male and female careers: sex-role and occupational stereotypes among high school students

Sex Roles, VoL 9, No. 6, 1983

Male and Female Careers: Sex-Role and Occupational Stereotypes Among High School Students 1

Sylvia Lifschitz: State University o f New York at Stony Brook

The present study combined two areas of research, occupational perceptions and sex-role stereotypes, in a 2 (Gender of Subject: male, female) × 2 (Gender of Character: male, Paul or David; female, Paula or Susan) × 2 (Gender of Occupa- tion: male, doctor or lawyer; female, nurse or secretary) between-sub/ects factorial design. High school students rated male and female characters in traditional and nontraditional occupations on the following six personality traits: ambitious, effective, emotional, intelligent, responsible, and traditional. The main finding was that occupational stereotypes were more prevalent than sex-role stereotypes. Reasons for this outcome are discussed and the implications o f the study in light o f past and future research are considered.

Inasmuch as research has indicated conclusively that people have definite, quan- tifiable attitudes about the personality characteristics associated with males and females (Broverman, Vogel, Broverman, Clarkson, & Rosenkrantz, 1972; Wil- liams, Bennett, & Best, 1975) and with various occupations (O'Dowd & Beards- lee, 1967), the next research issue should be the interface of these two stereotype systems. Limited research combining sex-role and occupational stereotypes has demonstrated that subjects' ratings of occupations are most influenced by the interaction of the gender of the stimulus person and the sexual dimension of the

i This article is based on part o f a doctoral dissertation submit ted by the author to the State University of New York at Stony Brook. The research was supported by U.S. Public Health Service Biomedical Support Grant 5 S07 RR 07067-12 to the State University of New York at Stony Brook during the summer of 1977. The author would like to thank Drs. Robert M. Liebert and Beverly Birns for their comments and the principal and teachers at Ward Melville High School in Setauket, New York, for their cooperation.

2All correspondence should be sent to Dr. Sylvia Lifschitz, 2130 Elder Street, Reading, Pennsylvania 19604.

725

0360-0025/83/0600-0725503.00/0 © 1983 Plenum Publishing Corporation

"726 Lifschitz

occupation (Shinar, 1978; Suchner & More, 1975; Touhey, 1974). Increased flexibility in occupational roles for both males and females raises questions about the limiting effects of existing sex-role and occupational stereotypes on career choice. Albrecht (1976) and Shinar (1975) found that male and female college subjects, as well as nonstudents, agreed on the characterization of over 100 occupations as masculine, feminine, or neutral. The specific parameters de- fining these gender-based distinctions in occupations need clarification.

Rosenkrantz, Vogel, Bee, Broverman, and Broverman '(1968) developed a sex-role questionnaire in the first study of a series that has since been used by other researchers (Touhey, 1974; Williams & Best, 1977). Bipolar adjectives are arranged with some number of dots between them, and the subject is asked to complete the task by marking between the two extremes the point that he or she feels is appropriate. The present study uses adverbs with adjectival bipolar opposites as rating scales in order to make the task more similar to the way people generally talk about others. People rarely talk about someone as being "7 on ambition" outside the laboratory, although use of the words "very" or "extremely" with adjectives is quite common. Empirical examination of the difference between rating scales with adverbial phrases and those with only adjectival points has yet to be conducted and is essential for determining the more accurate type of scale.

Each subject was asked to rate one stimulus person, either a traditional (male in male occupation, female in female occupation) or a nontraditional (male in female occupation, female in male occupation) character on the following six personality traits: ambitious, effective, emotional, intelligent, responsible, and traditional. These adjectives were chosen from studies of sex-role stereotyping (Broverman et al., 1972) and occupational stereotyping (O'Dowd & Beardslee, 1967) to be different from each other and to distinguish males from females.

Several considerations dictated the choice of occupations for this study. First was their usual characterization by subjects in several studies (Albrecht, 1976; Shinar, 1975) as male (lawyer, doctor) and female (secretary, nurse). Examination of the 1970 United States Bureau of the Census report revealed that there were an approximately equal percentage of people in the gender nontraditional roles represented by these four occupations. Another requirement was flexibility in which gender filled the role, In other words, it had to be relatively easy for subjects to imagine a female or male in the gender nontraditional occupation. Within the occupational sets (legal: lawyer/secretary; medical: doctor/nurse), people performed similar functions and worked in similar work environments. Although the traditionally female occupations in both pairs were the subordinate ones, this was more a reflection of the actual position of women in the workplace than of any shortcoming in the choice of occupations. The chosen occupations also needed to be familiar to high school students.

Male and Female Careers 727

The purpose of the present study was the investigation of persons as a function of gender and occupation. The study employed a 2 (Gender of Subject: male, female) × 2 (Gender of Character: male, David or Paul; female, Susan or Paula) × 2 (Gender of Occupation: male, doctor or lawyer; female, nurse or secretary) × 2 (Occupational Set: 1, doctor/nurse; 2 lawyer/secretary) × 2 (Name Set: 1, Paul/Paula; 2, David/Susan) factorial between-subjects design. The last two factors were considered to be solely methodological and not part of the experiment's core design. They were therefore used only in the preliminary sta- tistical analyses to determine the form of later analyses. Because name set was found to be a nonsignificant factor, subjects who rated Paula and Susan were combined, as were those who rated Paul and David. Occupational set was found to be a significant overall factor and it was therefore decided to separate the doctor/nurse occupational set from the lawyer/secretary occupational set, although a one-factor analysis of variance was computed within each trait to compare the four occupations.

METHOD

Subjects and Experimenters

The subjects were 77 male and 80 female students from grades 11 and 12 who attended a suburban high school and were enrolled in the psychology and sociology classes there. The population of this high school was primarily White and middle-class and the majority of students were college bound. Within gender of subject groups, students were randomly assigned to conditions. Two females, the author and an undergraduate research assistant, gave out the testing materials, read the instructions, and collected the completed sheets. Seventy-nine subjects, 39 males and 40 females, rated characters in the doctor/nurse occupational set; and 78 subjects, 38 males and 40 females, rated characters in the lawyer/ secretary occupational set.

Setting, Materials, and Procedure

Subjects were tested in groups of approximately 30 during the regular meeting time of the psychology or sociology class in which they were enrolled. Each participant was given a questionnaire booklet consisting of six sheets of paper. The character's name and occupation appeared at the top of each sheet with 12 verbal categories in the center. These categories were created by pairing an adjective and its bipolar opposite (expressed as the negative form of the ad-

728 Lifschitz

jec t ive) wi th six adverbs for each o f the six persona l i ty t rai ts l isted above. An

example of the ra t ing sheet for the adject ive " t r a d i t i o n a l " appears as Table I.

In o rder to prevent subjects f rom guessing the i n t en t o f the s tudy, t r ad i t iona l

was always the last t rai t for all subjects . The o the r five t rai ts were a r ranged in

two r a n d o m orders. The e x p e r i m e n t e r read the fol lowing s t a t e m e n t to the subjects :

Adverbs are often used when we describe people. For example, we can say someone is very smart or extremely nice. In the envelope I gave you, you will find six sheets of paper with different adjectives on each. The adverbs are the same on all the sheets. At the top of each sheet is the name of a person and some other infor- mation about him or her. When I tell you to begin, I 'd like you to go through the sheets in the set and choose the description from among the 12 provided that you feel best describes this person. Put a mark on the line next to the phrase that you feel best describes the person written on the upper right-hand part of each sheet. Everyone will have different opinions and I'm only interested in whatyou think so please don't look at anyone else's paper. If you are unsure what a certain adjective means, guess its meaning. When you have finished describing the person, please put all the materials in the envelope provided and I will collect them.

R E S U L T S A N D D I S C U S S I O N

The n u m b e r s 1-12 were arb i t ra r i ly chosen for ass ignment to the 12 verbal

categories o f each a t t r i bu te . People r anked high on an a t t r i bu t e were assigned

low n u m b e r s and people r anked tow were assigned high number s , so tha t 1 cor-

r e sponded to ex t r eme ly (adject ive) and 12 co r r e sponded to ex t r eme ly (nonad-

Table I. Sample Sheet from Questionnaire

extremely traditional very traditional quite traditional moderately traditional somewhat traditional

_ _ only slightly traditional only slightly nontraditional somewhat nontraditional

_ _ moderately nontraditional _ _ quite nontraditional

very nontraditional _ _ extremely nontraditional

Susan the doctor


jective). For an extended discussion of the alternate scoring method used on these data, see Lifschitz (1979).

The data were analyzed according to the StatisticalPackagefor the Social Sciences manual (Nie, Hull, Jenkins, Steinbrenner, & Bent, 1975) according to the least-squares estimation appropriate when unequal cell frequencies are ob- tained in an analysis of variance. Two three-factor analyses of variance were performed within each personality trait, one for subjects who rated doctor/nurse and the other for subjects who rated lawyer/secretary. Gender of Subject, Gen- der of Character, and Gender of Occupation are the relevant factors to be discussed separately for each trait. Means are given in Table 1I.

Traditional

For doctor/nurse, the analysis showed a gender of character by gender of occupation interaction, F(1, 71) = 31.195, p < .001. Subjects rated male nurses the leas( traditional and female nurses the most traditional. Male characters tended to be rated as less traditional than female characters, F(1, 71) = 3.021, p = .O87.

Table II. Cell Means for Trait Ratings

Doctor Nurse Lawyer Secretary

Trait being rated Male Female Male Female Male Female Male Female

Traditional Male subjects 4.9 7.8 8.4 4.8 5.3 7.6 9.6 6.2 Female subjects 4.4 5.9 8.6 3.7 5.2 4.9 7.6 4.7

Ambitious Male subjects 3.7 3.0 4.7 4.1 3.5 4.7 5.8 5.3 Female subjects 2.8 2.2 4.2 4.4 2.9 2.4 5.0 5.8

Intelligent Male subjects 3.8 2.6 4.7 4.0 3.4 4.6 4.5 4.1 Female subjects 2.3 2.3 3.5 3.6 2.5 3.0 4.2 4.9

Emotional Male subjects 6.1 4.0 5.7 4.9 6.6 6.1 5.3 4.2 Female subjects 5.2 4.9 5.3 5.1 7.1 5.3 5.8 3.7

Responsible Male subjects 4.1 2.9 4.7 2.1 3.3 4.9 3.8 4.6 Female subjects 2.2 1.6 2.9 2.7 2.9 2.8 2.6 2.5

Effective Male subjects 4.2 4.3 7.0 4.0 3.9 5.6 4.3 3.6 Female subjects 3.3 2.5 3.3 3.7 3.4 3.6 3.7 4.4

730 Lifschitz

For lawyer/secretary, there was a gender of character by gender of occupation interaction, F(1, 71) = 10.158, p = .002. Male secretaries were rated the least traditional and male lawyers were rated the most traditional. There was a main effect for gender of subject, F(1 ,71) = 5.918, p = .018, with male subjects rating characters as less traditional overall than female subjects. There was a main effect for gender of occupation, F(1, 71) = 4.524, p = .037, with lawyers rated as more traditional overall than secretaries.

A one-way analysis of variance performed on the four occupations showed 11o differences in traditionality among these groups.

As had been predicted, there was a gender of character by gender of occupation interaction in both occupational sets, with males in female occupations rated the least traditional. These data supported studies by Suchner and More (1975) and Touhey (1974) as well as prevalent stereotypes about males and females in nontraditional occupations. The two significant main effects can be explained by this two-factor interaction. Males in female occupations are still relatively uncommon, while females in male occupations are more common and possibly more accepted. Male occupations typically have more prestige than female occupations and the fact that male traits were found to be more positively valued than female traits (Broverman et al., 1972) may be a partial explanation for this phenomenon.

Ambitious

For doctor/nurse, there was a significant main effect for gender of occupation, F(1, 71) = 10.941, p = .001, with doctors rated more ambitious than nurses.

For lawyer/secretary, there was a significant main effect for gender of occupation, F(1, 71) = 15.558, p < .001, with lawyers rated more ambitious than secretaries.

A one-way analysis of variance performed on the four occupations showed a significant effect, F(3, 153) = 11.073, p < .001. A subsequent Newman-Keuls analysis showed doctor more ambitious than secretary, q(4, 1 5 3 ) = 103.75, p < .01; doctor more ambitious than nurse, q(3, 153) = 56.25, p < .05; lawyer more ambitious than secretary, q(3, 153) = 84.0 p < .01; and nurse more ambitious than secretary, q(3, 153) = 47.5, p < .05.

There was a significant main effect for gender of occupation both within and between occupational sets, with secretary rated as less ambitious than any of the other three occupations. Secretary was the only occupation that required no college education. It was clear from these data that ambition was a characteristic associated with occupations and not directly with gender. An employed woman was evidently as ambitious as an employed man in the same occupation, although when employment status was not specified, women were rated as less


ambitious than men (Williams & Best, 1977; Broverman et al., 1972). The underlying assumption is therefore that women differ among themselves in ambition, with the more ambitious women entering certain occupations. The same is true for men. Inasmuch as ambition is a characteristic of occupations, which are filled by people, ambition is indirectly a characteristic of people. But because there are few women in the occupations rated high in ambition, the implication is that there are in fact few ambitious women.

Intelligent

For doctor/nurse, there was a significant main effect for gender of subject, F(1, 71) = 4.753, p = .033. Females rated characters as more intelligent overall. There was a significant main effect for gender of occupation, F(1 ,71) = 9.668, p = .003, with doctors rated more intelligent than nurses.

For lawyer/secretary, there was a significant main effect for gender of occupation, F(1, 70) -- 8.873, p = .004, with lawyers rated more intelligent than secretaries. There was a marginally significant gender of subject by gender of occupation interaction, F ( 1 , 7 0 ) = 3.824, p = .055. Female subjects rated lawyers the most intelligent and secretaries the least intelligent.

A one-way analysis of variance performed on the four occupations showed a significant difference, F(3, 153) = 7.292, p < .001. A subsequent Newman- Keuls analysis revealed that doctors were rated more intelligent than secretaries, q(4, 153) = 67.25, p < .01; doctors were rated more intelligent than nurses, q(3, 153) = 46.25, p < .01; and lawyers were rated more intelligent than secretaries, q(3, 153) = 36.86, p < .05.

As with ambitious, there was a significant main effect for gender of occupation both within and between occupational sets, with secretary rated as less intelligent than any of the other three occupations. The present finding of the overall superiority of doctors was also found by O'Dowd and Beardslee (1967). These subjects' low opinion of secretaries may explain the presence of a gender of subject by gender of occupation interaction only in the lawyer/secretary occupational set. Female subjects rated lawyers as the most intelligent and secretaries as the least intelligent. Inasmuch as most of these students were college bound, their view of nontraditional high-prestige occupations such as law would increase in value, while occupations they would traditionally be expected to enter such as secretarial work would decrease in value.

Although males and females did not differ on ratings of intelligence, male occupations were rated as higher in intelligence than female occupations. Intelli- gence and gender were therefore indirectly connected. The underlying assumption, as with ambition, is that both inter- and intragender differences in intelligence coexist with occupational differences on this trait. In other words, a person, whether male or female, must attain a particular level of intelligence to

732 Lifschitz

enter certain occupations. However, the subtle implication is that few women have or believe they have the intelligence required to enter certain occupations.

Emotional

For doctor/nurse, there were no significant main effects or interactions. For lawyer/secretary, there was a significant main effect for gender of

character, F ( 1 , 6 8 ) = 6.144, p = .016, with females rated more emotional than males. There was a significant main effect for gender of occuaption, F(1, 68) = 7.108, p = .010, with secretaries rated more emotional than lawyers.

A one-way analysis of variance performed on the four occupations revealed a significant difference, F(3, 150) = 2.905, p = .0368. A subsequent Newman- Keuls analysis showed that secretary was more emotional than lawyer, q(4, 150) = 61.75, p < .05, and nurse was more emotional than lawyer, q(2, 150) = 44.25, p < .05.

Main effects for gender of character and gender of occupation appeared only in the lawyer/secretary occupational set, with females rated more emotional than males and secretaries rated more emotional than lawyers. The absence of a similar effect for the doctor/nurse occupational set may be explained by look- ing at the occupations involved. As members of helping professions, doctors and nurses are expected to be warm and understanding, regardless of gender. Yet they should also remain calm during crises and not be too emotional. These expectations were reflected in subjects' ratings, which were in the middle of the distribution of possible scores on emotion. Lawyers, however, may be seen as cold and calculating, while secretaries appear friendly and helpful. Apparently, emotionality was a trait more attributable to occupational roles than to persons, implying that males and females need not fall into stereotyped patterns. These high school students, possibly influenced by the feminist movement and popular trends, seemed to view emotionality as a trait equally characteristic of males and females, except where associated with particular occupational roles.

Responsible

For doctor/nurse, there was a significant main effect for gender of subject, F(1, 71) --- 7.436, p = .008. Female subjects rated characters as more responsible overall than male subjects did. There was a significant main effect for gender of character, F(1, 71) = 8.140, p = .006, with males rated more responsible than females.

For lawyer/secretary, there was a significant main effect for gender of subject, F(1, 70) = 14.739, p < .001. Female subjects rated characters as more responsible than male subjects did.


A one-way analysis of variance performed on the four occupations revealed no significant difference among the groups.

High school students, who have not yet entered the world of work, did not perceive differences in responsibility among working adults. Female subjects rated all characters, regardless of gender and occupation, as significantly more responsible than male subjects did. Perhaps these females were more naive and/ or polite than the males in attributing equally high levels of responsibility to all working adults.

Effective

For doctor/nurse, there was a significant main effect for gender of subject, F(1, 71) = 10.783, p = .002. Female subjects rated characters as more effective than their male counterparts did. There was a significant gender of subject by gender of character by gender of occupation interaction, F(1, 71) = 4.618, p -- .035. Males rated male nurses as the least effective, and females rated female doctors as the most effective.

There were no significant main effects or interactions in the lawyer/secretary occupational set.

A one-way analysis of variance performed on the four occupations revealed no significant difference among the groups.

One possible explanation of the inconsistent findings for ratings of effec- tiveness may be subjects' confusion about what "effective" meant. The author intended some measure of job competence by this term, but perhaps subjects defined it differently.

CONCLUSIONS

The overall pattern of responses indicated that where stereotypes appeared, they were based on occupation and not strictly on gender. Gender of occupation appeared as a significant factor (either main effect or interaction) in 8 of the 12 analyses conducted on the data; gender of character, in only 4 of the 12. While these occupational stereotypes are based at least partially on differen- tial expectations for males and females, the actual gender stereotypes seem less muted and more obvious. Perhaps the liberal or relatively sophisticated nature of the sample of subjects can explain this phenomenon. These middle-class, mostly White high school students probably would not want to admit obvious stereotyping of males and females in this age of the feminist movement on a test ad- ministered by a female psychologist from the university or they might be genu- inely affected by this movement. Occupational stereotypes are less blatant and

734 Lifschitz

therefore more readily expressed by this group. Even though the study was presented as an investigation of adverbs to disguise the true aim of investigating stereotypes, informal observation at the time the questionnaire was given revealed that some of the subjects recognized the actual intent. Subjects may have been reluctant to express their opinions honestly.

The results of the present study support those of a similar study by Shinar (1978), in which college students rated stimulus persons differing in gender and occupational role on 20 bipolar scales. Shinar found that of the three indepen- dent variables she used, gender of occupation had the greatest impact on ratings, with gender of character and gender of subject following as second and third in importance. The interaction of gender of occupation and gender of character (i.e., whether the person was presented in a sex-appropriate or sex-inappropriate occupation) also influenced subjects' ratings. Although this interaction did not appear as consistently in the study reported here as in Shinar's, careful examination of the two revealed that the data were parallel for traits which appeared in both studies. Different subject populations and methodologies may explain any disparities.

Following the implications of these data and especially the ratings on intelligence and ambition to their conclusion, we are confronted with an interest- ing logical argument. People believe that some professions require more of some positive traits than others, and that people in these professions have these qualities. Because few women are in these careers, the implication is that few women have these desired qualities. Yet subjects, at least those represented in this study, admit that women can possess these qualities. For example, although few women are apparently ambitious enough to become doctors, female doctors are seen as ambitious. Changing attitudes about sex roles may have influenced stereotypes about occupational roles, but this finding may also reflect the traditional way people have viewed women in nontraditional occupations, that is, as super- women who are more "masculine" than most males.

Interpretations of the data may be furthered by examining the specific adjectives and occupations included for study. Differences in ratings between occupational sets for traditional, emotional, and responsible indicate that the connotative meanings of these traits vary with the particular occupation and/or gender rated. An emotional doctor may be quite different from an emotional lawyer, for example. As discussed above, secretary seemed to be in a category separate from the others on some of the traits.

This study raised questions about the origin and nature of stereotypes in general, and about sex-role and occupational stereotypes in particular. The present author, in recognizing the complexities of the situation, believes that essen- tially a "chicken and egg" problem exists. Whether stereotypes are the result of gender-based occupational limitations or their cause is an unresolved issue. Stereotypes are both cause and effect, with changes in one stereotype system in- fluencing and being influenced by others,


R E F E R E N C E S

Albrecht, S. Social class and sex-stereotyping of occupations. Journal of Vocational Behav- ior, 1976, 9, 321-328.

Broverman, I., Vogel, S., Broverman, D., Clarkson, F., & Rosenkrantz, P. Sex-role stereotypes: A current appraisal. Journal of Social Issues, 1972, 28, 59-78.

Lifschitz, S. The interface of sex-role and occupational stereotypes in high school students. Unpublished doctoral dissertation, State University of New York at Stony Brook, 1979.

Nie, N., Hull, C., Jenkins, J., Steinbrenner, K., & Bent, D. Statisticalpackagefor the social sciences. New York: McGraw-Hill, 1975.

O'Dowd, D., & Beardslee, D. Development and consistency of student images of occupations (Monograph of the Department of Health, Education and Welfare). Washington, D.C. : U.S. Government Printing Office, 1967.

Rosenkrantz, P., Vogel, S., Bee, H., Broverman, I., & Broverman, D. Sex-role stereotypes and self-concepts in college students. Journal of Consulting and Clinical Psychology, 1968,32, 287-295.

Shinar, E. Sexual stereotypes in occupations. Journal of Vocational Behavior, 1975, 7, 99-111.

Shinar, E. Person perception as a function of occupation and sex. Sex Roles, 1978, 4, 679-693.

Suchner, R., & More, E. Stereotypes of males and females in two occupations. Journal of Voeational Behavior, 1975, 6, 1-8.

Touhey, J. Effects of additional women professionals on ratings of occupational prestige and desirability. Journal of Personality and Social Psychology, 1974, 29, 86-89.

Williams, J., Bennett, S., & Best, D. Awareness and expression of sex steteotypes in young children. Developmental Psychology, 1975, 11, 635-642.

Williams, J., & Best, D. Sex stereotypes and trait favorability on the Adjective Check List. Educational and Psychological Measurement, 1977, 37, 101-110.

male and female careers: sex-role and occupational stereotypes among high school students

Documents