reliance on a subset of titles and biases toward understating segregation due to aggregation of distinct jobs, but also involves a great deal of discretion and judgment in creating categories 2 through 4.

Finally, men's occupations tend to be more finely defined than are women's. Office work in particular encompasses relatively few, broad job titles, such as secretary or office assistant, which include positions that may be quite different as to skill requirements and responsibilities. Men's jobs, especially in the crafts, have been more sharply distinguished, perhaps as a result of union activity. Consequently, dissimilarity measures would reveal less racial segregation among women's jobs than among men's, because the paucity of categories obscures real differentiation. For instance, in 1940, the census occupational coding system included the title "secretaries, stenographers, and typists." The secretaries could have been all white and the typists all black, but the overall category would appear to be integrated.

Further, if there has been a relatively greater increase in the detail of women's job descriptions since 1940 than has been true for men's jobs, race segregation may appear to be decreasing more slowly among women than among men, but comparisons of gender segregation would not be distorted. Resolution of method questions To provide measures of occupational segregation that can be easily compared with those in the existing literature, this report continues the tradition of reliance upon a dissimilarity measure.

The problem of potential bias arising from changing numbers of job titles is not easily avoided by using other measures." As a check on the ability of the dissimilarity index to accurately reflect changes over time, the measure is calculated both on the entire sample, including all current job titles for each year analyzed, and on a subset of 159 occupational titles that are common to the census classification codes of every period, following the example of Blau and Hendricks, This subset, hereinafter referred to as the "core sample," includes all detailed job titles found in the coding scheme of each census; it excludes all others, including the miscellaneous categories, such as "clerical, not elsewhere classified."" The core sample includes about 60 percent of the total sample, varying from 50 percent to 83 percent for each group in each year."

The fourth method of constructing the dissimilarity index-that advocated by Williams-appears to muddy the waters with excessively discretionary judgments without providing much additional information, and has not been used.

Redefinition of occupations at the level of the individual occupation has been ignored. Redefinition and reclassification appear to be more problematic for research based on aggregate classifications. [14]


Table 2 presents dissimilarity measures calculated for each of the census years from 1940 through 1980, and for 1988. Figures for 1988, although they appear quite reasonable, are less reliable than those for earlier years, based as they are on the smaller sample of the Current Population Survey. The table presents results both for the entire sample and for the core sample for six different comparisons: black and white women, black and white men, black men and women, white men and women, black women and white men, and white women and black men. The indexes in the bottom half of the table are constructed using all job titles current in the given census year. The first set of results for each race-sex category is based on the core sample of 159 occupations found in each census year.

Several clear patterns emerge from these indexes. First, the two sets of indexes provide very similar descriptions of the pattern of occupational segregation throughout the period, indicating that the results for each index are robust, despite the concerns described in the methodology section above. Therefore, because indexes based on the core sample are best for comparisons over time, these are presented for the years 1940 through 1980 in the more detailed tables 3 and 4. Developments during the 1980's may be captured by the indexes derived from the total sample because the 1980 and 1988 occupational codes are the same.

Second, the absolute levels of occupational segregation revealed are remarkably high. Approximately two-thirds of men or women would have to change jobs to achieve complete gender integration. Nearly that proportion of black or white women would have had to shift to achieve a balanced occupational outcome among women in 1940, and 30 percent would have to move today.

Third, the greatest reductions in segregation are shown to have occurred between black and white women. In 1940, 0.618 of black and white women were overrepresented in their occupations, as estimated from the core sample. Between 1940 and 1950, this figure rose somewhat, by 3 percentage points, and then remained stagnant through the 1950's.

The index dropped precipitously between 1960 and 1970, by nearly 17 percentage points. The index fell another 15 points during the 1970's, reaching 0.326 in 1980. Although the core sample suggests that differentiation continued to drop between black and white women through the 1980's, the more accurate reading is found using the entire sample, which shows that movement ceased during that decade.

By contrast, the core sample segregation measure between black and white men stood at 0.383 in 1940, rose through the 1940's and 1950's by 2 to 3 points a decade, and then fell nearly 8 points during the 1960's and another 5 points through the 1970's, to 0.313 in 1980. Estimates for both the core sample and the entire sample indicate a slight decrease in occupational differentiation by race among men in the 1980's.

Fourth, the indexes of gender segregation within racial groups for this period are comparable to measures of gender segregation estimated from detailed occupations by other authors. Andrea H. Beller's 1984 paper, one of the most rigorous of the analyses of gender segregation, presents dissimilarity measures of between 0.687 in 1960 and 0.617 in 1981.15 These results are based on an occupational coding system that includes 262 titles, which is larger than that for the core sample for this study, but smaller than that for the total sample for the years from 1960 on.

Fifth, gender segregation among blacks and whites has remained quite comparable over time. The biggest decreases in occupational segregation by gender occurred during the 1970's. Whatever caused the large declines in occupational segregation by race during the 1960's had relatively little impact on the occupational differentiation between men and women.

Sixth, dissimilarity indexes are highest between black men and white women and between white men and black women, highlighting the importance of both race and gender in the allocation of workers among jobs.

Seventh, comparison with Albelda's results shows that employing a more detailed occupational coding system increases measures of differentiation on the order of 10 percentage points. Gender segregation appears to be more similar for blacks and whites than is indicated by Albelda's work, and occupational segregation by race has not decreased as much among either men or women as suggested by her analysis of very broad job categories. [16]

In short, it is clear that different dynamics are driving the changes in occupational differentiation by race experienced by men and women, although it is not obvious why women should initially have been so much more segregated by race, or why racial barriers have fallen so much more precipitously between women than between men. Greater detail may provide some insights into these questions. Tables 3 and 4 present additional segregation indexes for race-gender groups, broken down by age, region, economic sector, and educational level. Job segregation among women By far the biggest reduction in occupational segregation by race among women occurs in the 1960's, when the measure falls by approximately 25 percent. Clearly, the events of the 1960's-whether economic growth, the civil rights movement, or something else entirely-shook the longstanding structure of segregation of black women in the labor market. Segregation continued to diminish rapidly through the 1970's. Very little change was apparent in the 1980's.

The lowest levels of occupational segregation by race among women in the early years are found among women with college educations and among those employed by the public sector. The private sector remained more segregated in 1988 than the public sector had been in 1940. The effects of education and public sector employment may combine to discourage occupational differential by race. Black women with college degrees have been concentrated in teaching and nursing, occupations disproportionately located in the public sector." Segregation in schools and hospitals may have broadened the availability of teaching and nursing positions for black women.

It may also be that the government has pursued a more race-blind hiring policy than have private employers. The public sector has been a much more important source of clerical employment for black women than for white women. In 1960, 46 percent of black women working in clerical positions were employed in the public sector, 20 percent of all black women clerical workers by the Federal Government. By contrast, only 21 percent of white women clerical workers held public sector jobs at this time, and only 6 percent of white women clerical workers were employed by the Federal Government. This pattern was still in evidence in 1980, when 39 percent of black women clericals and 20 percent of white women in clerical positions were working in the public sector."

The equalizing effect of education among women may also be partially driving the dramatic reversal in the level of racial differentiation by age. Younger women were more segregated than women over age 35 from 1940 until 1960, when the pattern abruptly shifted so that younger women were much less occupationally differentiated in 1970 and 1980 than were older women. The effect of greater levels of education would have been more apparent in earlier years among women old enough to have complete college degrees; by the 1970's and 1980's, education levels were converging more rapidly for younger women at the same time that opportunities were widening for black women.

In the private sector, education may function as a signal of merit to employers who otherwise regard employing black women as more risky than hiring white women. A 1949 study found that, due to the proximity of clerical workers to management, the most important qualifications that employers sought in clerical workers were education, character," and particular personality traits, rather than either general or specific experience.[19]

E. William Noland and E. Wight Bakke, who conducted the study, suggested that "employers believe Negroes to be inferior to whites in these respects [education, character, and personality traits]." This attitude is demonstrated in an interview with the president of Pitney-Bowes, Inc., about the first black woman hired to work in the office, "The girl, a very fine individual and well educated, soon proved herself...... 211 In short, education may have been one of the few ways in which a black woman could demonstrate her employability.

Further, it apparently is easier for black women, and white women as well, to gain entry to occupations that require a formal credential than to those positions that require a more subjective assessment of candidates' training and abilities, such as managerial and upper-level clerical positions. Evidence for this conclusion may be found in the greater success of both black and white women in penetrating the professions than in gaining a foothold in management, although the preponderance of grade school teachers among women professionals somewhat skews this comparison.[21]

Segregation among male workers

Not only was occupational differentiation by race far less between men than between women in this study, but most of the trends visible in the detailed indexes of segregation between black and white women do not appear in those for black and white men. Further, similar breakdowns computed for black men and women, white men and women, black women and white men, and white women and black men, which are not presented here, show that the patterns observed between black and white women are unique.

The large differences in the level of occupational differentiation by race in the public and private sectors experienced by women do not appear for men. Further, the least educated men are the most integrated.

Once again, the hypothesis is suggested that white men, especially educated white men, enjoy greater access than do educated members of other groups to jobs that entail a degree of subjective evaluation for hiring and promotion. If white women are limited to a small number of formally credentialed professions, it should be easier for black women to achieve parity with white women than for black men to do so with white men.

Men were much less segregated by race than were women during the years from 1940 to 1960, despite the far smaller number of job categories commonly dominated by women, which should tend to bias downward estimates of race segregation among women. Since 1980, men have been more occupationally differentiated than are women, although the level is no longer substantially different between the genders.

Women seem to have been more segregated in the South throughout the period, a pattern that holds true for men only since 1960. It may be that black women held only "black women's jobs" in the South, while black and white men were employed in similar jobs, but in segregated production units. This pattern is suggested by a 1955 survey of the employment practices of three firms in New Orleans. Blacks constituted nearly half of the labor force of each company, but held no clerical or office jobs, which were presumably white women's jobs." One firm was completely segregated, although blacks and whites held many of the same jobs, because it maintained separate, but otherwise identical, production areas."


The two dissimilarity indexes generate very similar descriptions of the changing pattern of occupational segregation over the period, indicating that the results for each index are robust despite the concerns raised in the methodology section of this report.

Further, the dissimilarity indexes reveal a remarkable level of occupational segregation by race and sex. On the evidence of the segregation indexes alone, blacks and whites and men and women appear to have participated in labor markets which overlapped only partially in 1940. By 1980, these separate markets included more common territory, with the greatest merging occurring between the markets for black and white women, to form a larger women's labor market.

The greatest changes in the level of occupational segregation since 1940 occurred among women along race lines. Differentiation by race among men is falling more slowly than among women, and is now approximately equal to that among women. Occupational segregation by gender is falling more slowly still, and is approximately equal for blacks and whites.

The most significant changes in occupational segregation by race are shown to have occurred in the 1960's and 1970's. The tumultuous events of the 1960's provide several candidates as potential sources of the large shift. A common conclusion among students of black economic progress is that economic expansion is crucial to black advancement, and certainly the lowest unemployment rates of the study period occurred during the late 1960's." Michael Reich argues that the strength of sociopolitical movements concerned with black progress is a more fundamental source of improvements in black economic status.

If civil rights legislation was the key to black progress in the 1960's and 1970's, perhaps the legal system is also behind the advance made by women in the 1970's.

Further work, which focuses more tightly on the years since 1960, is needed to disentangle the possible effects of improving education for black women, tight labor markets, the civil rights movement and consequent legislation, the women's movement, and widening opportunities for white women. What this report has demonstrated is the complexity of the roles of race and gender in the labor market. Clearly, events since 1940 have affected relations between black and white women much differently than those between any other groups. Policy geared to improving labor market outcomes for women and ethnic minorities needs to be informed by a sophisticated understanding of the differences in race and gender dynamics in the labor market.


ACKNOWLEDGMENT: The author thanks Lee Badgett, Libby Bishop, Bill Dickens, Clifford Lehman, David Levine, Kathy O'Regan, Michael Reich, Lisa Saunders, and her editor for their very helpful comments.

1. See Heidi Hartmann and Donald Treimann, eds., Women, Work and Wages: Equal Pay for Jobs of Equal Value (Washington, National Academy Press, 1981).

2. James S. Cunningham and Nadja Zalokar, The Economic Progress of Black Women Since 1940: Wages and Occupations," Unpublished Paper (Washington, 1990).

3. See Richard B. Freeman, "Changes in the Labor Market for Black Americans, 19481972," Brookings Papers on Economic Activity, 1973, pp. 67-131; Timothy Bates, "Black Economic Well-Being Since the 1950's," The Review of Black Political Economy, Spring 1984, pp. 5-39,, Julianne Malveaux, "The Economic Interests of Black and White Women: Are They Similar?" The Review of Political Economy, Summer 1985, pp. 5-27; Margaret C. Simms and Julianne M. Malveaux, Slipping Through the Cracks: The Status of Black Women (New Brunswick, NJ, Transaction Books, 1987); Cunningham and Zalokar, "The Economic Progress"; and Nadja Zalokar, The Economic Status of Black Women: An Exploratory Investigation, U.S. Commission on Civil Rights Clearinghouse Report (Washington, U.S. Government Printing Office, 1990).

4. Randy P. Albelda, "Occupational Segregation by Race and Gender, 1958-8l," Industrial and Labor Relations Review, April 1986, pp. 404-11.

5. Current Population Reports: Consumer Income (Bureau of the Census), for the years 1948-86.

6 Otis Dudley Duncan and Beverly Duncan, "A Methodological Analysis of Segregation Indexes," American Sociological Review, April 1955, pp. 210-17. See also Andrea H. Beller, "Trends in Occupational Segregation by Sex and Race, 1960-1981," in Barbara Reskin, ed., Sex Segregation in the Workplace: Trends, Explanations, Remedies (Washington, The National Academy Press, 1984).

7. See Paula England, "Assessing Trends in Occupational Sex Segregation, 1900-1976," in Ivar Berg, ed., Sociological Perspectives on Labor Markets (New York, Academic Press, 1981), pp. 273-95; and Michael J. White, "Segregation and Diversity Measures in Population Distribution," Population Index, Summer 1986, pp.198-221.

8. Edward Gross, "Plus ;a Change ... ? The Sexual Structure of Occupations Over Time," Social Problems, Fall 1968, pp. 198-208.

9. Francine D. Blau and Wallace E. Hendricks, "Occupational Segregation by Sex: Trends and Prospects," The Journal of human Resources, Spring 1979, pp. 197-210.

10. Gregory Williams, "The Changing U.S. Labor Force and Occupational Differentiation by Sex," Demography, February 1979, pp. 7387; and Gregory Williams, "Trends in Occupational Differentiation by Sex," Sociology of Work and Occupations, February 1976, pp. 38-62.

11. See White, Segregation and Diversity Measures."

12. The 159-occupation core sample includes everyone classified under job titles from the 1940 census coding system for which a close approximation is found in the coding system of each succeeding census. Many of these job titles are quite straightforward. For example, the clerical occupations included in the core sample are: bookkeepers, accountants, and cashiers; store clerks; mail carriers; messengers and office boys; shipping and receiving clerks; stenographers, secretaries, and typists; telegraph operators; telephone operators; ticket and station agents; library attendants; bill and account collectors; express messengers and railway mail clerks; office machine operators; and telegraph messengers.

13. The core samples include the following fractions of the total samples:
 '40 '50 '60 '70 '80 88
 Total ....... .60 .58 .62 .56 .64 .59
Black women.......83 .79 .75 .58 .68 .58
White women.......61 .67 .62 .60 .68 59
Black men.........64 .52 .62 .50 .63 .57
White men.........58 .55 .60 .54 .61 .60

14. See Nancy Rytina and Suzanne Bianchi, "Occupational reclassification and changes in distribution by gender," Monthly Labor Review, March 1984, pp. 11-17; Technical paper 18: The Relationship Between the 1950 and 1960 Occupation and Industry Classifications (Bureau of the Census, 1968); Technical Paper 26: 1970 Occupation and Industry Classification Systems in Terms of Their 1960 Occupation and Industry Elements (Bureau of the Census, 1972); and Technical Paper 59: The Relationship Between the 1970 and 1980 Industry and Occupation Classification Systems (Bureau of the Census, 1989).

15. Beller, "Trends in Occupational Segregation."

16. Albelda, "Occupational Segregation."

17. Phyllis Wallace, Black Women in the Labor Force (Cambridge, MA, MIT Press, 1980).

18. 1960 Census of Population, Subject Report: Occupational Charactistics (Bureau of the Census, 1963); 1980 Census of Population, Volume 1, ch. D, pt. 1, section A (Bureau of the Census, 1984); and Economic Report of the President, 1990.

19. E. William Noland and E. Wight Bakke, Workers Wanted: A Study of Employers' Hiring Policies, Preferences and Practices in New Haven and Charlotte (New York, Harper & Brothers, 1949).

20. "Conference on Equal Job Opportunity," Monthly Labor Review, January 1956, pp. 3 1 33.

21. As the following tabulation demonstrates, the proportion of white men in managerial positions has always exceeded their fraction in the professions. The proportion of black men in management has just caught up with that fraction in professional jobs, whereas women of both races are far more likely to work as professionals than as managers.

Percent of the occupational category in - '40 '50 '60 '70 '80 88
Black women:
 Professional. . 4.2 4.8 7.7 9.4 10.4 9.6
 Managerial.. . .8 1.2 .9 1.1 4.1 6.4
White women:
 Professional. . 13.212.5 11.5 13.5 13.5 14.4
 Managerial ... 4.2 4.5 3.3 3.4 7.0 11.1
Black men:
 Professional. . 2.0 1.8 2.8 4.5 5.0 6.1
 Managerial ... 1.2 2.2 1.4 2.3 4.6 5.9
White men:
 Professional. . 5.4 7.1 9.0 11.6 10.4 11.9
 Managerial ... 9.6 10.8 10.9 10.8 12.4 14.0

21. Source: Public Use Micro-data Samples of the 1940-80 censuses and the 1988 Current Population Survey Annual Income Extract.

22. "Negro Employment in Three Companies in the New Orleans Area," Monthly Labor Review, September 1955, pp. 1020-23.

23. Gerald D. Jaynes and Robin M. Williams, Jr., eds., A Common Destiny: Blacks and American Society (Washington, National Academy Press, 1980).

24. Michael Reich, "Postwar Racial Income Differences: Trends and Theories," in Garth Mangum and Peter Philips, eds., Three Worlds of Labor Economics (Armonk, NY, M.E. Sharpe, Inc., 1988), pp. 144-67.
COPYRIGHT 1992 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:King, Mary C.
Publication:Monthly Labor Review
Date:Apr 1, 1992
Previous Article:Job training in Europe: lessons from abroad.
Next Article:Injuries and illnesses in the workplace, 1990.

Related Articles
Sex segregation.
Employment characteristics of older women, 1987.
Gender and economic outcomes.
Gender differences in occupational employment.
Wage gaps and segregation.
Gender and Jobs: Sex Segregation of Occupations in the World.
Occupational segregation: the position of women in the North Cyprus labor market.
Women and the paradox of economic inequality in the twentieth-century.
Industry representation, structural change and the older male worker in Australia (1985 to 2005).

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters