The effect of life domains on girls' possible selves.

Curry, Carol ; Trew, Karen ; Turner, Irene 等

One concern that seems to be resistant to satisfactory explanation concerns girls and careers; specifically, that females are under-represented in higher status, higher earning occupations, especially in the scientific and technology fields. Researchers have proposed explanations for this gender difference from three main perspectives. Cultural and socialization perspectives suggest that parental influence and sex-role stereotyping conspire to turn girls away from the "masculine" subjects of mathematics, physics, and chemistry (Kelly, 1982; McEwen & Curry, 1987; Pedersen, Elmore, & Bleyer, 1986). Biological perspectives suggest that girls' lack of success with higher level mathematics can be partially explained by differences in the male/female brain which account for boys' superiority on spatial-visual tasks (Maccoby & Jacklin, 1974; Connor & Serbin, 1980), while other explanations have focused on such factors as girls' fear of success, lack of confidence, self-efficacy, or competitiveness (Horner, 1972; Hackett & Betz, 1989).

The ages of sixteen to eighteen are crucial in the lives of girls and boys who are dealing with physical and psychological changes related to puberty. Students who continue with full-time education must strive for autonomy while remaining under the influence of their most significant socializers--parents and teachers--and at the same time, must make decisions about future plans concerning careers. It is clear that the many factors which have been found to influence girls' and boys' choices of subjects and careers are interrelated, and psychologists have turned to model building to explain the complexity of interweaving variables.

One model which relates specifically to career choice is that of Dick & Rallis (1991) in which students make their career choices on the basis of their beliefs about themselves, their abilities, and the relative values of different careers. Students' self-concept and career values are influenced by perceptions of their aptitudes, the cultural milieu, socializers, and past experience. Career values, in terms of interests and abilities, salary expectations, cost, and length of training affect career choice.

When this model was tested, the effect of socializers on career choice was found to be "subtle yet extremely powerful." Even young women with exceptional high school mathematics and science preparation had such different career plans from boys with similar attainment, with few girls planning careers in engineering and science, that the impact of socializers was considered to be felt quite early. Differences were also found for career values, with men considering salary as much more important in career choice than did women who considered a genuine interest in career as important. One problem with this model is that while a student's choice of career is considered to be heavily influenced by past and present experiences, the student's future plans are narrowly limited to career choice as if no other future considerations impinged on these decisions. The model is in accord with the "social mold" explanation which sees the student as a product of socialization and where motivations are not considered central to decisions.

The Role of "Possible Selves"

Motivation is the central concern of the model proposed by Markus and her colleagues (Markus & Nurius, 1986; Oyserman & Markus, 1990) which describes how a person's motivations influence thought, feelings, and actions. They propose the concept of "possible selves," which are the images or conceptions of what a person might become, would like to become, or is afraid of becoming. Possible selves have a very concrete impact on how actions are initiated and structured, both to realize possible selves or to prevent negative possible selves. Possible selves cause the person to focus on the activities necessary to achieve the desired goal, and to persevere in pursuing the goal. A given possible self will have maximal motivational effectiveness when it is balanced by a countervailing possible self in the same domain. For example, a positive possible self of "me getting a job" when matched with a negative possible self of "me being employed" may provide the spur to compete with a short-term possible self "me watching TV" instead of "me studying." Oyserman & Markus (1990) in a study on adolescent delinquency, found that although there were many similarities between "hoped-for" possible selves of delinquent and nondelinquent adolescents, a lack of balance between expected possible selves and feared possible selves distinguished the two groups. The nondelinquent group had more balance between expected and feared possible selves, resulting in more motivational control over their behavior in a particular domain. Those without such a balance had fewer motivational resources.

Taking a slightly different perspective, for a group of adolescents who are predominantly middle-class, well educated, and studying advanced level subjects, an alternative model is proposed. Possible selves may be balanced by "most probable" and "most promising" selves rather than the more dramatic hoped-for and feared possible selves of the delinquents. Possible selves still form the main reservoir and contain all the possible selves a person may have, but these are tempered and held in balance by cognitions about the degree of inevitability affecting "most probable" selves on the one hand and "most promising" selves on the other. Most probable selves would be influenced by the cultural milieu, socializers, and past experiences, and the perception and interpretation of these as explained by the Dick and Rallis model. They may also be influenced by "ought" selves (Higgins, 1987) which can be imprisoning selves, including those which may be imposed on the adolescent by well-meaning significant others. For example, they could be governed by cognitions about the realities of the employment market, "I would really like to be an actor, but knowing the figures for unemployment, I will probably take a secure job if I can get it." The future orientation or planning aspect which deals most effectively with the inevitability factor is most promising selves, which would be those that a person considered worth striving to obtain.

Self-Concepts and Life Domains

Although men and women do not differ on the level of overall self-esteem, Josephs, Markus, and Tafarodi (1992) suggest that the basis of their self-esteem may differ. Women are more likely to have a socio-centric or connected self schema in which relationships with other people are crucial and are included within the self. Men are relatively more likely to have an individualistic separate schema with "other" distinct from the self. These differences are thought to arise from different sources--from boys' early experience of having to differentiate themselves from their mothers, while girls experience similarity and continuance with their mothers (Chodorow, 1978); from an imbalance in social power, with women having to be attuned to dominant others who control their fate (Miller, 1986); or a difference in social roles assigned to men and women from birth which gives rise to different concerns and commitments which are connecting and relating for women and separating and individuating for men (Eagly, 1987). Therefore, when adolescents are considering their future selves, girls and boys have different cognitions about most probable and most promising selves which may affect their future career aspirations.

Research has consistently shown that boys and girls differ in their pattern of career choice which still reflects traditional occupational choices. In attempting to investigate why, for example, few girls want to enter the object-oriented field of engineering or why few boys want to enter the person-oriented field of nursing, many factors have been investigated, but one of the most contentious motivations that have been considered in relation to gender differences in employment is that of work commitment. Hakim (1991) who has researched occupational segregation asked why, when women are to be found in the lowest status and lowest paid jobs, they were disproportionately satisfied with their jobs. After re-examining some of the most substantial research concerning women and employment in recent years, Hakim concludes that women's work commitment was found to be markedly lower than men's and a powerful predictor of women's work decisions and job choices. Work commitment is typically measured by the answer to the question of whether the respondent would engage in paid work if the financial necessity was removed. Hakim concludes that the majority of women aim for a homemaker career in which paid work is of secondary or peripheral importance and they receive strong support from their husbands for this strategy. For example, she quotes the 1980 Women and Employment Survey which found that "paid work was 'rarely' a central life interest for women; for 'most' women work took a secondary role to family concerns, and had to be accommodated to and balanced with domestic priorities." Career women, on the other hand, were in the minority--those who were committed to work as a central life goal, achieving higher level jobs and earnings, and whose work commitments were similar to those of working men.

The existence of these two discrete groups also helps explain the persistence of job segregation, in that women tend to be satisfied with low-status, low-paid jobs because such jobs tend to offer easy access to the work place, part-time hours, and social factors such as friendly people. The sex differential in work orientations, work commitment, and job satisfaction disappeared only in higher grade, male-dominated, professional and managerial occupations which normally require a major investment in skills, experience, and full-time work. Hakim's analysis might suggest that, as with other psychological models, womens' motivations are being measured against a male model of the world and found wanting, but she argues that women should be regarded as self-determining actors who make choices about their life styles and may indeed make these choices as young women.

The choices young women make may concern a wider range of life domains than do those of boys. Adolescent girls have been found to be more mature than boys and may therefore be more concerned with their future at an earlier age. Cohn (1991) conducted a meta-analysis of sex differences in personality development in sixty studies containing 9,000 subjects, all of which reported findings using Loevinger's (1976) system of ego development. It was found that girls achieved developmental milestones earlier than did boys, and that their greater maturity was not an artifact of superior verbal abilities. These sex differences were moderately large and stable among adolescents. The girls were less egocentric, less concrete in their thinking, and showed greater movement toward self-awareness and reflection. In behavioral terms, Abrams (1989) found a two-year lag between males and females in the shift toward heterosexual relationships. While girls were dating older boys by the ages of 15 or 16, boys were not catching up until 17 to 19.

Girls are clearly aware of the "dual responsibility" of having a career and maintaining the home and family (Curry & McEwen, 1989). There is also evidence that teenage girls are still being socialized into accepting the main burden for household responsibilities by doing much more housework than boys (Kelly, 1982), even when they are students preparing for higher education and future careers (McEwen, Curry, & Watson, 1986); Post-Kammer and Smith (1986) also suggest that women still persist in choosing careers which are more easily interrupted, less demanding, and which require fewer educational prerequisites than do careers chosen by males. While the role of women with regard to family and career has changed over the years, Burnhill and McPherson (1984) exploring the expectations of able Scottish school-leavers in 1971 and 1981 found that women in 1981 "more closely resembled men in the clarity, quantity, and quality of their occupational ambitions," but that "educationally able young women in general are just as uncertain today about the difficulties and successes women may have in combining a familial with an occupational career, as were their counterparts a decade earlier."

How does all this relate to the fact that girls are under-represented in mathematics and science at advanced levels, and what light can this kind of analysis shed on subject choice and career plans of highly educated "A" level pupils? In examining the factors which may influence boys and girls in their subject choices we investigated students' aptitudes, the cultural milieu, the effect of socializers, and past experiences on student's subject choice--all of which have an influence but fail to explain satisfactorily why so many able girls shy away from science and technological fields. The Dick and Rallis model takes us so far, but it is suggested that considerations about careers are inextricably linked with cognitions and motivations about other life domains for girls more so than boys. Since feelings of connectedness and relatedness (Chodorow, 1978) are part of girls' self-concepts, other life domains such as marriage and family must be factors in girls' future plans and may influence more immediate career planning. In terms of possible selves, since a major task of adolescence is to create the self one is going to become, boys are more likely to have a balance among their possible selves in terms of future careers with hope for a successful career balanced by fears of failing and being unemployed. It is also more likely that since girls lack such a straightforward balance, hope for a successful career may be complicated by probable and "ought" selves concerning future roles as primary caregivers for young children, while for boys the role of "good father" does not need to unbalance or interfere with hope for successful careers but is indeed incorporated into the ideal of "good provider."

Hakim suggests that we can identify "careerist," "adaptive," and "home-centered" work orientations among both men and women. Following from this, our hypothesis would be that careerist girls with work as a central life domain should be more confident about nontraditional masculine-style careers, and be less concerned with future family commitments. Adaptive and home-centered girls may already be thinking in terms of several life domains and be less committed to a central life domain of work. We tested this explanatory framework of work orientation, and using the concept of "possible selves," examined our data from a sixth-form study of pupils taking advanced-level subjects in grammar schools within Northern Ireland.

METHOD

Design of the study. This study forms part of the LAMBDA PROJECT (the Longitudinal Assessment of Mathematical Beliefs, Development and Attainment) (Turner et al., 1989) which has been investigating a sample of pupils from age 7-8. The sample at age 13-14 contained the longitudinal sample plus a cross-sectional sample of their fellow pupils (N = 1,500). Those who remained at school into sixth-form (age 16-17) form the sample for this study (N = 520, 240 girls and 280 boys), and at the time of contact were embarking on a two-year advanced level course.

Quantitative data: The questionnaire. Pupils completed a comprehensive questionnaire in school, supervised by the researchers. The questionnaire focused on mathematics and gathered information on such topics as educational and home background, advanced subject choices, future career and other plans, in addition to a battery of psychological measures.

Defining work orientation. As these pupils still have several years of schooling before they enter the labor market, our measurement of future work orientation was necessarily rather crude. Nevertheless we attempted to measure one aspect of work orientation by categorizing responses to the following question: "What do you think will be the likely pattern of your working life?" Those who chose working "full-time for most of my married life" were categorized as careerist. Those who chose "full-time from time to time" or "part-time from time to time" were adaptive. Those who choose "part-time for most of the time" or "do not intend to work for more than a few years at most" were considered home-centered.

Qualitative information: Group discussions. A series of group discussion were held (average size 8 pupils) each lasting approximately one hour. Groups were defined by sex and whether the pupils were taking mathematics at an advanced level. Discussions on topics such as mathematics and future careers were recorded for subsequent analysis.

RESULTS

Work orientation. When the responses to the question on work orientation were examined, it was found that 71% of all the pupils were defined as careerist, 21% as adaptive, and 8% as home-centered. As Table 1 shows, however, gender differences in the distributions of career type were large and obvious. While 86% of boys are careerist and thought they would be working full-time for most of their married lives, just over half (54%) of girls thought this was the likely pattern of future employment for them. More girls were adaptive (30% compared to 12% for boys) and 16% of girls were categorized as home-centered as compared to 2% of the boys (||Chi~.sup.2~ = 70.8, df = 2, p |is less than~ .0001). Table 1: Career Type by sex Girls Boys % % Careerist 53.8 86.1 Adaptive 30.4 12.1 Home-centred 15.8 1.8 Total numbers 240 280 Significance: chi-square = 70.8, d.f. = 2, p |is less than~ .001

While it is possible that pupils' thoughts about full-time and part-time work could be influenced by current and projected unemployment figures, because these pupils are among the elite in terms of education, we do not think this would be a serious concern. Secondly, although the question was framed within the context of married life, because less than 4% of pupils said they had no wish to get married or have children, these responses represent most pupils. Because of the small number of pupils who were in the home-centered category (girls, n = 38; boys, n = 5), pupils were divided into two categories: careerist or noncareerist--which comprises the adaptive and home-centered categories.

Support for differences in the careerists' and noncareerists' thinking about various life domains is shown in Table 2. Using a four-point scale from 1 (indicating "a great deal") to 4 (indicating "hardly at all"), Table 2 gives the mean responses where lower means indicate greater time spent thinking about each issue. Careerist girls show a pattern of responses that is similar to that of boys on questions about time spent thinking about the age at which to marry, and the age at which to start a family. The noncareerist girls spent significantly more time thinking about age to marry (p |is less than~ .05) and age to start a family (p |is less than~ .01) as compared to the careerist girls. Even at this early age, there is some evidence that girls have different cognitions about their probable selves in relation to marriage, family, and career. Since the boys were predominantly careerist, the remainder of the paper focuses on the girls in order to examine whether careerist and noncareerist categories provide useful insights into the future career planning of academically able girls. Table 2: Means of time spent thinking about age to marry and age to start a family GIRLS BOYS Careerist Non-careerist Careerist Non-careerist Age to marry 2.9 2.7(*) 3.0 3.0 Age to start family 3.2 2.9(**) 3.3 3.2 Significance levels, t-test * = p |is less than~ .05, ** = p |is less than~ .01 Note: lower means = greater time spent thinking about the issue.

Attainment and subject choice. Are careerist girls different from non-careerist girls in terms of attainment and subject choice? Results from a national examination (General Certificate of Secondary Education) taken at age 15-16 and covering a wide range of curriculum subjects were examined. Grade points were calculated by awarding 5 points to an A grade to 1 point for E, and zero for the failing grades for all subjects taken. Careerists had a higher mean of 39 points as compared to the noncareerists mean of 36 points, though this difference did not reach significance level.

In terms of A-level subject choice (advanced level examinations taken at the end of the sixth-form at age 17-18), as Table 3 shows, careerists were significantly (||Chi~.sup.2~ = 10.3, df = 3, p |is less than~ .05) more likely to be taking science only or science-dominant subjects (40%) as compared to noncareerist (22%), while noncareerists were more likely to be taking arts or art-dominant subjects (61%) compared to careerists (49%). More of the noncareerists (16%) had a balance of subjects as compared to the careerists (11%).

Careerists were more confident about their academic abilities, expecting to achieve higher grade points in their future advanced-level examinations. When expected A-level grade points were calculated, careerists expected a mean of 10.6 A-level points as compared to 9.6 for the noncareerists (p |is less than~ .05).

Parental influences. In terms of parental influence, the effects of mothers' working status on daughters' future plans for working were examined. Burnhill and McPherson (1984) found that for a Scottish sample of school-leavers, the girls responses to their expected pattern of working life mirrored their mothers' experience. For this sample the pattern is repeated since more of the careerists had mothers who were in full-time work (29%) than did the noncareerists (23%), while more of the noncareerists had mothers who were at home full-time (29%) as compared to the noncareerists (16%). Table 3: Advanced level subject choices by career type for girls Careerists Non-Careerists % % Science only/science dominant 40.3 22.5 Arts only/arts dominant 48.9 61.2 None dominant 10.9 16.2 Total numbers 129 111 Significance: chi-square = 10.4, d.f. = 4, p |is less than~ = .05

With regard to parental identification, more girls overall identified with their fathers (61%) as compared to their mothers (36%). However, more of the noncareerists identified with their mothers (43%) than did the careerists (31%), and more of the careerists identified with their fathers (65%) as compared to the noncareerists (56%).

Attitudes toward careers, marriage, and family. To examine the interplay of career, marriage, and family intentions, we used nine items to measure attitudes, applying a five-point scale anchored at "strongly agree" and "strongly disagree." The pupils' instructions defined "career" as follows: "This question asks you to look to the future and think about your life as a young adult when you leave school. We would like your views on careers, marriage, and family life. By "career" we mean putting continuous energy over a long period into achieving some success in an occupation rather than simply taking short-term occasional jobs. There are no right or wrong answers; we would like you to give us your frank opinions."

While attitudes toward issues concerning women and work in general do not necessarily match womens' personal behavior, as Table 4 shows, responses to the attitude items show that the careerist and noncareerist categories do seem to be tapping some aspects of work orientation rather than simply measuring responses to likely patterns of working. In items concerned with "equality for women," all the girls rejected notions about men being the main breadwinner, women being unable to successfully have a career and a family, or that men were reluctant to marry career women, though they were unsure whether marriage and children would have to fit into their career plans. However, chi square showed significant differences in five of the items more related to career behavior. More careerists agreed that it was important for them to establish a career as soon as possible (p |is less than~ .001), and that both parents' incomes were necessary to support a family (p |is less than~ .001). More of the noncareerists thought that young children needed a parent at home full-time (p |is less than~ .01), agreed that they would be happy to stay at home if their spouse earned more than they did (p |is less than~ .01), and thought that a woman can get as much satisfaction from a family as from a career (p |is less than~ .05). Table 4: Percentages "strongly agreeing" or "agreeing" with each item by career type Careerists Non-careerists % % It is very important for me to try to establish myself in a career as soon as possible 72 59(***) Marriage and children would have to fit into my career plans 57 50 I think when children are young they should have a parent at home full-time to look after them 48 63(**) It is important for the man to be the main breadwinner in the family 5 7(**) If my spouse earned more than I did I would be quite happy to be the one to stay at home to look after the children 24 42(**) Nowadays, both parents need to work to support the family 57 33(***) Women can get as much satisfaction from having a family as from having a career 52 67(*) It is rare for a woman to combine a career and a family and make a success of both 10 9 Men are not keen to marry "career" women 18 9 Significance levels for chi-square * = p |is less than~ .05, ** = p |is less than~ .01, *** = p |is less than~ .001

Career self-efficacy. Hackett and Betz (1985) have extended Bandura's (1977) theory of self-efficacy to career behavior and have investigated the importance of the construct of self-efficacy in understanding the differences in the educational and career choices of women and men. They argue that, largely as a result of socialization experiences, women exhibit lower expectations than do men for many achievement-related behaviors and thus fail to fully realize their capabilities and talents in career pursuits. Further, if individuals lack self-efficacy in career-related behavior, career planning, such as enrolling in mathematics and science classes, is less likely to occur, and therefore high self-efficacy is theoretically related to making effective career choices.

Hackett and Betz (1985) suggest that cognitions concerning competence may be a much more critical factor than measured abilities in both educational and career choice processes, particularly for women pursuing nontraditional careers. Self-efficacy can be affected by factors other than past performance such as sex-role socialization messages from society, family, classmates, and teachers, discouraging women from achieving in traditionally masculine domains such as mathematics (Boswell, 1985; Eccles & Jacobs, 1986). Lent, Brown, and Larkin (1987) compared self-efficacy with several other models such as interest congruence (Holland, 1985) and consequence thinking (Janis & Mann, 1977) in explaining career relevant behavior in students considering science and engineering fields, and found that self-efficacy was the most useful of the three in predicting grades and persistence in technical/scientific degree subjects.

To measure self-efficacy in relation to careers, pupils were asked to rate a number of careers along several dimensions, two of which are reported here. Pupils were asked to rate how interested they would be in the occupation, from 1 (very interested), 2 (undecided), to 3 (not interested) and, secondly, how confident they would be in obtaining the educational qualification necessary for each job if they wanted it--whether or not they had previously said they were interested from 1 (confident), 2 (unsure), 3 (not confident). The sixteen occupations are among those which have been used in several research projects (Hackett & Betz, 1985; Post-Kammer & Smith, (1986) and reflect male and female traditional occupations.

Since the girls in the study were all taking advanced-level courses, it is not surprising that over three-quarters of them were planning to enter higher education. Slightly more careerists (81%) were intending to enter higher education than were noncareerists (71%), while more of the noncareerists (13%) were planning to start a career immediately after leaving school than were the careerists (9%), though the differences were not significant. Although few of the occupations aroused interest among the girls, certain differences are evident. In five of the eleven nontraditional occupations, there were significant differences, with careerists more interested than noncareerists. The noncareerists showed a higher level of interest in all the traditional occupations than did the careerists.

TABULAR DATA OMITTED

In terms of self-efficacy, however, the careerists showed a higher level of confidence in obtaining the educational qualifications necessary for all the nontraditional occupations, a difference which was significant for all but one of the occupations. Careerists also showed the same levels of confidence about the nontraditional occupations as did the noncareerists. In other words, in terms of self-efficacy, the careerists showed higher levels of confidence than did the noncareerists over the whole range of nontraditional occupations associated with higher status, higher income careers.

Psychological profiles of careerists and noncareerists. Up to this point our attempts to measure work orientation in sixth-form girls have produced two groups who are different in their patterns of attainment, academic expectations, subject choice, and interest and confidence in nontraditional occupations. The question remains: Would other psychological measures show any differences between the careerists and noncareerists? As Table 6 shows, the profile of the careerists as less stereotypically "feminine" than the noncareerists is emphasized by the finding that they are significantly less person oriented and more object oriented than the noncareerists as measured by 10 items from the Smithers and Collings (1984) scale, and they also had a higher masculine score on the Bem sex-role orientation measure (Bem, 1974).

On factors associated with mathematical attainment, careerists had a higher mathematical competence score (on 5 items from Harter's Adolescent Profile, 1986) and were more autonomous and independent of their teacher's judgment (on 5 items from Harter's Intrinsic/Extrinsic Orientation in the Classroom, 1980). Their general self-efficacy was also higher (on 8 items from Sherer & Maddux, 1982). In terms of femininity and self-worth, there was no difference between the two groups on the Bem femininity scores or on 5 items from Harter's measure of global self-worth, indicating that self-esteem may be derived from different sources for different groups. Table 6: Psychological measures for careerists and non-careerists Careerists Non-careerists Score Score Masculine Sex-role Orientation (Bem, 1974) (higher scores=higher masculinity) 4.8 4.5(***) Person-object Orientation (Smithers & Collings, 1978) (higher scores=higher object orientation) 23.5 21.9(*) Scholastic Competence (adapted from Harter, 1986) (higher scores=higher mathematical competence) 15.5 14.0(***) Independent Judgement (from Harter, 1980) (higher scores=higher independence) 16.6 15.6(**) General Self-efficacy (from Sherer & Maddux, 1982) (higher scores=higher self-efficacy) 30.7 29.5(*) Global Self-worth (from Harter, 1986) (higher scores=higher self-worth) 16.9 17.1 Feminine Sex-role Orientation (Bem, 1974) (higher scores=higher femininity) 4.9 5.0 Significance levels for t-tests: * = p |is less than~ .05, ** = p |is less than~ .01, *** = p |is less than~ .001

Gender differences in possible selves. This paper has presented some tentative suggestions that girls who are studying advanced-level subjects at school and considering their future plans may have a different set of cognitions and a wider range of life domains to consider than do similarly qualified boys. It is suggested that girls' self-concepts are more likely to encompass several life domains, and the importance girls attach to the different domains may influence their future career planning. It also may help explain why some girls are reluctant to enter high-profile, "masculine" careers which may be less accommodating to those who may wish to consider part-time work while their children are young, and may be careers where those taking "career breaks" would be most likely to suffer in terms of promotion.

Most of the girls, as would be expected from this sample of A-level students, are defined as careerists, but a substantial minority may not have a career as a central organizing life domain. There is some evidence for Hakim's suggestion that life choices may be made relatively early, and that girls, just like adult women, are not a homogeneous group in terms of career planning. Whether the differences found between the careerists and noncareerists can be accounted for by future intentions (most promising selves) or are simply a reaction to the realities of the life they envisage (most probable selves), they must be a factor in girls' self-concepts about their possible selves. Some of the quotes from a series of group discussions with these sixth-form girls illustrate differences in their thinking about career and families. From a group of the best mathematicians in a girls' school, the following are some noncareerist views:

"My mother was always there so I feel if I had children I would give up whatever career I had."

"You'd look at it |your career~ in a different way, I think, whenever you came back, because firstly you were able to dedicate yourself to a career but whenever you come back with a child you would have to slow down a bit."

The careerists thought it was "an unfair world" where working mothers were concerned:

"I don't think there should be any problem. I think it's up to the government to provide a refresher course or something like that. Husbands, they stay at work; they don't have to miss out. I think that's really unfair."

"They should make more facilities for mothers with young children because it's not fair to disadvantage them ... if there's a man in the same position as a woman, if she's going to have a baby it's only fair that they provide facilities for the baby so she can go back to work and I think there are very, very few of them and there should be more. I don't think it's fair at all."

The girls had no doubts about the impact that career breaks can have on careers:

"My mother works in the hospital and because she took time off to be with my sister and me she is still now in the same position as she was. She went back in the same position and she's now seeing people that she trained up, higher than her because she took time out."

"That's the same as my mother. She was head of a department; she's a teacher, and she took time off to have me and now she's just an ordinary teacher again. I think she's a bit annoyed ... to see people a lot younger than her and less qualified than her in higher positions when she took time off to have me which meant she had to go back as an ordinary teacher."

CONCLUSIONS

Judging from the differences found in attitudes, interests, attainment, subject choice, parental identification, influence of mother as a role model, and psychological profiles, it would appear that some girls may be making early choices about the kind of life they would like to live--choices that encompass career, marriage, and family. While boys appear to be a more homogeneous group concerning career orientation, girls appear to be heterogeneous, with differing orientations toward full-time and part-time work and differing attitudes toward the role of parenthood. While a longitudinal study would be needed to examine how early work orientation translates into future plans for career, marriage, and family, it is suggested that even at this relatively early age in terms of planning future lives as independent adults, these factors may play an important part in the considerations. If horizontal and vertical job segregation is resistant to change because many women (with the support of their partners) actively choose marriage and family as a central life domain, with employment a secondary domain as Hakim suggests, it is nevertheless surprising to find evidence that some girls taking an advanced-level academic course and preparing for higher education and future careers may already be thinking in this way.

REFERENCES

Abrams, D. (1989). Differential association: Social developments in gender identity and intergroup relations during adolescence. In S. Skevington, & D. Baker (Eds.), The social identity of women. London: Sage Publications.

Bandura, A. (19877). Self-efficacy: Towards a unifying theory of behavioral change. Psychological Review, 84, 191-125.

Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.

Boswell, S. L. (1985). The influence of sex-role stereotyping on women's attitudes and achievement in mathematics. In S. Chipman, L. Brush, & D. Wilson (Eds.), Women and mathematics: Balancing the equation. Hillsdale, NJ: LEA.

Burnhill, P., & McPherson, A. (1984). Careers and gender: The expectations of able Scottish school-leavers in 1971 and 1981. In S. Acker, & D. W. Piper (Eds.), Is higher education fair to women? London: Nelson.

Chodorow, N. (1978). The reproduction of mothering: Psychoanalysis and the sociology of gender. Berkley: University of California Press.

Cohn, L. D. (1991). Sex differences in the course of personality development: A meta-analysis. Psychological Bulletin, 109(2), 252-266.

Connor, J., & Serbin, L. (1980). Mathematics, visual spatial ability and sex roles. Report to the UN National Institute of Education.

Curry, C. A., & McEwen, A. (1989). The "Wendy House" syndrome: A teenage version. Research in Education, 41, 53-60.

Dick, T. P., & Rallis, S. F. (1991). Factors and influences on high school students' career choices. Journal for Research in Mathematics Education, 22(4), 281-292.

Eagly, A. (1987). Sex differences in social behavior: A social role interpretation. Hillsdale, NJ: Erlbaum.

Eccles, J. S., & Jacobs, J. E. (1986). Social forces shape math attitudes and performance. Signs: Journal of Women in Culture and Society, 11, 21.

Hackett, G., & Betz, N. E. (1985). A self-efficacy approach to the career development of women. Journal of Vocational Behavior, 18, 326-339.

Hackett, G., & Betz, N. E. (1989). An exploration of the mathematics self-efficacy/mathematics performance correspondence. Journal of Research in Mathematics Education, 20(3), 261-273.

Hakim, C. (1991). Grateful slaves and self-made women: Fact and fantasy in women's work orientations. European Sociological Review, 7(2), 101-121.

Harter, S. (1980). A scale of intrinsic versus extrinsic orientation in the classroom. Denver, CO: University of Denver.

Harter, S. (1986). Self-perception profile for adolescents. Denver, CO: University of Denver.

Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340.

Holland, J. L. (1985). Making vocational choices: A theory of careers (2nd ed). Englewood Cliffs, NJ: Prentice-Hall.

Horner, M. S. (1972). Towards an understanding of achievement-related conflicts in women. Journal of Social Issues, 28, 157-175.

Janis, I. L., & Mann, L. (1977). Decision making: A psychological analysis of conflict choice and commitment. New York: Free Press.

Josephs, R. A., Markus, H., & Tafarod, R. W. (1992). Gender differences in the source of self-esteem. Journal of Personality and Social Psychology, 63, 391-402.

Kelly, A. (1982). Gender roles at home and school. British Journal of Sociology of Education, 3(3), 281-295.

Lent, R. W., Brown, S. D., & Larkin, K. C. (1987). Comparison of three derived variables in predicting career and academic behavior: Self-efficacy, interest congruence and consequence thinking. Journal of Counseling Psychology, 34 293-298.

Loevinger, J. (1976). Ego development: Conceptions and theories. San Francisco: Jossey-Bass.

Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, CA: Stanford University Press.

McEwen, A., Curry, C. A., & Watson, J. (1986). Subject performances at A-level in Northern Ireland, European Journal of Science Education, 8(1), 39-50.

McEwen, A., & Curry, C. A. (1987). Girls' access to science. In R. D. Osborne, R. J. Cormack, & R. L. Miller (Eds.), Education and policy in Northern Ireland. Belfast Policy Research Institute.

Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954-969.

Miller, J. B. (1986). Toward a new psychology of women (2nd ed.). Boston: Beacon Press.

Oyserman, D., & Markus, H. (1990). Possible selves and delinquency. Journal of Personality and Social Psychology, 59(1), 112-125.

Pedersen, K., Elmore, P., & Bleyer, D. (1986). Parent attitudes and student career interests in junior high school. Journal for Research in Math Education 17(1), 49-59.

Post-Kammer, P., & Smith, P. L. (1986). Sex differences in math and science career self-efficacy among disadvantaged students. Journal of Vocational Behavior, 29, 89-101.

Sherer, M., & Maddux, J. E. (1982). The self-efficacy scale: Construction and validation. Psychological Reports, 51, 663-671.

Smithers, A., & Collings, J., (1984). Person orientation and science choice. European Journal of Science Education, 6(1), 55-65.

Turner, I.F., Trew, K., Hunter, J., McQuoid, J., & Mangan, C. (1989). Mathematics attainment and its concomitants. Unpublished report to Government of Northern Ireland, Department of Education.

Carol Curry, B.A., M.S., Research Fellow; Irene Turner, Ph.D., Senior Lecturer; Jennifer Turner, B.A., M.S., Research Fellow, The School of Psychology The Queen's University of Belfast.

Reprint requests to Karen Trew, Ph.D. School of Psychology, The Queen's University of Belfast, Belfast, BT7 INN Northern Ireland.