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.
~ Job Changing and Occupadonal 7 Sex Segregation: Sex and Race Compansons RACHEL A. ROSENE7ELD The U.S. occupational structure is and has been extremely sex segregated (see Beller, in this volume). The extent and sta- bility of such sex segregation prevents most individuals from considering possible mo- bility between sex-typical and atypical oc- cupations. Such mobility does, however, oc- cur. World War II was a dramatic example of a situation where many women changed from typically female to typically male jobs. Even under less extreme circumstances, the sex composition of a person's occupation is not a constant throughout one's work life, as will be shown here and as others have demonstrated (Wolf and Rosenfeld, 1978; England, 1982b; Corcoran et al., in this vol- ume; Jusenius, 1975; Sociology of Work and Occupations, vol. 9, number 3, 19821. Using 1973 data on job changers, this paper will focus on change in occupational sex com- position that people experience with a change in employer. It will describe the movement by black and white women and mend among occupations with clifferent sex compositions and will predict movement to or from sex- atypical occupations using a range of indi- vidual and job history variables. EXPLANATIONS OF SEX SEGREGATION Explanations for why women end up in typically and predominantly female occu- pations vary depending on whether the ex- planations focus on labor supply or labor demand. Both types of explanations usually fail to consider that a person might break through the sex segregation barrier. Labor Supply Explanations Labor supply arguments about sex segre- gation often give the impression that a person makes a once-in-a-lifetime and usually sex- typical occupational choice. Socialization ex- planations, one type of supply side argument, suggest that women are socialized to plan for ~ There are, of course, other ethnic groups whose occupational sex segregation would be interesting and useful to study. One might like, for example, to follow Malveaux's (1982) lead and look at the position of His- panics and their mobility. The data set used here, how- 56 ever, identifies race/ethnicity only as white, black, and other. The other group is extremely small and probably quite heterogeneous. In what follows, those identified as other are dropped from the analysis, leaving com- parisons between blacks and whites.
JOB CHANGING AND OCCUPATIONAL SEX SEGREGATION _ 57 and enter occupations that society considers appropriate for women, while men are so- cialized to choose Tom a wider range of suit- ably male occupations. Consistent with this explanation is the evidence of students' early expectations of and aspirations for sex-typical occupations (see Marini and Brinton, in this volume, for a review of this literature). Hu- man capital explanations of sex segregation, another supply side argument, see the choice and preparation for a sex-typical occupation as part of economically rational planning. Since women expect to take time out from the labor force for fur-time work at home, they train for ant! later enter occupations that over easy reentry ant! low depreciation of their skills and training while they are out of the labor force. Such occupations, of course, become predom- inantly female (Polachek, 1979, 1981a; Op- penheimer, 19701. (See the detailed cliscus- sion of this approach and evidence contrary to its assumptions and hypotheses in Corcoran et al., in this volume.) Supply side arguments carry with them as- sumptions about people's education and train- ing. Much formal training for work careers takes place early in a person's life, often before one starts full-time work because in part, as the human capitalist explains, early training provides a longer time over which to receive the returns to this training. Women's occu- pations are not necessarily low-skilled, but they are described as occupations in which a woman must bring her training with her to the job because her expected short tenure does not allow enough time to receive returns on her training there (Oppenheimer, 19701. An early choice of a typically female occu- pation, therefore, must be to some extent a choice about the type of training to get or the type of major to take in college (Polacheck, 19781. Tra~n~ng for a typically female job might preclude training for a typically male job; for example, when a woman decides or is advised to attend nursing school rather than medical school. Thus prepared, the woman lacks the credentials to enter a medical occupation atypical for her sex. Likewise, a young man who does not take clerical courses in high school or vocational school may not be able to get a secretarial job. An early choice of training thus determines a person's later oc- cupational career. Labor Demand Explanations Labor demand arguments say that the ex- clusion of women from traditionally male jobs, especially the ones that supposedly require continuous commitment, is largely a result of employers' and male workers' preferences, not women's choices. The internal labor markets' literature hypothesizes that on some career ladders that are protected from outside mar- kets and for which workers receive on-the- job training, employers are unwilling to take a chance on losing their training investment by hiring members of high-risk groups; that is, members of groups known to be unstable workers (e.g., Doeringer and Piore, 1971; Ed- wards, 19791. Because of their propensity to marry and bear and raise children, women are considered one of these groups. On the basis of their group characteristics, then, all women might be screened from certain male jobs by what is known as statistical discrimi- nation (Phelps, 19721.2 Employers may also 2 It is not necessary for the beliefs about certain groups to be true for statistical discrimination to occur (Spence, 1974; see also England, 1982a). If the supply of appro- priate labor is great enough, and in the absence of other pressures, the employer will not be hurt even if the stereotypes about groups are objectively incorrect. At times, employers have even created the link between women's domestic roles and intermittent labor force participation, reinforcing stereotypes about women as workers. It is not that women have always chosen to leave the labor market when they marry or have chil- dren; it has also been the case that employers required that women who marry or have children leave their jobs. For example, secretaries, teachers, and flight at- tendants the latter as late as 1972 have been re- quired to leave their jobs when they marry (Davies, 1975; Cohn, 1982; see also Cook and Hayashi, 1980, on forces in contemporary Japanese firms pushing or encouraging women to leave the labor force at marriage or childbirth).
58 RACHEL A. ROSENFELD feel that on-thejob training is easier if the work force is homogeneous. Even for low- skiDed jobs for which a long-term commit- ment is not expected, however, employers may hire on the basis of the sex appropriate- ness of the applicant to the sex type of the job (Levinson, 19751. In addition, male workers in competition with female workers have excluded women from predominantly male training programs and jobs (see Baker, 1964; Hartmann, 1976~. Such exclusion may be hidden in seemingly universal entrance requirements and hiring and promotion procedures (Roos and Res- kin, in this volume). Reinforcement of Sex Segregation Once women have selected out of, or are excluded from, male occupations, some ex- planations go on to say, their tenure in typ- ically female occupations furler reduces their ability to change to a typically male occu- pation. In their typically female jobs, they get returns from the skills in which they have already invested; they do not have the chance to learn new skills necessary for male jobs; they may be isolated from information networks about typically male job openings (e.g., Boos and Reskin, in this volume); and therefore they do not have access to the male career ladders that provide better ad- vancement than do female job ladders. As a result of few advancement opportunities and the demands of both a family and an outside job, women may indeed lack com- mitment to their jobs and decide to drop out of the labor force. EXPLANATIONS OF MOBILITY To begin to study mobility to and from a sex-typical occupation, rather than stopping with the conclusion of immobility, one must (1) examine more carefully the stereotypes of typically female and male jobs, (2) con- sider how supply side characteristics might change over a person's work life, and (3) consider changes in demand for certain kinds of labor. Occupations and Sex Stereotyping The contrast often made is between typ- ically female white-collar occupations and typically male professions and crafts. One is led to forget the variation among typically male jobs and among typically female jobs. Some typically male jobs do not require much skill or continuity, although they may pay more than a predominantly female job (Eng- land and McLaughlin, 1979; England et al., 1982~. Early decisions about future occu- pations and occupational training alone can- not account for the level of sex segregation one observes. Not all female jobs are lower in their oc- cupational rewards when compared with all male jobs, either. While female occupations pay less, on the average, than male occu- pations pay to either male or female incum- bents, certain white-colIar female occupa- tions have relatively high status and relatively good working conditions. And while female occupations are described as giving little chance for advancement, some upward mo- bility may be possible. Men in these occu- pations are often the ones who take advan- tage of these possibilities. Oppenheimer (1970) has suggested that for some men of lower socioeconomic status, such female jobs may be seen as a move up, perhaps a step on the way to a managerial position. For example, men in teaching tend to end up disproportionately as principals and super- intendents (Schmuck et al., 1981~. In gen- eral, one finds that men in the female semi- professions are overrepresented in admin- istrative jobs (Grimm and Stern, 19741. (See also discussion in Sokoloff, 1980: 55-63.) In a case study of one firm that had encouraged people to move to jobs atypical for their gender, Schreiber (1979) found that men in clerical positions, in contrast with the women in those jobs, felt that these were jobs that
BOB CHANGING AND OCCUPATIONAL SEX SEGREGATION 59 would increase their chances to move into management. What is interesting about their perception is that it was not supported by the promotion records of men who had been clerical workers within the firm. Individual Life Changes Inclividuals' needs, employment behav- ior, and human capital can change over their lives. Not all men, for example, are highly committed to the work force over all their lives. Some are employed while still in school, although often in sex-typical occupations (Greenberger and Steinberg, 1983; Lewin- Epstein, 19811. Still, at this stage, a young man might end up in a typically female oc- cupation that is convenient for him e.g., the graduate student who is also a depart- mental secretaryand then change to a more typically male occupation after graduation. Many women have fairly long periods of their lives when they are not involved with childrearing and when they would be suit- able candidates for jobs that require a rel- atively long commitment (Kreps and Leaper, 19761. Schooling and training necessary to make an occupational change are possible later in life, as evidenced by the increasing numbers of women attending colDege at older ages (Heyns and Bird, 1982~. Other changes in a woman's life circumstances could lead her to seek a job considered atypical for her. Many of the descriptions of women's careers implicitly place women in families where the husband provides the main economic support; in such a setting the woman can make her job decisions using criteria other than income maximization. With increasing divorce, inflation, and unemployment rates, however, more women find themselves without husbands or with husbands who do not earn enough, or anything, to support the family. Under such circumstances, women may decide they can no longer afford to remain in low-paying, though sex appro- priate, jobs. Changes in Labor Demand Another source of individuals' mobility between sex-typical and atypical jobs is from changes in labor demand. A person may be able to fill an occupation usually held by the opposite sex, if the employer is experiencing a demand for labor in that occupation. For example, the feminization of clerical work and school teaching seems to have resulted from a scarcity of willing and suitably edu- cated men (Davies, 1975; Strober and Tyack, 1980; Oppenheimer, 1970), although the women who entered these jobs usually were not moving from another job but from out- side the labor force. World War II has al- ready been mentioned as an example where women were sucIdenly brought into men's jobs. After World War II, of course, women were just as brusquely forced out of the la- bor force or into the usual female jobs (An- derson, 19811. Dramatic changes in de- mand, then, can bring about equally dramatic changes in the sex distribution of occupa- tions, which strongly suggests the impor- tance of demand for maintaining and chang- ing occupational sex segregation. Legislation can also affect labor demand and, as a result, movement from sex-typical to sex-atypical jobs. The enactment of the Equal Employment Opportunity (EEO) ti- tle ofthe 1964 Civil Rights Act, its expansion and strengthening in 1972, and Affirmative Action legislation may have permitted and encouraged at least some women to move to jobs that previously were closed to them. Beller (1982, and in this volume) did find dramatic decreases in occupational sex seg- regation in the 1970s and some links of this decrease to legislative enforcement of EEO. At the same time, some movement back to sex-typical occupations may occur after individuals experience harassment by co- workers or difficulties in adjusting to work that was designed for the opposite sex (Schreiber, 1979; Gruber and Bjorn, 1982; Harlan and O'Farrell, 1982; Kanter, 1977; Boos and Reskin, in this volume).
60 RACHEL A. ROSENFELD The meaning of a move to a typical or atypical occupation might be quite different depending on whether the person is of a minority race. In the past, because of race discrimination in lower-level, male occu- pations, a black woman or man moving to a typically female occupation might find not oniv an increase in status. but also an in- thors make some initial analyses of the de- terminants of such change for women. ORGANIZATION OF THE PAPER This paper will examine both patterns and determinants of changes of occupational gentler typicality by sex and race. The next crease in pay and a job more in line with section describes the data for the analysis. her or his educational credentials. Many of The third section shows the tYnes of occu- the typical occupations for blacks, however, were sex typed as well as race typed; for example, private household workers were black women anti porters were black men. Among blacks, therefore, one might still ex- pect to see differences in occupational lo- cation and mobility by sex. It has been sug- gested that one result of the civil rights movement of the 1960s was the movement of some black males into higher level posi- tions typically occupied by white males and the movement of black women into the lower- paying and lower authority positions typi- cally filled by white women (Lewis, 1977; Jusenius, 1975; Malveaux, 19821. Analysis ~ ~ of mobility across sex-type boundaries needs ^ - ~ - ~ ~ ~ ~ to include differences related to race. Why a person has either a sex-typical or sex-atypical occupation has`;received re- markably little investigation (see Polacheck, 1979, 1981b; England, 1982b; Jusenius, 1975; Daymont and Statham, 1981; Corcoran et al., in this volume; and Beller, 1982 for some of the exceptions to this statement). Even less has been studied regarding the extent and determinants of changes in occupational sex-type during the adult years. Some case studies (e.g., Schreiber, 1979; Kanter, 1977; McIlwee, 1982; Dresse} and Petersen, 1982) provide interesting hints about what hap- pens, but they do not generalize their find- ings to the occupational structure as a whole. Using national data, Jusenius (1975) and Corcoran et al. (in this volume) have shown that some black and white women move among male occupations; Corcoran and co- workers show that at least some black and white men change, as well. Both sets of au- . , , ~ pational mobility undertaken by women and men, black and white. The fourth section examines the individual and job-level de- terminants of a move to or out of a sex- atypical occupation. The final section sum- marizes the results and discusses their im- plications. DATA The data for this paper come from the January 1973 Current Population Survey (CPS) and its supplemental questionnaire that surveyed persons with new jobs since January 1972. Data were selected on out- of-schoo! whites and blacks 20 to 50 years of age who had new employers at the time of the CPS and who had been employed at some time in the preceding 5 years.3 The data set includes information on the re- spondents' previous jobs, their 1973 jobs, the way in which they spent time between jobs (if there was such a period), the reasons 3 In January 1973, respondents were asked regarding each r`^rc~n in the CP~ "Wac ~7 ~ an_.. ~~ ^ .~ ^ v, ~ . . . doing the same kind of work a year ago, in January 1972?" The supplemental questionnaire was distributed to those over 16 who had new employers since January 1972 and who were not self-employed or working without pay in a family busi- ness in January 1973; the questionnaire asked about the previous job and about the job search. Although the question was meant to include persons who changed occupation or job with the same employer, almost all respondents seemed to interpret the question as re- ferring to employer shifts. Of the 102,374 people about whom the January 1973 CPS inquired, only 326 were reported as changing jobs with the same employer, too few to include in the analysis here.
OB CHANGING AND OCCUPATIONAL SEX SEGREGATION 6 that they changed jobs, and such personal characteristics as marital status, age, and ed- ucation. These data enable one to examine the job-changing process and the extent and determinants of changes in the sex compo- sition of occupations held by individuals. The timing of this survey on employer changers was fortunate, since January 1973 is a particularly interesting period in which to examine gender differences in job shift- ing. It was just after EEO legislation en- forcement became stricter. It was also a time of recovery in the business cycle, when jobs may have been more available and the ef- fectiveness of EEO legislation enhanced. The availability of data on a relatively large sample of persons that changed employers during a particular period is also fortunate. Selecting only employer changes avoids confounding general inertia with the bar- riers to changing occupational type faced by movers. Of course, people can change jobs within the same employer. (Data on job shif- ters within a firm were not available from the CPS; see footnote 3.) One might expect, however, that the chances for changing from a sex-typical occupation to a sex-atypical oc- cupation would be greatest across employ- ers. Those changing jobs within the same firm might be more likely to continue along some job ladder rather than changing job ladders. Jusenius (1975) found that, among Some of those with a new employer had never been employed before or had not been employed within the last 5 years. New labor force entrants will, by defini- tion, not have had a previous job and so are not really job shifters. They are not included in the analysis. Fur- ther, the CPS supplemental questionnaire asked for detailed information about the job held previous to the January 1973 job only if it was one on which the re- spondent worked in 1968 or later. The data for the analysis presented in this paper are thus on those who had changed employer some time in the 5 years pre- ceding January 1973. Previous job refers to that job immediately before the job in January 1973. For most people, this w`as a job held during 1972. For others, it may be separated from the J`anu`ary 1973 job by a period of up to 5 ye`ars out of the labor force or unemployed. white women aged 30 to 44 in 1967, about 50 percent of those who moved from an oc- cupation predominantly of one sex to one predominantly of the other sex between 1967 and 1971 did so within a given firm. For black women, however, she found "chang- ing type of occupation either to or from a typical one was much more likely to be accompanies! by an employer change." Fur- ther, "these results are in part explained by the listing of occupational changes.... Among white women, there were those whose movement into an atypical occupa- tion appeared to be an intrafirm promotion, from operative to foreman, for example. Among the black women, however, the atypical jobs into which women moved were considerably different from their previous, typical employment, from private house- hold worker to operative, for example" Ju- senius, 1975:28~. There is considerable sex segregation within firms (Blau, 1977~. Occupations that are atypical for a given sex in the labor mar- ket as a whole may be predominantly of that sex within a particular firm. Some of the intrafirm job changes that Jusenius labelled as changes in type of occupation may have actually been changes from one predomi- nantly female (male) job to another predom- inantly female (male) job in that particular firm. While using data on employer chang- ers does not solve the problem of possible inconsistency between the sex composition of an occupation as a whore and the sex com- position for a given job, it does increase the chances that a change in occupational type represents a real change on the job. More will be said about this in the conclusions. While there are advantages to using this particular sample, some important statistical problems are raised in limiting the study to those persons changing employers. In ana- lyzing the outcomes of job changes across employers, one would like to be able to gen- eralize these results to all potential job-shift- ers, including: (1) those persons who re- mained with their current employers from
62 RACHEL A. ROSENFELD . . January 1972 until January 1973 but who might, if observed longer, change employ- ers after January 1973; (2) those persons pre- viously employed who were not in the labor force or who were unemployed and between jobs as of January 1973; and (3) those persons who did not have any previous job. Select- ing only those persons for study who changed employers over a given period could bias coefficients estimated on the selected sam- ple (Berk and Ray, 1982; Heckman, 1979; Barnow et al., 1980; Olsen, 1980~. The au- thor was not able to investigate all forms of selectivity in these data. The author did, however, use procedures described by Ray et al. (1981) to create a selectivity measure for moving from, as compared to staying with, an employer, 1972 through 1973. This meas- ure was highly collinear with tenure on pre- vious job for those who shifted jobs, not a surprising result given the decline of job mobility with tenure (Hall, 1980~. Tenure on the job, then, differentiates those who changed employers from those who stayed with the same employer between January 1972 and January 1973. Since the correction proposed by Ray et al. (1981) is for an or- dinary least squares model and this study used logits for much of the analysis, a direct measure of tenure on previous job to control for selectivity bias is used here rather than the derived selectivity measure. The selection of those who change em- ployers over those who remained with an employer is probably most significant for adult males, especially white adult males. For women, especially white women, the dis- tinction between being in the labor force at all and not in the labor force can be equally important. The women in the sample were employed at a given time, January 1973. Only about 50 percent of all adult women are employed at a particular time. However, selecting on employment at any given time does not seem to bias cross-sectional anal- ysis of women's occupational rewards and characteristics (Corcoran et al., in this vol- ume; Fligstein and Wolf, 1978~. They were also required to have had some previous job in the last 5 years, perhaps thus selecting on those with more continuous labor force participation. At the same time, since this previous job could be one held at any time between 1968 and January 1973, the selec- tivity bias is probably less than if the re- quirement for inclusion in the analysis had been employment on two particular dates (Corcoran et al., in this volume). DESCRIPTION OF OCCUPATIONAL SEX COMPOSITION AND EMPLOYER CHANGING Table 4-1 describes the current and pre- vious occupations of the sample in terms of the average percent male.4 White males were most segregated on this measure: On the average, they held occupations composed of over 80 percent of the same sex. Black men were in somewhat less sex-segregated oc- cupations. Women, both white and black, at the mean tended to be in occupations that were only 26 to 28 percent male (or, con- versely, 74 to 72 percent female). As a result of changing employers, there was little change in average sex composition of occupations for any group.5 The overall low average change in percent 4 In general, percent male was calculated for each 3- digit occupational code from the 1970 U. S. census. Ten large, miscellaneous groups of occupations (including clerical workers not specific and miscellaneous, man- agers and administrators not elsewhere classified, and inspectors n.e.c.) were further broken down by indus- try; see Spenner, 1977, for further details. Using the 1970 data to categorize occupations is problematical in that at least some occupations changed their sex com- position between 1971 and 1974 (Belier, in this vol- ume). It is not clear that such changes affect a large number of people. One should keep in mind, however, as discussed above, that just as individuals can change to occupations with different sex types, so whole oc- cupations over time can change their sex composition and sex label. 5 Since the CPS was a multistage rather than simple random sample, the significance levels in the tables are only approximate. Using the weights provided with the CPS, however, did not change the distribution of the dependent or independent variables.
IOB CHANGING AND OCCUPATIONAL SEX SEGREGATION 63 TABLE 4-1 Description of lob Shifts and Occupational Sex Composition by Race and Sex: 1973 CPS Employer Changers Black White Black White Women Women Men Men Mean percent male of previous occupation 26.4 27.3 76.0 81.8 Mean percent male of 1973 occupation 28.9 27.3 76.6 82.1 Average change in percent male (percentage points Correlation between percent male of previous and 1973 occupationsb N 2.23 34 . - 44 . .32 .86 .26 185 2009 176 2358 .29 a None of the changes is statistically significantly different from O at the .05 or .1 level. b All correlations significantly different from 0 at the .05 level. male might be interpreted as indicating lit- tle change of occupational sex type with a change of employer. This would be consist- ent with the representation of the labor mar- ket as sex-segregated and preventing any change to less sex-typical jobs by members of a given sex. Averages, though, can hide large changes in both directions, i.e., to and away from occupations with high propor- tions of the respondents' sex. As Table 4-1 also shows, there are surprisingly low cor- relations between the sex composition of the previous and the January 1973 occupations of the CPS employer changers, from a high of .34 for black women to a low of .26 for black men. Such low correlations suggest considerable change in occupational sex type with a change of employer, perhaps some- what less so for women (black and white) than for men. One hypothesis to account for these results is that something is wrong with the data. Others (e.g., England, 1982b), though, have found similarly low correla- tions. While lower than might have been ex- pected, the correlations are positive. One can imagine that what is happening is that people are changing occupations within ranges of sex composition. For example, women could be shifting easily among oc- cupations with anywhere from 0 to 30 per- cent male but be stuck at a hypothetical 30 percent male barrier. The image of the labor market is of barriers across which it is dif- ficult to move (e.g., Sokoloff, 19801. One would like, therefore, to go from a contin- uous measure of occupational sex composi- tion to a categorical or ordinal measure of occupational sex-type or typicality. While most occupations can be iclentified as male or female, the particular coding of occupations into these categories is some- what arbitrary. In the remainder of this pa- per, occupations in which men are the ma- jority (at least 51 percent of those in the occupation) will be labelled "male," "atyp- ical for women," or "male dominant," while those with less than 51 percent male will be called "female," "atypical for men," or "not male dominated."6 Table 4-2 describes the employer changes of the 1973 CPS sample in terms of these categories. While there is considerable sex segregation, some people do move from or to sex-atypical jobs. Ten to 15 percent move from a sex-typical to a sex-atypical occupa- tion with a change of employer; over 60 per- cent move back to a sex-typical one. While 6 This categorization is similar to that used by Cor- coran et al. (in this volume). The distribution of race and sex groups over the occupational percent male did not show any natural breaking points, but it was ex- tremely skewed. The median percent male of the 1973 occupation was 16 for black women, 16 for white women, 87 for black men, and 93 for white men. Since most people are in the tails of the distributions, the exact points chosen to separate male from female occupations should have relatively little importance as long as they are somewhere in the middle of the distribution.
64 RACHEL A. ROSENFELD TABLE 4-2 Sex-type of 1973 Occupation by Sex-type of Previous Occupation by Race and Sex: 1973 CPS Employer Changers Black Women White Women Previous Occupation Previous Occupation Male- Not Male- Male- Not Male- Dominated Dominated Total Dominated Dominated Total 1973 Occupationa (%) (%) (%) (%) (%) (%) Male-dominated 39.0 15.3 20.5 36.2 15.9 21.0 Not male-dominated 61.0 84.7 79.5 63.8 84.1 79.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 N 41 144 185 503 1506 2009 % 22.2 77.8 100.0 25.0 75.0 100.0 Black Men White Men Previous Occupation Previous Occupation Male- Not Male- Male- Not Male- Dominated Dominated Total Dominated Dominated Total 1973 Occupation (%) (%) (%) (%) (%) (%) Male-dominated 86.5 62.9 81.8 90.2 71.6 88.4 Not male-dominated 13.5 37.1 18.2 9.8 28.4 11.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 N 141 35 176 2136 222 2358 % 80.1 19.9 100.0 90.6 9.4 100.0 a Male-dominated occupations are those that are at least 51 percent male. the flow is stronger in the direction of sex- typical occupations, at least at the level of census occupational cocles, the barriers be- tween male and female occupations are not impermeable. 7 In Table 4-2, results are gen- erally consistent with other descriptions of occupational sex type by race. Black women as compared with white women, and black men as compared with white men, are 7 The movement across occupational sex types as de- fined here usually involves a relatively large change in the sex composition of a person's occupation; see Ap- pendix A. (Results for blacks, not shown, are similar to those in Appendix A.) The majority of those who stay with an occupation labelled male or female hold occupations after their employer shift that are within 10 percentage points of the percent male of their pre- vious occupation. A majority of those crossing the sex- type boundaries move to an occupation with a sex com- position that differs by more than 50 percentage points from that of their previous occupation. Further, the direction of the change is as one would expect: Those women going from typical to atypical occupations, for example, are going to occupations that are considerably more male than their previous occupations. somewhat more likely to be in female oc- cupations, which is what Malveaux (1982) reports. Here, though, one floes not see a greater tendency for black women than white women to move to female occupations. The mobility patterns of women by race are sim- ilar. Black men, on the other hand, show greater movement than white men out of male-dominated occupations and less move- ment to male occupations. As Malveaux (1982) and others have em- phasized, in making comparisons of sex type of occupations (and mobility between types) by race, one needs to keep in mind the race differences in occupation within the cate- gories of male and female occupations. One way of understanding race and sex differ- ences in mobility across or within occupa- tional sex-type categories is by examining the outcomes of such mobility i.e., the types of occupations where people work after a change. Table 4-3 shows these outcomes by race, sex, and sex type of 1973 occupa- tion. Typical occupations for the white women
OB CHANGING AND OCCUPATIONAL SEX SEGREGATION 65 job shifters are clerical occupations, al- though 13 percent end their employer changes as health workers or teachers and 19 percent as service workers. Atypical oc- cupational outcomes for white women tend to be within the professional and managerial categories, though 16 percent are durable and nondurable goods operatives. While al- most 40 percent of the black women who end up with sex-typical 1973 occupations are clerical workers, more of them as comparecl with white women have sex-typical blue- colIar jobs as operatives, nonhousehold service workers, and household workers. Black women's atypical occupations are roughly comparable to white women's atyp- ical occupations. Among white men, sex-typical occupa- tional destinations are most often skilled ant] semiskilled blue collar. Almost one-quarter of those with male-dominated 1973 occu- pations have crafts occupations, while an- other one-fifth are operatives. Atypical oc- cupations for white men after a job shift are in lower white-collar retail sales and clerical occupations, as well as in durable and non- durable operatives and service categories. For black men, both sex-typical and sex- atypical occupational outcomes tend to be lower skilled, as compared with the out- comes for white men. More of those with sex-typical destinations, as compared with white men (or women), are laborers or serv- ice workers. While 18 percent of the black men with sex-atypical outcomes are clerical workers and 12 percent are health care workers and teachers, another 35 percent are service workers. Another approach to understancling race and sex differences in mobility across and within sex-type boundaries is to look at the outcomes of the mobility in terms of relative job rewards. Table 4-4 shows changes in sta- tus and wages by race and sex for different types of moves. Consistent with Wolf and Rosenfelcl's (1978) findings, all sex and race groups mov- ing to a male-dominated occupation gained the most, or close to the most, in terms of status, despite the greater proportion of white-collar jobs held by white women and black and white men who were in female occupations. Going to or even staying in male- clominated occupations, however, does not necessarily result in the greatest increases in wages. For white men, moving to a fe- male occupation may not increase status, but it does increase wages more than any other type of move. For black women, those going to typically female occupations ex- perience the greatest gains. Moving to a male- dominated occupation does provide greater wage increases for white women than other types of moves, but white women do not significantly gain in wages by going from one male occupation to another. The failure of male occupations to be wage growth occu- pations for women has been found else- where (Corcoran et al., in this volume; Ju- senius, 1975; Rosenfeld, 1983~. At the same time, those who stayed in male-dominated occupations as compared with others in their race/sex category have the highest 1973 wage levels. Even among people changing employers, then, there is a considerable amount of sex segregation in both the origin and the des- tination occupations. At the same time, be- tween 10 and 15 percent, depending on race and sex category, go from a sex-typical to a sex-atypical occupation with their employer move. A considerably larger proportion, 60 to over 70 percent, of those who previously had a sex-atypical occupation had a sex-typ- ical one in 1973. The sex-type distribution and mobility across occupations differ be- tween the sexes and between the races. Black women resemble white women in their dis- tributions and mobility more than black men resemble white men. Yet even among women the characteristics of jobs labelle(1 male or female and the consequences of mobility within and across categories differed by race. Such differences need to be kept in mind when interpreting sex and race differences and similarities in individuals moving to or between sex-atypical occupations, the sub- ject of the next section.
RACHEL A. ROSENFELD TABLE 4-3 Occupational Destinations by Sex, Race, and Sex-type of 1973 Occupation: 1973 CPS Employer Changers Black Women Occupational Destinations Typicalb Atypical 1973 Occupation (%) (%) Total Engineer, physician, dentist 0 2.56 .54 Health worker, teacher (except college) 2.04 10.26 3.77 Engineering and science technicians 0 0 0 Other professional, salaried 1.36 17.95 4.84 Manager, salaried, manufacturing 0 0 0 Manager, other, salaried 0 7.69 1.61 Sales retail 3.40 0 2.69 Sales other .68 2.56 1.08 Clerical 38.10 20.51 34.41 Crafts 0 7.69 1.62 Durable and nondurable goods operative 15.64 17.95 16.13 Other operative 2.04 10.25 3.76 Nonfarm labor , 0 0 0 Private household worker 10.88 0 8.60 Service worker 25.85 2.56 20.97 Farmer, farm manager 0 0 0 Farm laborer, foreman 0 0 0 Totala 99.99 99.98 99.99 N 147 39 186 Percent white collars 46 62 White Women Engineer, physician, dentist 0 .71 .15 Health worker, teacher (except college) 12.82 11.85 12.61 Engineering and science technicians 0 1.42 .30 Other professional, salaried 1.32 17.54 4.72 Manager, salaried, manufacturing 0 .71 .15 Manager, other, salaried 0 13.03 2.73 Sales retail 5.40 1.18 4.52 Salesother 0 6.64 1.39 Clerical 47.36 12.32 40.02 Crafts .25 3.09 .85 Durable and nondurable goods operative 9.74 16.12 11.07 Other operative 1.76 5.45 2.54 Nonfarm labor .06 4.26 .95 Private household worker 2.32 0 1.84 Service worker 18.98 4.97 16.04 Farmer, farm manager 0 0 0 Farm laborer, foreman 0 .71 .15 Totala 100.01 100.00 100.03 N 1596 422 2014 Percent white collars 67 65 a Totals differ from 100 due to rounding. b A typical occupation is one that is male dominated (more than 50 percent male) for men and one that is not male dominated for women. c White-collar occupations are those in the major occupational categories of professional and technical, managerial and administrative, clerical, and sales.
TOB CHANGING AND OCCUPATIONAL SEX SEGREGATION - 67 TABLE 4-3 (Continued) Black Men 1973 Occupation Engineer, physician, dentist Health worker, teacher (except college) Engineering and science technicians Other professional, salaried Manager, salaried, manufacturing Manager, other, salaried Sales retail Sales other Clerical Crafts Durable and nondurable goods operative Other operative Nonfarm labor Private household worker Service worker Farmer, farm manager Farm laborer, foreman Toting N Percent white collars Occupational Destinations Typicalb Atypical (~) (%) Total .68 1.35 o 7.43 1.35 5.4 o 1.35 4.05 4.86 0.81 12.84 22.30 o 4.86 o 2.70 99.99 48 22 Engineer, physician, dentist Health worker, teacher (except college) Engineering and science technicians Other professionals, salaried Manager, salaried, manufacturing Manager, other salaried Sales retail Salesother Clerical Crafts Durable and nondurable goods operative Other operative Nonfarm labor Private household worker Service worker Farmer, farm manager Farm laborer, foreman Totala N Percent white collars o .76 o 2 94 . o o 5.88 o 7.65 o 23.52 2.94 o o 35.28 o o 99.97 34 38 White Men 3.09 1.97 2.34 8.25 2.48 9.14 1.73 5.90 3.98 24.46 8.10 2.93 8.9s o 5.00 .19 1.45 99.96 .55 3.30 o 6.59 1.10 4.40 .10 1.10 6.59 2.09 3.18 0.99 8.14 o 8.69 o 2.20 00.02 82 o 9.22 o 3.55 o o 5.60 o 23.05 .71 27.30 3.19 o o 7.38 o o 100.00 2134 282 39 51 2.73 2.82 2.07 7.70 2.19 8.07 3.35 5.22 6.21 21.69 0.35 .80 7.90 o 6.48 .17 .28 100.03 2416 a Totals differ from 100 due to rounding. b A typical occupation is one that is male dominated (more than 50 percent male) for men and one that is not male dominated for women. c White-collar occupations are those in the major occupational categories of professional and technical, managerial and administrative, clerical, and sales.
68 RACHEL A. ROSENFELD TABLE 4-4 Mean Status and Wage Changes With Employer Changes by Occupational Sex-type, Sex and Race: 1973 CPS Employer Changers Black Women White Women Previous Occupation Previous Occupation 1973 Occupation Atypical Typical Atypical Typical Atypical Previous wagea 4.25 2.97 3.30 2.68 Current wage 3.84 3.70 3.54 3.10 Wage change -.42 .73 .24 .41* Previous SKI 53.7 32.3 48.0 39.6 Current SKI 60.4 39.3 48.9 46.4 SKI change 6.71 7.0 .9 6.8* Typical Previous wage 2.29 2.62 2.71 2.50 Current wage 3.30 2.59 2.54 2.83 Wage change 1.01* -.03 -.16 .33* Previous SKI 33.5 30.5 44.2 40.3 Current SKI 37.4 30.2 40.5 41.8 SKI change 3.9 - .3 - 3.7* 1.6* Black Men White Men Previous Occupation Previous Occupation 1973 Occupation Typical Atypical Typical Atypical Typical Previous Wage 3.54 2.98 3.98 2.75 Current wage 4.73 3.56 4.35 3.30 Wage change 1.181 .59 .37* .55* Previous SKI 26.5 28.9 36.0 28.8 Current SKI 25.3 32.1 38.3 37.7 SKI change 1.2 3.2 2.2* 8.9* Atypical Previous wage 3.42 2.81 3.16 3.11 Current wage 3.39 2.77 4.01 3.40 Wage change -.53 .25 .851 .30 Previous SKI 27.7 22.7 32.7 27. ~ Current SKI 30.0 25.9 33.9 30.0 SKI change 2.2 3.2 1.2 2.2 a All wages are expressed in 1972 dollars. * Change significantly different from 0 at the .05 level. t Change significantly different from 0 at the .1 level. PREDICTING MOBILITY TO AND BE117VEEN SEX-ATYPICAL OCCUPATIONS In this paper occupations have been di- chotomized into sex-typical (for men, oc- cupations with a male majority; for women, occupations with ~50 percent male) and sex- atypical jobs. Mobility between occupations for different race and sex categories was shown in Table 4-2 as the probabilities that either (1) persons who begin their employer change from a sex-typical occupation will end up in a sex-atypical one or (2) persons who start from a sex-atypical occupation will go to a sex-typical one. In this section the two kinds of probabilities are taken as the
TOB CHANGING AND OCCUPATIONAL SEX SEGREGATION 69 depenclent variables, to be explained by a set of individual and job characteristics.8 Since the dependent variables are dichot- omous, ordinary least squares is not an ap- propriate method of analysis. Logistic regression is used instead. The dependent variable with logit analysis is actually the logged odds of having one kind of occupa- tional destination versus the other kind. This paper, however, will sometimes refer to ef- fects of independent variables on the prob- ability of being in an atypical occupation, since a simple transformation of the coeffi- cients that does not change their signs does give the elect of the independent variable on the probability (see Hanushek and lack- son, 19771. Independent Variables While the various explanations of sex seg- regation usually focus on its extent and per- sistence rather than on individuals' mobility from or to sex-typed occupations, they do suggest what might affect mobility across occupational sex-type boundaries. For ex- ample, many labor supply explanations of why women are in female occupations em- 8 An alternative to analyzing mobility across occu- pational categories would be to use as the dependent variable the distance moved, measured by the differ- ence in percent male between the previous and the 1973 occupation. Percent male on the previous occu- pation could be included as one of the independent variables to control for the degree of sex segregation on the origin occupation. Such equations, however, are likely to violate the assumptions of ordinary least squares (OLS) regression. At least one of the independent var- iables (the lagged dependent variable) is usually cor- related with the error term. OLS regression then gives biased and inconsistent estimates. The degree of bias may vary over groups, making comparisons risky. If one wants to keep the emphasis on mobility, and es- pecially if the values on independent variables do not change (see Corcoran et al., in this volume), one can do little to correct this problem with data at only two times. (See Rosenfeld and Nielsen, 1984, for a discus- sion of this problem and of solutions when one has data over more than two times.) phasize sex differences in continuity, exten- siveness, and intensity of employment. If such explanations are valid, then both women and men who have less continuous, exten- sive, or intense employment should be more likely to go to or stay in female occupations. The variable marital status (presently married or not) used here indicates the ex- tent of a women's family responsibilities, since responsibilities accompanying marriage are often the reason for women's more inter- mittent and less intense employment (e.g., Sweet, 1973; Shaw, 19811. For women this variable would be expected, if anything, to decrease the probability of going to a sex- atypical occupation and to increase the probability of moving to a sex-typical one. For men, on the other hand, one would expect effects in the opposite direction, since responsibility to provide for a family has been described as part of men's motivation to re- main continuously employed in good jobs (see Duncan et al., 1972; Rumberger and Carnoy, 1980; and Rosenfeld, 1980, for some evidence consistent with this view).9 For women, continuity of employment and extent of family responsibilities are also measured by whether a woman was out of the labor force caring for her family be- tween her previous and 1973 job. Of course, it is possible that a woman reentering the labor force after a period of caring for a fam- ily is embarking on a fairly continuous work life; the data, however, do not give sufficient information on numbers or ages of children to predict this possibility. To the extent that there is statistically significant discrimina- tion by sex and age, one would expect women who leave employment for family care to be 9 An alternative explanation of this evidence is that men who are more motivated, who remain employed, and who achieve higher occupational rewards, are also those who are motivated toward and successful at get- ting and remaining married. A preferable variable would be changes in marital status rather than simply marital status as of 1973. The CPS, however, did not give information on previous marital status.
70 RACHEL A. ROSENFELD less likely to move out offemale occupations and more likely to move to them.~° This variable is not included in models for men, since almost none of them had taken occu- pational leaves for care for a family. Another kind of break in employment is unemployment, searching for a job an- other of the independent variables. Such a period between jobs does not necessarily lead people to jobs that allow or encourage low labor force attachment; a period of searching for a job may be necessary to find a better job or one that is unusual for a person of a given gender. Extent of at- tachment to a particular job is measured by whether a person had more than two jobs in 1972. (One generally would expect those with any jobs in 1972 to have at least two" one held previously and one held as of lan- uary 1973.) Tenure on previous job is in- i° Daymont and Statham (1981), who included num- ber and various age groups of children in an analysis of middle-aged black and white women's occupational atypicality, did not find significant negative effects of children on sex-atypical occupation. Beller (1980, 1982), using data on a less restricted age range, likewise found that a large number of children did not significantly decrease the probability of a woman being in a male occupation. ii Length of time unemployed between jobs was di- chotomized into no time and some time unemployed because the distribution was extremely skewed away from 0. The data contained a general measure of time not in the labor force between jobs, which was also positively skewed. This information was used to con- struct a variable representing whether persons had a period out of the labor force before their 1973 jobs (net of a period out for caring for a family by women). A third variable was created, indicating whether a job search had begun because a person left a previous job involuntarily, something that might be expected to modify the effects of having a period of unemployment. These three variables some unemployment between jobs, some time out of the labor force, and an invol- untary job search were highly intercorrelated. In- clusion of them in various combinations showed no significant effects of an involuntary search or a period out of the labor force on subsequent job attachment, nor modification of the effect of a period of unemploy- ment. eluded, but its interpretation as an indicator of job attachment is complicated by its role as a measure of selectivity into the sample. Intensity of employment is measured by whether the previous and the 1973 job were full-time (at least 35 hours a week). Women's occupations have been described as offering more flexibility in hours than typically male jobs. This flexibility, however, benefits em- ployers by enabling them to pay lower wages, give fewer fringe benefits, and more easily hire and fire (see, for example, Sokoloff, 1980:106-107~. Women are overrepresented among part-time workers, but they are also overrepresented among the involuntarily part-time (Barrett, 1979~. To the ex- tent that typically female occupations offer less than fi~-time employment, one might expect those male or female who seek or find full-time work after being employed part-time to be those who are also moving from female to male occupations. Con- versely, those taking part-time employment may be going to a more typically female occupation. With cross-sectional data, Beller (1980, 1982) found some evidence to sup- port this speculation. In general, though, the evidence for the effects of home re- sponsibility, continuity, and extent of em- ployment on the sex composition of wom- en'sand in the case of Beller's research, on men's occupations is weak (Belier, 1980, 1982; Corcoran et al., in this volume; Po- lachek, 1981b; England, 1982b; Daymont and Statham, 19811. One might expect sim- ilarly weak effects of these variables on mo- bility among occupations. Cohort, age, and stage in the work career may all affect the nature of one's occupation. Older women ancl men may have sex-role attitudes that make them less willing than younger people to enter atypical occupa- tions; they may also be less acceptable to employers as candidates for such occupa- tions. Beller (in this volume) shows that younger age cohorts as compared with older ones are less sex-segregated. Young persons who are new to the labor market, however,
71 may not yet have learned about the possi- bility of atypical jobs (Malveaux, 19821. Age is included as a measure of both cohort and age per se, while whether a respondent was in school in 1972 indicates stage in the work career. Many persons are employed while attending school (Lewin-Epstein, 19811; more persons have at least one period of employ- ment or other activity between leaving high school and entering college. Thus, attend- ance in school the previous year and em- ployment and out-of-school in 1973 do not necessarily mean that those persons have just entered the work force for the first time. At the same time, for men who have had a sex-atypical job while attending school, leaving school could mean acquiring a more sex-typical occupation. Education has been shown to be a major determinant of occupational achievement, ant] is among the independent variables here. In general, though, occupations with more women do not have a less educated work force and do not require less education than those occupations with more men (England et al., 1982~; this might lead one to hypoth- esize that education does not affect mobility between types of occupations. Research to clate, however, hints at curvilinear effects of education: Both low and high levels of ed- ucation increase the chances having an atyp- ical occupation (Belier, 1980; Polachek, 1981b; Daymont and Statham, 19811. Those with the least education would tend to ap- ply for and obtain jobs that required less skill or training and among which transfer- ability is relatively great. The atypical jobs for which they qualify would have as few skill requirements as the typical jobs they already held Jusenius, 1975:241. Jusenius found support for this idea among white middle-aged women: Those with fewer than 12 years of schooling as compared with those who had at least 12 years were more likely to go from typical to atypical occupations. At the other end of the spectrum, those with the credential of at least a college degree as compared with those without a college diplomamight be considered qualified to move into more traditionally male mana- gerial or administrative positions or may have taken majors that prepared them explicitly for atypical careers. Getting additional education or training could also influence whether a person moves to an atypical occupation. The CPS supple- mental questionnaire asked respondents, Did you take any occupational or educa- tional training courses or programs during the time you were looking for work?" The response to this question is included with the independent variables. Relatively few people did have any training, which pre- cluded coding the responses into general types of training. Training is a change in human capital; it allows one to change sex type of occupation only if one is trained for a sex-atypical occupation and if the training that the person took is accepted by em- ployers. lob training programs often offer sex-stereotyped training that is not useful in getting a private sector job (see Boos and Reskin, in this volume; Waite and Berry- man, in this volume). The CPS employer changers could have gotten their additional training from other than job programs (the source of the training is not specified), but if training was not aimed at sex-atypical work, it would not necessarily increase movement to or between sex-atypical jobs. Since skills required by particular jobs limit mobility and since the relatively high status of some female jobs encourages mobility, Specific Vocational Preparation (SVP) re- quired for the previous occupation and whether this occupation was white collar were also included among the independent var- iables. Especially among men, those who have white-collar jobs that require more skill couicl be those less likely to change their occupation, let alone the sex type of their occupation, across firms. Region (South or not) is included as a con- trol variable. In cross-sectional analysis of oc- cupational atypicality, Daymont and Statham (1981) found black middle-aged women in the
72 RACHEL A. ROSENFELD TABLE 4-5 Logistic Regressions for Moving From a Sex-typical Occupation to a Sex- atypical Occupation by Sex and Race: 1973 CPS Employer Changersa Black Women (N= 138) White Women (N= 1485) Intercept .45 - .96 Independent Variables Highest year of education completed 0-11 - 1.08 .32 1-3 years of college .002 .06 >3 years of college 2.80* 1.19** Training while job seeking .69 .14 Age 20-29 - 1.29 - .30 30 39 - 1.11 - .28 Currently married -.98 .06 In school in 1972 - 3.70* - .43 Taking care of a family before 1973 job - 1-94t .02 >2 different jobs in 1972 .64 .431 Some time unemployed before 1973 job .26 .11 Going from part-time to full-time work 1.92** .18 part-time to part-time work 2.081 -.15 full-time to part-time work .83 - .53* SVPC of previous job .28 - .05 Previous job white collar -.19 -.06 Tenure on previous jobs .341 .07 Region= South .95 - .06 x2 32.0 51.4 df 18 18 p .02 .00 D .21 .03 _ . . . * .01 < p ~ .05. ** p ~ .01. ~ .05 < p < .10. a A sex-typical occupation is one that is at least 51% male (for men) or less than 51% male (for women). Only those who were previously in a sex-typical occupation are included. The outcome variable = 0 if the 1973 occupation is sex-typical, = 1 if it is not. b Dummy variables are coded 1 = Yes, 0 = No. South less likely to be in sex-atypical occu- pations. The independent variables thus include measures of employment continuity and in- tensity, changes in life cycle stage, age, ed- ucation, training, and level of previous job. They do not, however, include labor demand variables, which will be discussed in the con- clusions. Results Table 4-5 gives the results by race and sex for moving from a sex-typical occupation to a sex-atypical one. The coefficients show the effects of the independent variables on the logged odds of ending up in a sex-atyp- ical rather than a sex-typical occupation. The chi-square statistic compares the estimated model with one including only the inter- cept, thus enabling one to calculate whether the estimated mode! as a whole is significant in predicting the outcome. The D statistic is equivalent to R2 and measures the overall fit of the mode} (Harrell, 19801. Means and standard deviations for the variables are in Appendix B. For all groups except white males, having
lOB CHANGING AND OCCUPATIONAL SEX SEGREGATION 73 TABLE 4-5 (Continued) Black Men White Men (N = 139) (N = 2105) Intercept Independent Variables Highest year of education completed 0-11 1-3 years of college >3 years of college Training while job seeking Age 20-29 30-39 Currently married In school in 1972 Taking care of a family before 1972 jobs >2 different jobs in 1972 Some time unemployed before 1973 job Going from part-time to full-time work part-time to part-time works full-time to part-time work SVPC of previous job Previous job white collar Tenure on previous jobs Region= South x2 df p D -.73 .20 1.05 3.75* 1.13 2.09t -.90 -.36 .93 .49 ·24 .17 1.37 - .14 -1.34 - .17 .16 30.02 16 .02 .20 88 22 .18 .11 .21 .41 ·33 -68** .16 .10 .16 .30 ·60* 13** .26 09t 13 7 16 .00 .03 c SVP = Specific Vocational Training. From Temme's (1975) aggregation from the Dictionary of Occupational Titles (3rd ed.~. 1 = short demonstration only, 10 = over 10 years of specific training required. ~ Tenure on previous job: 1 = less than 4 months; 2 = 4-6 months; 3 = 7-11 months; 4 = at least 1 year but less than 3; 5 = at least 3 years but less than 5; 6 = at least 5 years but less than 10; 7 = at least 10 years but less than 15; 8 = 15 or more years. e Not included for men because there were too few cases in this category to give a stable estimate. more than 3 years of college (in contrast with some other level of educational attainment) increases the chances of going from a sex- typical to a sex-atypical occupation. Remem- ber that for white women, their male-clom- inated and other occupations were about equally likely to be white collar, and for black women and men, sex-atypical occu- pations, as compared with sex-typical ones, were more likely to be white collar, and perhaps more likely to require college cre- dentials. In contrast, then, with what Ju- senius (1975) found in her cross-cIassifica- tion of education categories by type of occupation, the results here suggest that higher, rather than lower, educational levels enable one to break the sex-type barrier, unless one is a white male. For white men, education has no effect, although it is also true for them that those who hold typically female occupations are more likely to be in white-collar occupations. For white men, skill level of their pre- vious job is significant. The higher the skill level on their previous job, the less likely they are to go to an atypical job when they change employers. This result fits with the image of white men going to atypical jobs
74 RACHEL A . ROSENFELD only if such moves are to better jobs, an image suggested by other authors (e.g., Op- penheimer, 1970), as well as by the average wage increase white men earn with a shift from male-dominated to other occupations (Table 4-41. Getting additional training or education while looking for a job has no ef- fect for any race and sex group on chances for moving to an atypical occupation, per- haps because of the sex typing of such train- ~ng. For women and white men previously in sex-typical occupations, changes in amount of labor supplied had an effect on the type of occupational destination. Whitesmen as well as women who went from full-time to part-time work were also likely to be going to typically female occupations. Black women who went from part-time to full-time work (as compared with those with other patterns of hours) had an increased probability of going to occupations in which men were a major- ity. It is not clear whether part-time employ- ment represents involuntary underemploy- ment or a choice to work less than full-time. For both white women and the relatively few white men who go to part-time em- ployment, the move to part-time (and fe- male) work could be due to (1) other op- portunities closing (although whether the job search was involuntary did not change the size of the work intensity variables) that is, the choice might be between part-time work and no work or (2) white men and women might choose to take part-time em- ployment, which is more readily available in typically female occupations, in order to combine employment with other activities. One can speculate that black women are motivated by underemployment when they seek full-time work that is more often as- sociated with typically male occupations than with typically female ones. White women going from part-time to full-time work could be those who have taken time out from a full-time work career to raise their families and now are reentering the labor force in a typically female occupation. The movement from part-time to fil11-time employment thus has no effect on the sex type of their occu- pation. For black men, the move from full-time to part-time work or part-time to full-time work has no effect on the sex type of their 1973 occupations. Black men tended to have both typical and atypical occupations that were Tower in status and skill than other groups. For black men, even more than for black women, male and female jobs may be equally likely to require or provide less than full-time employment. In trying to interpret the effect of a change in hours, one needs to keep in mind that part-time work is not simply redundant with female occupations: The correlation be- tween percentage male of 1973 occupation and whether the job is full-time is only .14 for white men, .12 for black men, .05 for white women, and .06 for black women. Human capital explanations for why women are in typically female jobs predict that being presently married and having taken time out for family care reduce the chances that a woman will move from a sex-typical to a sex- atypical occupation. If employers exercise statistical discrimination on the basis of more than sex alone that is, to the extent that they see all married women or all women who have been out of the labor force as less committed to employment one would also predict negative effects ofthese variables on going to an atypical occupation. The only effect of family responsibility, however, is for white men: They are less likely to change from a sex-typical to a sex-atypical occupa- tion if they are married. In this analysis of movement from a sex- typical occupation, significant effects of age or work career stage are present only for blacks, and only strongly for black women. Black women who were in school the pre- vious year and who tended to be younger were less likely than others to leave female occupations (see also Malveaux, 19821. If op- portunities in traditionally female occupa- tions were just opening up to black women, one would not be surprised to find that those
[OB CHANGING AND OCCUPATIONAL SEX SEGREGATION TABLE 4-6 Logistic Regressions for Moving From a Sex-atypical Occupation to a Sex- typical Occupation by Sex: 1973 White CPS Employer Changersa White Women White Men (N = 493) (N = 222) Intercept 2.09t 2.20 Independent vanablesb Highest year of education completed 0-11 - .66* - .11 1-3 years of college - .26 - 1.11* >3 years of college - 1.66** - .53 Training while job seeking .27 .60 Age 20-29 ·57t .17 30-39 .31 1.14 Current married .39 -.32 In school in 1972 .55 1.84** Taking care of a family before 1973 jobs .24 >2 different jobs in 1972 .61t .62 Some time unemployed before 1973 job .04 .32 Going from part-time to full-time work .18 -.22 part-time to part-time works .45 full-time to part-time work 1.04** -.40 SVP of previous job - .27** - .09 Previous job white collar .60* .82t Tenure on previous job .09 - .05 Region = South .13 .28 x2 86.7 29.3 df 18 16 .00 .02 D .16 .13 .05 < p < .10 * .01 < p ~ .05 ** p 6 .01 a Only those whose previous occupation was sex-atypical are included. The outcome variable = 0 if the 1973 occupation is sex-atypical, = 1 if it is sex-typical. b See Table 4-5 for coding. c There were too few cases in this category to include men. 75 just leaving school were taking advantage of female occupations and keeping them. There is also a marginal negative effect of being under 30 on the chances that black men will stay with a male-dominated occupation. Since predominantly male occupations are the ones in which both black men and black women have the most to gain in terms of wage level, the younger blacks were at a disadvantage in the outcome location of their employer changes. What about going the other way, from a sex-atypical to a sex-typical occupation? Table 4-6 presents the results for those who have previous occupations that are sex-atypical. The coefficient estimates give the effects of the independent variables on the logged odds of ending the employer change in a sex- typical rather than a sex-atypical occupation. Table 4-6 does not include results for blacks, because of the small number of blacks who begin their job shifts from sex-atypical oc- cupations. Means and standard deviations for the variables are in Appendix B. Again, education helps predict the oc- cupational outcome. Here one finds the cur- vilinear effect of education reported from cross-sectional analyses. Women with less
76 RACHEL A. ROSENFELD than 12 years of schooling or more than 3 years of college are more likely to remain with a sex-atypical occupation than those with intermediate amounts of education. Fur- ther, the higher the specific skill prepara- tion required on the previous job, the lower the odds that a woman will go to a sex-typical occupation with a change of employer. As Table 4-5 showed, higher previous SVP also decreases the probability that white men will leave sex-typical occupations. Male oc- cupations requiring higher skills or educa- tion may be especially likely to keep their incumbents even across employers, al- though women with lower education stay with their atypical occupations, too. For men who began their employer change from a typically female occupation, it is having only a few years of college, as compared with other levels of education, that inhibits movement to a sex-typical occupation. The interpretation may again have to do with credentials. If some women s occupations offer relatively good positions to men, one might have expected that those men who had fifed higher-skiDed and white-collar jobs and who had more than 3 years of education would also be less likely to move from a female to a male occupation. This does not seem to be the case. For white men, having held a white-collar occupation actually in- creased the chances of leaving a sex-atypical occupation, although this effect was of mar- ginal statistical significance. If the mana- gerial and administrative women's occupa- tions do indeed offer men chances for promotion, perhaps men are moving to them in increasing numbers, resulting in occu- pations that are now more male, though con- sidered female. For women, having a pre- vious atypical occupation that was white collar, net of everything else, increases mo- bility to a sex-typical occupation. i2 |2 In Table 4-6, the opposite signs for white women of whether the previous job was in a white-collar oc- cupation and of its SVP look suspiciously like the result Some life cycle and age effects appear. Men who had sex-atypical occupations and were previously in school have greater odds of going to sex-typical occupations than those who were out of school in both 1972 and 1973. Younger women, as compared with older women, may have a somewhat greater tendency to stay with a sex-atypical occu- pation even across firms, although here the effect is only marginally significant. While marital status and family related interruptions in labor force participation again have no significant effects on women's oc- cupation-type mobility, changes in hours do. The women who go from full-time to part- time work are again more likely to move to typically female occupations. Thus, as in the analysis of mobility from sex-typical occupations in Table 4-5, edu- cation, skill requirements, life cycle stage, and hours employed play a part In precl~ct- ing who will move across sex-type bound- aries, although the way in which these var- iables explain mobility is somewhat different here. SUMMARY AND DISCUSSION The labor market is extremely sex seg- regated. Some people, however, do move across the barriers built upon sex-typicality of occupations. This paper used date on 1972 to 1973 black and white employer changers to examine this phenomenon. At the level of 3-digit occupational codes, this study found that of those persons beginning a job shift of multicollinearity. The correlation between whether the previous job was white collar and its SVP is mod- erately high about .48 for white women. For black women and men it was approximately .43 and for white men, .34. When white-collar occupation for the pre- vious job was dropped, the effect of SVP continued to be significantly negative for white women. When SVP was dropped, the effect of whether the previous job was white collar was positive, but not significant. For other groups and other kinds of mobility, including only SVP or only white collar did not change the re- sults.
JOB CHANGING AND OCCUPATIONAL SEX SEGREGATION 77 from a sex-typical occupation (defined in terms of whether the occupation had a male majority), 15 percent of both black and white women, 13 percent of black men, and 10 percent of white men moved to an occu- pation atypical for their sex. There was also a strong flow from atypical to typical occu- pations: Somewhat over 60 percent of women and black men made such a move, as well as over 70 percent of the white men. In general, the mobility patterns across sex types of occupations for black ant! white women resembled each other closely. Indeed, sex differentiated occupational locations and types of mobility to a much greater extent than did race. At the same time, there were race differences. Even among the women, the black women's female occupations were much less likely to be white collar ones than those held by white women, in part because of the overrepresentation of black women in typically female service occupations. Black women who were able to remain in pre- dominantly male occupations actually had higher wage levels and status gains than white women in male occupations. Among men, blacks were more likely to be in or move to female occupations, although those who re- mained in predominantly male occupations received the greatest wage gains of any of the four race by sex groups. For black men, too, both the male and female occupations they held were relatively unlikely to be white- colIar ones. For neither white nor black women was there much support for the idea that extent of family responsibilities influences the chance to move from or to a sex-typical oc- cupation. One could argue that this is be- cause the variables measuring family re- sponsibilities are not detailed enough. These results, however, are consistent with a num- ber of other studies that show the effects of marriage and children on women's labor force participation but not on the status, income, or sex-type of the occupation women hold once they are in the labor force. Much of the lore about why women get lower wages than men and about why they are in wom- en's occupations focuses on women's roles within the home. For men, on the other hand, responsibilities for a family are usually ignored. Finding effects of family status for men but not for women suggests that our stereotypes about the interface offamfly and employment need to be reexamined with respect to both women and men. Other variables, though, that indicate changing commitment to the labor force did have effects on the nature of women's and men's mobility between occupations with different sex types. Moving to part-time work was associated with moving to or staying with a typically female occupation for white women and men, while moving to a full- time job from one that had been part-time increased the probability of going to a typ- ically male occupation for black women. Having been in school the previous year also increased the chances of moving to a sex- typical job for black women and white men previously in typically female occupations. As discussed in the preceding section, it is not possible to determine whether part-time employment represents a choice about how many hours to spend on market work versus other activities or whether it represents in- voluntary underemployment. For those persons who have demands on their time beyond employment, it would be possible, in some cases, to extend the range of jobs open to them through such options as flex- time. Level of education influenced the types of occupation changes that persons made. Having more than 3 years of college, which in most cases would indicate having a col- lege degree, made it more likely that women and black men would go from sex-typical to sex-atypical occupations and that white women would go from one atypical occu- pation to another. Unfortunately, the CPS did not give information on college major. It is not clear, therefore, whether it is the degree as a credential or the substance of the degree that enables someone to go to
78 RACHEL A . ROSENFELD and stay with an atypical occupation. If de- mand is sex-segregatecI, then even having the training for an atypical occupation may not help the individual get the desired job; by virtue of gender, he or she will be con- sidere`1 inappropriate. If employers are pressured to desegregate the workplace, they may be more willing to hire people from other fields and train them on the job. With the decreasing sex segregation of college majors (Belier, in this volume; Heyns and Bird, 1982), it will be interesting to measure whether persons use such majors to follow atypical careers.l3 White women and men who held male occupations that required more skill were more likely to stay with a typically male oc- cupation when they changed employers. This effect, together with the effect of having a college education, suggests the attraction and retentive power of the higher-level male jobs. The effects of having less than a high school degree on a woman remaining with atypical occupations emphasize again the range of occupations that are predominantly male. It is not skill and training alone that keep women out of male occupations. It could be that the less-educated women who stay in sex- atypical occupations are trapped in low-level jobs; they remain in these jobs because the pay is higher than that which they could earn elsewhere. Daymont and Stream (198lj Fund that it is precisely among the blue-collar oc- cupations, which often have lower educational requirements, that one finds an advantage to \3 Using data on female college graduates in 1961, Bielby (1978) showed that those who had sex-atypical college majors were more likely than other women to stay with sex-atypical careers. Conversely, Hearn and Olzak (1981) reported data on 1976 college seniors that showed women were actually more likely than men to study vocationally specific majors. The jobs for which these typically female majors were preparing them, however, were lower in status than those anticipated by typically male vocationally specific majors. Hearn and Olzak were not able to follow these seniors to see how closely the various majors predicted types of ca- reers. being in a male-typed rather than a female- typed job. The analysis did not present strong age effects, such as one might have expected from changes in the degree of sex segrega- tion over cohorts (Belier, in this volume). It may be that both the strong movement back to sex-typical occupations and the failure to find age effects are the result of using data from a period when changes in the climate facilitating sex desegregation were just underway. These analyses provide some insights into the nature of male and female typical jobs by race and sex and into the mechanisms by which persons change from an occupation of one sex type to one of another sex type. The explanatory power, however, of the in- dependent variables taken together is low, as indicated by the D statistic. It.is espe- cially difficult to predict, for whites, who will leave a sex-typical for a sex-atypical oc- cupation, at least using these individual and job-shifting characteristics as the independ- ent variables. One reason for this could be that the individual and job-shifting variables were not detailed and extensive enough. Another reason could be that the factors that affect whether a person is in or moves to a sex-atypical occupation are outside the per- son. As already discussed, individual char- acteristics can seem to produce effects that, in reality, reflect outside forces. The de- mand for certain kinds of labor, e.g., the demand for women in atypical occupations, can work against even the strong sex role socialization that most of us receive. Kanter (1977), for example, recounts how within a given firm, management persuaded women against their initial resistance to move to more typically male positions. Within and among firms, employers vary in their en- couragement of persons of the wrong sex to apply and be hired for jobs; this variation could explain whether an individual ends up in a sex-typical or atypical job. Once a per- son is hired into a sex-atypical job, whether that person stays with the job may depend
IOB CHANGING AND OCCUPaTIONAL SEX SEGREGATION 79 on management's efforts to curtail harass- ment and to adapt work conditions so that both men and women can continue em- ployment there (Roos and Reskin, in this volume). Such efforts at desegregating the work- place usually depend on whether it is prof- itable to make them. Legislation can make it unprofitable not to hire and retain a sex- integrated labor force by enacting direct monetary sanctions and by creating a cli- mate in which expensive sex discrimination suits are possible. An unfavorable economy such as we have now, however, can redirect emphasis toward providing jobs and away from efforts to integrate jobs. While women may have a greater need to support their families now and may be more attracted to typically male jobs, competition for jobs in- creases the possibility of reverse discrimi- nation charges by male and white workers, as well as informal efforts to exclude women from male occupational territory. Rapid changes in decreasing occupational sex seg- regation and improving sex role attitudes occurred in the 1970s; social scientists and policy makers need! to monitor carefully what happens in the unfavorable economic cli- mate of the early 1980s if these gains are to persist. The dependent variables in this analysis were movements across occupation sex types that were created from the sex composition of 3-digit occupational codes. Information at the 3-digit census code level is generally the most detailed available for national samples. Sex segregation occurs within 3-digit codes. Firms are sex segregated (Blau, 1977), and within firms, women rarely work in the same jobs as men (Bielby and Baron, in this volume). It is not clear, then, that the women who move to male occupations or the men who move to female ones are really going to jobs shared with members of the opposite sex. One would hope that sensitivity to this level of measurement problem would lead more of those who design large surveys, es- pecially longituclinal ones, to include ques- tions on the sex composition of the respond- ents' specific jobs. The conditions on a person's job can be important in determin- ing that person's rewards and motivation to stay with the job. O'Farrell and Harlan (in this volume) have urged that serious efforts be made to study sex segregation within or- ganizations, where advancement and work conditions can be traced more precisely. The rewards of a given occupation over time may indeed depend on its overall sex composi- tion, with those occupations dominated by women offering lower rewards than other aspects of the occupation would predict (England and McLaughlin, 19791. Future research needs to include not only the ques- tion of why inclividuals enter and stay with sex-typical as compared with sex-atypical oc- cupations, but also the larger questions of why and how women, whose work is un- dervalued, have been excluded from the work that society values highly. ACKNOWLEDGMENTS The support of the Carolina Population Center and NSF grant DAR-79-1585 made this paper possible. I am grateful to referees and participants from the Workshop on Sex Segregation of lobs (National Research Council, Washington, D. C., May 24-25, 1982) for comments on an earlier paper that informed the work on this one, to David Maume and Roberto Fernandez for merging the data, to Kenneth Spenner for providing the sex composition data, to Franpois Niel- sen and Kathy Ward for their advice, and especially to Barbara Reskin for her exten- sive and helpful suggestions.
80 RACHEL A. ROSENFELD APPENDIX A Extent of Change in Percentage Male by Type of Occupation Change: White 1973 CPS Employer Changers White Women Type of Movea Typical to Typical to Atypical to Atypical to Typical Atypical Typical Atypical (%) (%) (%) (%) 1973 percentage male-previous occupation percentage male = (1973 occupation greater % male) 91 to 100 0 0 1.2 0 81to90 0 0 34 0 71 to 80 0 0 8.1 0 61to70 0 0 9.7 0 51to60 0 0 19.3 0 41to50 .9 0 22.7 .5 31to40 2.2 0 14.3 2.7 21to30 2.9 0 14.0 4.9 llto20 8.9 0 5.6 8.8 1 to 10 11.8 0 1.6 12.1 (No change) 0 42.4 0 0 34.1 (1973 occupation smaller % male) -1 to -10 13.1 2.1 0 13.2 -11 to -20 9.6 3.8 0 8.2 -21 to -30 4.7 5.0 0 10.4 -Otto -40 2.8 12.1 0 3.8 -41to -50 .7 20.5 0 1.1 -51 to -60 0 20.9 0 0 -61 to -70 0 16.3 0 0 -71 to -80 0 12.1 0 0 -81to -90 0 4.6 0 0 -9lto -99 0 2.5 0 0 Totals 100.0 99.9 99.9 99.8 N 1267 239 321 182
[OR CHANGING AND OCCUPATIONAL SEX SEGREGATION 81 APPENDIX A (Continued) White Men Type of Movea Typical to Typical to Atypical to Atypical to Typical Atypical T,vpical Atypical (%) (%) (%) (%) (1973 occupation greater % male) 91 to 100 0 1.9 0 0 81to90 0 5.2 0 0 71 to80 0 5.7 0 0 61 to 70 0 17.1 0 0 51to60 0 18.1 0 0 41 to 50 1.5 25.7 0 0 31to40 2.2 10.5 0 4.8 21to30 3.5 12.9 0 7.9 11 to20 5.9 2.9 0 3.2 1 to 10 18.6 0 0 11.1 (No change) 0 31.1 0 0 47.6 (1973 occupation smaller % male) - 1 to - 10 18.6 0 .6 15.9 -11 to -20 7.4 0 2.5 1.6 -21to -30 5.1 0 5.0 4.8 -31 to -40 4.9 0 10.7 3.2 -41 to -50 1.2 0 23.3 0 -51to -60 0 0 21.4 0 -61to-70 0 0 15.7 0 -71 to -80 0 0 11.3 0 -81 to -90 0 0 6.9 0 -91 to -99 0 0 2.5 0 Totalb 100.0 100.0 99.9 100.1 N 1926 210 159 63 a A sex-typical occupation is one that is at least 51 percent male for men or less than 51 percent male for women. b Totals differ from 100 due to rounding.
82 4- o C) ·= en an 8 C. C) . so a; C) Cal en c) ._ Cal I o - :> i_ o o d C ._ Ad, C ~ ~ An Cal Ct _ .. .w ~ ~ .o Ct o z.O ~ Pa .~ 3 .o.- =P Cal .o~ .~ C5g E E a .~ 3 .o~,4 ~ ·< Cal C) O ~ _ o ~ S ·o _ c~ "o ~ ~ ~ ~ o ~o · ~ c~ ~ o o - - ~ - c) - - - - o - - ~ - c~ ~ - c~ - c~ - c~ · - · - · - · - ~ ~ ~ ~ - ~ ~ ~ - - - - oo cv' ~ c~ ~ - . o ~ - - - - cs) ~ - o u, O - - - - ~ ~ ~ ~ o ~ ~ oo O - cO - · - · - - a, CD co C~ di _ CO ~ _ - - - - · - · - · ~ ~ ~ ^ O ~ ~ ~ C~ _ _ ~ ~ ~ ~ c~ ~ ~ ~ - cO ~ o - - - - - ~ ~ ~ ~ ~ ~ o ~ · - · - · co o ~ ~ - ~ ~ 'o ~ ~ c~ ~ ~ ~ o c~ · - · O - ~ cO ~ ~ ~ ~ ~ ~ - ~ ~ - - - - - ~ = - ~ - ~ ~ =^= ~ ~ - - - - - io c~ O c~ C~ ~ o° '~ ~ u ) - c~ - ~ - - - - - - - - - - ~Q ~ o ·~ O ~ - co) ~ o o c) . , E E ~ E ,6 <ts 3 ~ .= ~ ~ ~ ~ ~ ~, =, ~ =o ~
83 He ~ ~ ~ Cal Cal Go ~ on ~ o ~ ~ oo ~ U. ~ CO _ Cal _ Cal o ~ o Cal o ~ ~ ~ o ~ Cal ~ . . . . . . . · . · . . . · . . . _ _ _ _ _d _ C~~ _~ ~ C~ ~ ~C~ ~ ~ ~ ~ ~ = - ~ ~ ~m ~ C~ ~ _ o ~ o ~ o _ ~ C~ ~ _ ~ C~ ~ · _ _ ~ _1 C'd:, _ <e - C) E ~1 c P. o .m C . C~ ~ _ · _ · _ ·U. ~ _ o =~ C ==oo~ - ~^~ ~ ~ ~ o _ C~ o ~ _ _ C~ _ ~ ~ ~ ~ ~ o · _ _ · _ · _ · _ "t, _ ·C~ _ · _ ~ :- o _. C ~ ~ U) o ~ .Cq ~ ~ O 0 0 ~ ~ O C~ C~ _ ~ 0 0 C5' ~ 0 et ~ O C] ~ ~ _ ~ O C~ _d e~ b- C~ ~ ~ ~ ~ ~ ~ ~ O · _ · · CO _ $ C . _ C~ ~ ~ ~ ~ ~ · _ · ·O ~0 ~ - U, ~ ~ ~ ~ ~ ~ ~ h~ ~ c~ ~ ~ ~ ~ C~ ~ U, _ C~ _ C~ ~ ~ O _ . . . . . . . . . . . . _ _ _ _ ~ _ ~ ~ _ ~ · _ _ ~U, CD O $ - ~D O ~oo ~ 0 co ~ O O ~ _ C~ _ CO O C~ _ C~ 00 ~ ~ ~ ~ ~ ~ ~ _ _ ~ __ 3~ 3 :>~ ' o ~o~ ' _ C 8 ~ 3 j S S D 8 S - S c S . ~ 8 , z A ° c' ~ P. ~ ~ ~ ~ ~
84 RACHEL A. ROSENFELD REFERENCES Anderson, Karen 1981 Wartime Women: Sex Roles, Family Rela- tions, and the Status of Women during World War II. Westport, Conn.: Greenwood Press. Baker, Elizabeth 1964 Technology and Women's Work. New York: Columbia University Press. Barnow, Burt, Glen Cain, and Arthur Goldberger 1980 "Issues in the analysis of selectivity bias." In Ernst Stromsdorfer and George Farkas (eds.), Evaluation Studies Review Annual, Vol. 5. Beverly Hills, Calif.: Sage, pp. 43-59. Barrett, Nancy S. 1979 "Women in the job market: Unemployment and work schedules." In Ralph E. Smith (ed.), The Subtle Revolution: Women at Work. Washington, D.C.: The Urban Institute, pp. 63-98. Beller, Andrea 1980 "Occupational segregation by sex: Determi- nants and changes." Paper presented at the Population Association of America annual meeting, Denver, April 12. 1982 "Occupational segregation by sex: Determi- nants and changes." Journal of Human Re- sources 17:371-92. Berk, Richard, and Subhash Ray 1982 "Selection biases in sociological data." Social Science Research 11:352-98. Bielby, Denise 1978 "Career sex-atypicality and career involvement of college educated women: Baseline evidence from the 1960s." Sociology of Education 51:7- 28. Blau, Francine 1977 Equal Pay in the Office. Lexington, Mass.: Lexington Books. Cohn, Samuel 1982 "Synthetic turnover: A reinterpretation of hu- man capital models of occupational sex-typ- ing." Unpublished manuscript. Department of Sociology, University of Wisconsin-Madison. Cook, Alice, and Hiroko Hayashi 1980 Working Women in Japan. Ithaca, N.Y.: Cor- nell International Industrial and Labor Rela- tions Report #10. Davies, Margery 1975 "Women's place is at the typewriter: The fem- inization of the clerical labor force. " In Richard Edwards, Michael Reich, and David Gordon (eds.), Labor Market Segmentation. Lexing- ton, Mass.: D. C. Heath, pp. 279-95. Daymont, Thomas, and Anne Statham 1981 "Occupational Typicality: Changes, causes, and consequences." In Lois Shaw led.), Dual Ca- reers, Vol. 5. Columbus, Ohio: Center for Hu- man Resources Research, Ohio State Univer- sity, pp. 107-139. Doeringer, Peter, and Michael Piore 1971 Internal Labor Markets and Manpower Anal- ysis. Lexington, Mass.: D. C. Heath. Dressel, Paula, and David Petersen 1982 "Becoming a male stripper: Recruitment, so- ciali7~tion, and ideological development." Work and Occupations 9:387-406. Duncan, Otis Dudley, David L. Featherman, and Bev- erly Duncan 1972 Socioeconomic Background and Achievement. New York: Seminar Press. Edwards, Richard 1979 Contested Terrain. New York: Basic Books. England, Paula 1982a "Explanations of occupational sex segregation: An interdisciplinary review." Unpublished manuscript, University of Texas at Dallas, De- partment of Sociology and Political Economy. 1982b "The failure of human capital theory to explain occupational sex segregation." Journal of Hu- man Resources 17:358-70. England, Paula, Marilyn Chassie, and Linda Mc- Cormack 1982 "Skill demands and earnings in female and male occupations." Sociology and Social Research 66:147-68. England, Paula, and Steven D. McLaughlin 1979 "Sex segregation of jobs and male-female in- come differentials." In Rodolfo Alvarez, Ken- neth Lutterman, and Associates (eds.), Dis- crimination in Organizations. San Francisco: Jossey-Bass, pp. 189-213. Fligstein, Neil, and Wendy Wolf 1978 "Sex similarities in occupational status attain- ment: Are the results due to the restriction of the sample to employed women?" Social Sci- ence Research 7:197-212. Greenberger, Ellen, and Lawrence Steinberg 1983 "Sex differences in early labor force experi- ence." Social Forces 62:467-86. Grimm, James, and Robert Stern 1974 "Sex roles and internal labor market structures: The 'female' semi-professions." Social Prob- lems 21:690-705. Gruber, James, and Lars Bjorn 1982 "Blue-collar blues: The sexual harassment of women auto workers." Work and Occupations 9:271-98. Hall, Robert E. 1980 "The importance of lifetime jobs in the U.S. economy." Cambridge, Mass.: National Bu- reau of Economic Research Working Paper #560. Hanushek, Eric A., and John E. Jackson 1977 Statistical Methods for Social Scientists. New York: Academic Press. Harlan, Sharon, and Brigid O'Farrell 1982 "After the pioneers: Prospects for women in nontraditional blue collar jobs." Work and Oc- cupations 9:363-86.
TOB CHANGING AND OCCUPATIONAL SEX SEGREGATION -r' Harrell, Frank 1980 "The LOGIST procedure." In SAS Supple- mental Library User's Guide: 1980 Edition. Cary, N.C.: SAS Institute, Inc., pp. 83-102. Hartmann, Heidi 1976 "Capitalism, patriarchy, and job segregation by sex." SIGNS 1:137-69. Polachek, Solomon Hearn, James, and Susan Olzak 1981 "The role of college major departments in the reproduction of sexual inequality." Sociology 1979 of Education 54:195-205. Heckman, James 1979 "Sample selection bias as a specification error." Econometrica 47:153-61. Heyos, Barbara, and Joyce A. Bird 1981a 1982 "Recent trends in the higher education of women." In Pamela Perun (ed.), The Under- graduate Woman: Issues in Educational Equity. Lexington, Mass.: Lexington Books, pp. 43- 69. Jusenius, Carol 1975 "Occupational change, 1967-71." In Dual Ca- reers, Vol. 3. Columbus, Ohio.: Center for Hu- Ray man Resource Research, Ohio State Univer- Ace; sity, pp. 21-35. Kanter, Rosabeth Moss 1977 Men and Women of the Corporation. New York: Basic Books. Kreps, Juanita, and R. John Leaper 1976 "Home work, market work, and the allocation of time." In Juanita Kreps (ed.), Women and the American Economy. Englewood Cliffs, N.J.: Prentice-Hall, pp. 61-81. Levinson, Richard M. 1975 "Sex discrimination and employment prac- tices: An experiment with unconventional job 1980 inquiries." Social Problems 22:533-43. Lewin-Epstein, Noah 1981 Youth Employment During High School. Re- port to National Center for Education Statis- tics, U.S. Department of Education. Chicago: National Opinion Research Center. Lewis, Diane 1977 "A response to inequality: Black women, rac- ism and sexism." SIGNS 3:339-61. Malveaux, Julianne 1982 "Recent trends in occupational segregation by race and sex." Paper presented at the Work- shop on Job Segregation by Sex, National Academy of Sciences, Washington, D.C., May 24-25. McIlwee, Judith 1982 "Work satisfaction among women in nontra- ditional occupations." Work and Occupations 9:299-336. Olsen, Randall 1980 "A least squares correction for selectivity bias." Econometrica 48:1815-20. Oppenheimer, Valerie 1970 The Female Labor Force in the United States: Demographic and Economic Factors Govern- 85 ing Its Growth and Changing Composition. Population Monograph Series, No. 5. Berke- ley: University of California. Phelps, Edmund 1972 "The statistical theory of racism and sexism." American Economic Review 62:659-61. 1978 "Sex differences in college major." Industrial and Labor Relations Review 31:498-508. "Occupational segregation among women: Theory, evidence, and a prognosis." In Cyn- thia Lloyd, Emily Andrews, and Curtis Gilroy (eds.), Women in the Labor Market. New York: Columbia University Press, pp. 137-57. "Occupational self-selection: A human capital approach to sex differences in occupational structure." Review of Economics and Statis- tics, pp. 60-69. 1981b "Occupational segregation: A defense of hu- man capital predictions." Unpublished man- uscript, University of North Carolina-Chapel Hill, Department of Economics. Subhash, Richard Berk, and William Bielby "Correcting sample selection bias for bivariate logistic distribution of disturbances." Unpub- lished manuscript, University of California-Santa Barbara. Revision of a paper presented at the 1980 meeting of the American Statistical As- sociation. Rosenfeld, Rachel A. 1977 "Breaks in women's employment and labor market re-entry." Paper presented at the an- nual meetings of the Canadian Sociology and Anthropology Association, Fredericton, New Brunswick. "Race and sex differences in career dynamics." American Sociological Review 45:583-609. 1983 "Sex segregation and sectors: An analysis of gender differences in returns from employer changes." American Sociological Review 48:637- 55. Rosenfeld, Rachel A., and Frangois Nielsen 1984 "Inequality and careers: A dynamic model of so- cioeconomic achievement." Sociological Meth- ods and Research 12:279-321. Rumberger, Russell, and Martin Carnoy 1980 "Segmentation in the U.S. labour market." Cambridge Journal of Economics 4:117-32. Schmuck, Patricia, W. W. Charters, Jr., and Richard O. Carlson (eds.) 1981 Educational Policy and Management: Sex Dif- ferentials. New York: Academic Press. Schreiber, Carol Tropp 1979 Changing Places: Men and Women in Tran- sitional Occupations. Cambridge, Mass.: MIT Press. Shaw, Lois 1981 "Causes of irregular employment patterns." In Lois Shaw (ed.), Dual Careers: A Decade of Changes in the Lives of Mature Women, Vol. 5. Columbus, Ohio: Center for Human Re-
86 RACHEL A. ROSENFELD source Research, Ohio State University, pp. 79-106. Sokoloff, Natalie 1980 Between Money and Love: The Dialectics of Women's Home and Market Work. New York: Praeger. Spence, Michael 1974 Market Signaling: Information Transfer in Hir- ing and Related Screening Processes. Cam- bridge, Mass.: Harvard University Press. Spenner, Kenneth 1977 From Generation to Generation: The Trans- mission of Occupation. Unpublished Ph.D. thesis, University of Wisconsin-Madison. Strober, Myra, and David Tyack 1980 "Why do women teach and men manage?" SIGNS 5:494-503. Sweet, James A. 1973 Women in the Labor Force. New York: Sem- inar Press. Temme, Lloyd 1975 Occupation: Meanings and Measurement. Washington, D.C.: Bureau of Social Science Research. Wolf, Wendy, and Rachel A. Rosenfeld 1978 "Sex structure of occupations and job mobil- ity." Social Forces 56:823-44.