See all ›
1 CitationSee all ›
30 ReferencesNurse or Mechanic? Explaining Sex-Typed Occupational Aspirations amongst Children
Abstract
There is a high degree of sex-typing in young children's occupational aspirations and this has consequences for adult occupational segregation. Yet we still know surprisingly little about the mechanisms involved in the formation of sex-typical preferences and there is considerable theoretical controversy regarding the relative role of parental socialization and individual agency in this process. This study analyzes the determinants of sex-typed occupational aspirations amongst British children aged between 11 and 15. We develop a model of parental socialization and test for different channels and mechanisms involved in the transmission of sex-typical preferences. We also propose an innovative definition of personal agency that is anchored in observable psychological traits linked to self-direction. We find that parental influences on occupational preferences operate mainly through three distinctive channels: 1) the effect that parental socio-economic resources have on the scope of children's occupational aspirations, 2) children's imitation of parental occupations, and 3) children's learning of sex-typed roles via the observation of parental behavior. We also find a strong net effect of children's own psychological predispositions —self-esteem in particular— on the incidence of sex-typical occupational preferences. Yet large sex-differences in occupational aspirations remain unexplained.
Full-text (PDF)
Available from: Javier Polavieja, Jan 08, 2015Other full-text sources
© The Author 2014. Published by Oxford University Press on behalf of the
University of North Carolina at Chapel Hill. All rights reserved. For permissions,
please e-mail: journals.permissions@oup.com.
Social Forces 00(00) 1–31, Month 2014
doi: 10.1093/sf/sou051
Research for this paper has been supported by the European Centre for Analysis in the Social Sciences,
ECASS; the Spanish National Research Council, CSIC; IMDEA-Social Sciences Institute; and the
Institute for Economic Analysis, IAE-CSIC. Platt carried out part of this research during a Marie
Curie Intra-European Fellowship at the Institute for Economic Analysis, IAE-CSIC. She is grateful to
the European Commission for funding the fellowship. The authors also thank Michael Biggs, Stephen
Jenkins, and Tak Wing Chan for their helpful comments. The usual disclaimer applies. Address cor-
respondence to Javier G. Polavieja, University Carlos III de Madrid, Department of Social Sciences;
Calle Madrid 135, 28903 Getafe, Madrid, Spain; email: javier.polavieja@uc3m.es.
Nurse or Mechanic?
Nurse or Mechanic? The Role of Parental
Socialization and Children’s Personality in the
Formation of Sex-Typed Occupational Aspirations
Javier G. Polavieja, Universidad Carlos III de Madrid
Lucinda Platt, London School of Economics and Political Science
B
oys and girls with sex-typical aspirations are significantly more likely to end up
in sex-typical jobs as adults. Preference formation among children is therefore
relevant for subsequent occupational outcomes. This study investigates the role
of parental socialization and children’s agency in the formation of sex-typed occupa-
tional preferences using data for British children aged 11 to 15. We anchor agency
in observable psychological attributes associated with children’s capacity to act in
the face of constraints. We focus on two such attributes, motivation and self-esteem.
Our findings identify two main sources of parental influence: (1) parental sex-typical
behaviors, from which children learn which occupations are appropriate for each sex;
and (2) parental socio-economic resources, which affect children’s occupational ambi-
tion. We find, additionally, that girls with high motivation and both girls and boys with
high self-esteem are less likely to aspire to sex-typical occupations, net of parental
characteristics. Motivation and self-esteem help girls aim higher in the occupational
ladder, which automatically reduces their levels of sex-typicality. For boys, however,
self-esteem reduces sex-typicality at all levels of the aspired occupational distribu-
tion. This suggests that boys with high self-esteem are better equipped to contradict
the existing social norms regarding sex-typical behavior. Implications are discussed.
Introduction
Even today, most people work in jobs occupied largely by persons of their own
sex (see, e.g., Chang 2004; Tomaskovic-Devey et al. 2006). Although this is
true for both men and women, segregation is more acute for the latter, as they
Nurse or Mechanic? 1
Social Forces Advance Access published May 12, 2014
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
tend to concentrate in fewer occupations. Predominantly female occupations
offer lower wages and fewer opportunities for career advancement, and hence
segregation is often regarded as the main source of women’s labor-market dis-
advantage (see, e.g., Maume 1999; Tomaskovic-Devey 1993). It is therefore not
surprising that the study of gender segregation has for long been placed at the
center of gender stratification research.
Gender segregation in occupations is the result of the actions and interactions
of both firms and workers. Discrimination and social closure explanations focus
on the role that employers, managers, and male coworkers play in hindering
women’s access to particular jobs (Roscigno, Garcia, and Bobbitt-Zeher 2007).
However insightful, demand-side approaches cannot explain the existence of
significant sex differences in career preferences and occupational aspirations,
not only among adults, but also among young children who lack labor-market
experience (Harper and Haq 2001; Okamoto and England 1999).
Sociologists have long stressed the crucial role that socialization processes
play in the transmission of sex-specific norms, values, and aspirations leading
to segregated occupational outcomes (Hitlin 2006; Okamoto and England
1999). According to classical socialization approaches, early childhood expe-
riences would have a prime impact on the formation of gendered preferences,
leaving a long-lasting imprint on people’s lives. Gender socialization models
provide a supply-side alternative to human capital and sphere specializa-
tion models in economics (Polavieja 2009) as well as to socio-biological and
evolutionary explanations of gender-role differentiation (Kanazawa 2001;
Udry 2000).
The existing sociological literature on gender socialization suffers, how-
ever, from two important empirical limitations. First, research has been much
more concerned with establishing empirical associations, typically associations
between parents’ and children’s characteristics, than with explaining the mecha-
nisms whereby socialization influences operate (Reskin 2003). Consequently, we
still know little about the actual channels and processes involved in the inter-
generational transmission of sex-typed preferences. Second, empirical studies
often draw on adult samples to address socialization processes that are thought
to take place during childhood, which further complicates the identification
of transmission mechanisms. As a result of these caveats, socialization is still
largely a black box in gender stratification research.
Conventional socialization approaches have also been subjected to two
important theoretical criticisms. First, it has been argued that classical
approaches overemphasize the importance of early childhood experiences as
primary sources of socialization (see, e.g., Corsaro and Fingerson 2003; Elder
1994). New approaches in developmental psychology, social psychology, and
life-course research contend that socialization is a lifelong process and stress the
continuing socializing role of small-group interactions, contextual influences,
and peer effects (for a review, see Correll and Ridgeway 2003; Corsaro and
Fingerson 2003; Elder 1994). The degree to which early socialization experi-
ences have a lasting effect on subsequent adult behavior is today a contested
theoretical question, which is still open to empirical testing.
2 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
A second criticism of classical socialization approaches is that they leave very
little room for individual agency. It has been argued that socialization mod-
els typically portray actors as passive receptors of gender values and norms,
and assume that all individuals are equally malleable by social influences. This
leads to an over-socialized conception of human behavior (Gecas 2003; Hakim
1991; Hays 1994). In our view, this neglect of agency constitutes a fundamental
theoretical shortcoming of conventional approaches. Understanding what the
role of individual agency is and how it interacts with the social environment in
the formation of sex-typical preferences is not only a crucial question for the
development of gender socialization theory, but one that can shed light on the
structure/agency debate that runs so deep in the sociological discipline (Corsaro
2005, chapter 1). Yet such a task poses one fundamental methodological chal-
lenge; namely, how to measure human agency.
In much of the existing empirical literature, agency has been equated with
preference heterogeneity (see, e.g., Hakim 1991, 2000). Since individual pref-
erences are seldom observed, it is often assumed that agency is to some extent
represented by the amount of unexplained variance in empirical models (Hitlin
and Elder 2007). In other words, individual agency is typically not measured
but only inferred. This indirect approach carries with it the serious risk of over-
individualization; that is, magnifying individuals’ capacity to make indepen-
dent choices. In order to shed empirical light on the socialization versus agency
debate, it is therefore essential to find more direct ways of measuring the role of
individual agency in preference formation.
This paper investigates the degree of sex-typicality in the occupational aspira-
tions of British children under 16 and tests for different mechanisms involved in
the acquisition of sex-typical occupational preferences. We establish that early
occupational preferences have a real impact on occupational outcomes in adult
life. We then address the following research questions: First, we want to know
whether parental characteristics and parental behavior influence the degree of
sex-typing in children’s occupational aspirations, and if so, how. To this end,
we propose an eclectic theory of parental socialization that incorporates explicit
channels and mechanisms, which are empirically testable.
Second, we investigate the role of children’s agency in the formation of occu-
pational preferences. Hitlin and Elder (2007) argue that current sociological
treatments of agency are too abstract to offer guidance for empirical research
but can be illuminated by social psychology. They call for anchoring the “slip-
pery concept” of agency in measurable psychological attributes. We put their
recommendation into practice. We expect that children’s heterogeneity in occu-
pational preferences is associated with the distribution of certain psychological
characteristics in the population. We are interested, specifically, in those psycho-
logical attributes that can exert an influence on individuals’ capacity to act in
the face of constraints—and hence to resist socialization pressures. We focus on
two such attributes: motivation and self-esteem. We argue that if agency plays a
role in the formation of occupational preferences, we should find an association
between these personality attributes and the level of sex-typicality in children’s
occupational aspirations.
Nurse or Mechanic? 3
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
We test our model using information on parental, relational, and psycho-
logical variables for a representative sample of over 3,000 British children aged
11 to 15. This sample is drawn from waves 4 to 18 of the British Household
Panel Survey (1994–2008). Over one-third of the children in this sample can
be followed into their early occupational outcomes as adults. By investigating
early gender differences in occupational aspirations, our approach helps open
the black box of parental gender-role socialization, sheds light on the agency-
structure debate, and fills an important gap in the sociological literature on
gender segregation.
Theoretical Framework
Parental Socialization
Following Arnett (1995, 618), we can define socialization as “the process by
which people acquire the behavior and beliefs of the social world—that is, the
culture—in which they live.” The most important—but certainly not the only—
agent of primary socialization in gender roles is the family (Bandura 1977;
Cunningham 2001; Hitlin 2006; Okamoto and England 1999). But how can
families shape children’s occupational aspirations? Drawing on social stratifica-
tion, social learning, and developmental psychology, we identify two main chan-
nels of parental influence: (1) parental behavior in the economic and domestic
spheres; and (2) parental socio-economic resources.
Behavioral role-modeling: Occupational imitation and sex-role learning According
to role-model theories, children first learn about gender roles by observing and
emulating the behaviors of their parents (Bandura 1977). Empirical studies
have found a significant statistical association between the present behavior
of daughters and the past behavior of their mothers in areas such as family
formation, housework distribution, and female labor-market participation (see,
e.g., Cunningham 2001; van Putten, Dykstra, and Schippers 2008). This evidence
has been interpreted as proof of behavioral role-modeling. Yet it is still unclear
how role-modeling actually operates. This is partly due to the shortage of data
that can measure parental behavior contemporaneously with the formation of
children’s preferences.
We distinguish between two different forms of sex-role-modeling: simple
imitation and behavioral sex-role learning. Imitation is an essential compo-
nent in children’s observational learning based on live models (Bandura 1977).
Developmental psychologists have shown that pure imitation of same-sex par-
ents plays a crucial role in infants’ sex-role learning (Bussey and Bandura 1999).
The essential precondition for same-sex imitation is children’s identification with
their same-sex parent. Today there is a growing consensus among developmental
psychologists that same-sex identification is probably innate, as it requires some
form of preexisting gender identity (Martin, Ruble, and Szkrybalo 2002). Same-
sex identification with peers is also known to be strong among infants and pre-
adolescent children (Corsaro and Fingerson 2003, 143–44).
4 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
We propose to test for direct occupational imitation as one potential mecha-
nism of occupational socialization. Occupational imitation is expected to be
homo-lineal; that is, daughters are expected to aspire to their mother’s occu-
pation, while sons are expected to aspire to their father’s. Direct occupational
imitation will lead to sex-typed aspirations among daughters (sons) insofar
as their mothers (fathers) work in segregated occupations themselves (H1).
Occupational reproduction through imitation could therefore be the simplest
form of intergenerational transmission of sex-typed occupational aspirations.
Behavioral sex-role learning is the process by which children discover and
absorb what the prescribed behavior for their sex is by observing the actions of
their parents (see, e.g., Crouter, Manke, and McHale 1995). This learning pro-
cess is indeed more complex and cognitively demanding than simple imitation.
Children must first identify socially prescribed gender roles by examining the
behavior of their own parents and then form expectations about the costs and
benefits of deviating from sex-typical behavior. Parents can be active or passive
gender-role models for children. Active parents stimulate children’s compliance
with gender norms directly by using sanctions and rewards, which can be more
or less subtle (Bandura 1977). But parents can also enact gender roles insen-
tiently, simply because their sex-patterned behaviors embody the social structure.
Eagly (1987) builds on social-role theories to explain how sex-differentiated
behaviors are replicated and sustained within an unequally structured society.
The thrust of her argument is that sex-patterned behaviors provide crucial infor-
mation about the social structure because they encapsulate the social constraints
under which men and women carry out their lives (see also Eagly, Wood, and
Diekman 2000). By observing parental sex-typical behaviors in both the domes-
tic and the public spheres, children learn about the social distribution of oppor-
tunities between the sexes. This implies that parental sex-typical behaviors can
foster adaptive gendered processes of aspiration and sex-role assumptions even if
parents do not actively seek to transmit traditional gender norms. We thus expect
parents’ sex-typical behaviors to promote sex-typical occupational aspirations
among children (e.g., nurse for girls, mechanic for boys) even if such aspirations
do not entail copying the exact occupations of their same-sex parents (H2).
More precisely, we expect that girls (boys) whose mothers (fathers) are
employed in traditionally female (male) occupations develop more sex-typical
occupational aspirations than girls (boys) whose mothers (fathers) are employed
in less traditional jobs (H2a). Similarly, we expect that children living in house-
holds with a traditional distribution of housework—that is, where mothers
do more than fathers—(H2b) and children of mothers with low labor-market
attachment (H2c) develop more sex-typical occupational aspirations than chil-
dren living in households with less traditional arrangements.
Parental resources, occupational ambition, and sex-typing It is well known that
children’s educational and occupational attainment is highly dependent on
parental background (see, e.g., Gamoran 1996). Families with greater cultural
and economic resources tend to have higher attainment aspirations for their
offspring and to transmit these aspirations to children themselves. They are also
Nurse or Mechanic? 5
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
in a position to directly support their children’s occupational ambition through
increased opportunities and investment. We call this the ambition effect of
parental resources. This ambition effect, we believe, can have implications for
the degree of sex-typicality of girls’ occupational preferences but not necessarily
of boys’. This is because top-level occupations are traditionally male dominated.
Hence for girls, aiming high on the occupational ladder typically means aspiring
to occupations that are not female dominated. Yet boys have many sex-typical
occupations to choose from at both ends of the occupational distribution and
hence increasing occupational ambition has no obvious bearing on the degree
of sex-typicality of their aspirations. Parental resources affecting children’s
occupational ambition are therefore expected to affect the degree of sex-typing
in daughters’ occupational aspirations (H3a), while having a neutral effect on
sons’ (H3b).
The role of personality In recent years, research in economics and sociology has
paid increasing attention to certain psychological attributes that are shown to
be relevant to socio-economic success (see, e.g., Heckman, Stixrud, and Urzua
2006; Jackson 2006). In research practice, these attributes are often reduced
to composite indices that tap on the correlation between various measures of
personal drive, motivation, and self-esteem (Carneiro and Heckman 2005). In
competitive environments, such personality characteristics, often referred to
as noncognitive skills, are expected to exert a crucial influence on individuals’
attainment chances. The idea that personality attributes and dispositions might
critically influence goal-oriented behavior comes from social and developmental
psychology (Bandura 1997; Jacobs et al. 2002; Wigfield and Eccles 2000).
In this study, we focus on two psychological attributes, motivation and self-
esteem, which are relevant in influencing children’s capacity to act in the face
of constraints. While both motivation and self-esteem promote achievement-
oriented behavior, and can be regarded as partially overlapping, we expect self-
esteem to have the additional effect of enhancing children’s capacity to make
independent choices. This expectation follows directly from Bandura’s self-
efficacy theory (Bandura 1977, 1990), which sees individuals’ beliefs about their
own capabilities as the core psychological determinant of human agency, under-
stood as human’s capacity for action in the face of constraints (see also Gecas
2003; Hitlin and Elder 2007).
While Bandura (1990) argues that self-efficacy is different from self-esteem,
a number of authors have highlighted the intrinsic connection between the two,
via self-competence, a critical dimension of self-esteem.
1
Tafarodi and Milne
(2002), among others, argue that feelings of self-competence are so deeply inter-
twined with self-efficacy that the conceptual distinction between the two should
be relaxed in practice.
2
For them, self-efficacy and self-competence are but
two sides of the same cumulative process: that of exercising efficacious action.
Similarly, Gecas (2003, 371) argues that self-efficacy is a prime source of self-
esteem and hence considers both concepts as crucial components of agency.
3
We concur with these approaches and treat self-esteem as (indirectly) reflect-
ing people’s capacity for autonomous action. This interpretation is in line with
6 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
accumulated empirical evidence in psychology showing that people with high
self-esteem have greater initiative and hence greater capacity to deviate from the
group’s consensus (see Baumeister et al. 2003).
4
We thus posit that both motivation and self-esteem will reduce the sex-
typicality of occupational aspirations, but that it will do this through two
distinctive mechanisms: ambition and autonomy. Children with high levels of
motivation or self-esteem are expected to aim “higher” in the occupational
structure (Carneiro and Heckman 2005; Heckman, Stixrud, and Urzua 2006).
This ambition effect should reduce the level of sex-typicality in girls’ occu-
pational aspirations (H4a), but not in boys’, since top-level occupations are
typically male dominated (H4b). At the same time, children with high levels of
self-esteem are expected to be better equipped to make independent choices,
and hence to act against existing social norms, than their low-esteem counter-
parts. This autonomy effect should make children of both sexes more likely to
choose occupations that are outside the range of what is socially prescribed for
their sex (e.g., nurses for boys, mechanics for girls). We therefore expect more
autonomous children to be more likely to choose sex-atypical occupations
whatever their occupational ambition. However, since girls who select higher-
ranking occupations are necessarily less likely to select female-dominated occu-
pations, this autonomy effect might be indistinguishable from occupational
ambition for girls (H5a). It is only among boys that we will be able to identify
clearly whether self-esteem enables them to deviate from expected behaviors
independently of occupational ambition (H5b).
In sum, motivation (trait) is expected to reduce sex-typicality through greater
occupational ambition (mechanism), while self-esteem (trait) is expected to
reduce sex-typicality through both greater ambition and greater autonomy
(mechanisms). Given that patterns of occupational sex segregation differ by sex,
empirical predictions are sex specific: Ambition is expected to reduce the degree
of sex-typicality of girls’ occupational aspirations but not necessarily of boys’
(since top-level occupations are typically male dominated), while autonomy is
expected to reduce sex-typicality for both girls and boys. Finally, since in the
case of girls both autonomy and ambition mechanisms are expected to work
in the same direction, their empirical effects are likely to be confounded. This
implies that the effect of autonomy will most probably be identifiable only for
boys. Table 1 summarizes our hypotheses, including channels and mechanisms,
for both parental and personality effects.
Data and Methodology
Data Sources
British Household Panel Survey The British Household Panel Survey is a
longitudinal study of individuals who were living in private households in Great
Britain in 1991 (University of Essex 2010). The original sample comprised
around 5,500 households, with around 10,300 respondent adults. These original
sample members are followed over time and reinterviewed each year, along with
Nurse or Mechanic? 7
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Table 1. Parental Socialization and Personality Effects on Children’s Sex-Typed Occupational Aspirations: Channels, Mechanisms, and Hypotheses
Parental socialization Personality traits
Channels Parental occupation
Parental domestic
behavior
Parental socio-
economic status Motivation Self-esteem
Mechanisms -Imitation
-Sex-role learning
-Sex-role learning -Ambition -Ambition -Ambition
-Autonomy
Hypotheses -Homo-lineal
occupational imitation
transmits sex-
typicality (H1)
-Parents in
sex-segregated
occupations increase
children’s sex-
typicality homo-
lineally (H2a)
-Traditional
distribution of
housework increases
children’s sex-
typicality (H2b)
-Mothers with
low labor-market
attachment increase
children’s sex-
typicality (H2c)
-Parental SES
increases attainment
aspirations
Negative effect
on girls’ sex-
typicality (H3a)
Neutral effect
on boys’ sex-
typicality (H3b)
-Motivation
increases attainment
aspirations
Negative effect
on girls’ sex-
typicality (H4a)
Neutral effect
on boys’ sex-
typicality (H4b)
-Self-esteem increases attainment
aspirations
Negative effect on girls’ sex-
typicality (H4a)
Neutral effect on boys’ sex-
typicality (H4b)
-Self-esteem increases children’s
capacity to make independent
choices
Negative (but confounded)
effect on girls’ sex-typicality
(H5a)
Negative effect on boys’ sex-
typicality (H5b)
8 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
other members of their households aged 16 and over. Data are available for all
years up to 2008 (or wave 18).
In 1994, a self-completion questionnaire was introduced for children in the
panel aged 11 to 15. This questionnaire (the youth panel) has been administered
annually up to 2008. The youth panel provides the empirical backbone of this
study. We are able to link information from the youth panel to household and
individual adult respondent files in order to relate children’s and their parents’
responses to one another, to include family context, and to apply appropriate
weights. Having contemporaneous self-reported data from both parents and
children provides us with a distinctively rich resource of family information.
Given the longitudinal nature of the survey, we can also link children’s responses
in the youth panel to their post-16 outcomes as, at age 16, they become eligible
for the main adult interview.
Overall, just over 5,000 individual children were surveyed in the youth panel
over the 15 waves.
5
However, many of the questions, including those of particu-
lar interest to this study, are not asked in every sweep, meaning that some chil-
dren are missed altogether for some questions and others have varying numbers
of repeated observations on any particular measure. In order to ensure that an
appropriately complete array of variables is available for each child in our study,
and to exploit the value of panel data in providing repeat measures, we utilize
information across all the sweeps in which they were observed. This enables us
to provide rich information on their occupational aspirations and to construct
measures of their psychological characteristics based on repeat observations.
Around 3,700 boys and girls provided a valid response to an open-ended
question on occupational aspirations at some point. This question forms the
basis of our dependent variable (see below). The question was not asked in
waves 9, 10, or 11, so we do not have observations for those years. We utilize
the latest valid response. For nearly half of the children, this was at age 15. Since
different questions are asked in different years, answers to other variables may
have taken place at other ages.
For child-level independent variables, such as age, where possible we measure
them concurrently with the measure of occupational aspirations. Where they
occurred only in prior waves, we utilize the latest observation. However, for the
psychological variables, where we expect them to capture underlying, stable dis-
positions, such as with our measures of motivation and self-esteem, we exploit
the advantages of repeat measures in panel data by utilizing measures across
all observations on each child to construct a child-specific measure (see below).
By these means, we construct a cross-sectional data set, which accommodates
the distinctive structure of the study, but which utilizes as much information as
possible from across the observations. An illustration of this structure is given
in figure 1.
Information from co-resident parents of each child was matched into the
youth data using a similar approach. Allowing for missing data and questions
not asked of particular children or parents because of the question cycles, our
final analysis sample comprises 1,693 boys and 1,667 girls, which amounts to
91 percent of those for whom we have valid coded occupational aspirations.
Nurse or Mechanic? 9
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Figure 1. Example of data set structure
Child
ID
Original data Analysis sample
Waves
observed
Age Response on
occupational
choice
Response
on VarY
Response on
VarZ
(psychological
variable)
Age when
latest
occupation
measured
Last valid
response on
occupational
choice
Last
valid
response
on varY
Response on
varZ
(psychological
Variable)
1F 11 fireman Yes Important
1G 12 fireman Not asked Important
1H 13 Police officer Not asked Missing 13 Police officer Yes
1I 14 Not asked Not asked Not asked
1J 15 Not asked Not asked
2J 11 Not asked Not asked Not asked
2L 13 Actress Yes Not asked
2M 14 Actress Yes Not asked 14 Actress Yes Missing
3K 12 Nurse Not asked Important
3L 13 Nurse No
3M 14 Non -response No
3N 15 Teacher No 15 Teacher No
4Q 11 Air pilot Yes Not asked
4R 12 Air pilot Missing Important 12 12 Yes Important
Mean of valid
responses
Mean of valid
responses
Very
important
Very
important
Very
important
Very
important
Note :These cases are illustrative only and do not represent genuine respondents and their responses. The highlighted cells indicate the responses from
the original longitudinal sample of children 1–4 that are utilised in the analytic sample of unique observations per child.
10 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Measuring sex-typicality: The Labour Force Survey In order to measure the level
of sex-typicality in children’s favored occupations, we calculated segregation
measures using the UK Labour Force Survey (LFS). We used 28 pooled quarters
of the LFS, from the first quarter of 1994 (which corresponds to the start of the
BHPS youth panel) to the last quarter of 2000 (Office for National Statistics
1994–2000). This gives us a pooled nationally representative sample, with
current occupational information for 367,006 working-age adults across 371
occupations. From this, we calculated the average proportion of women/men
for each three-digit occupation
6
and then matched this information to children’s
identified job preferences as well as to each parent’s job.
7
We also matched the
proportion of women and men in a given occupation to the realized occupational
outcomes for those 567 girls and 620 boys from the BHPS youth panel who
were both interviewed as adults and employed at the time.
We also use the LFS to calculate the average wage for each three-digit occupa-
tion in the data set. This provides a measure of the relative position of respon-
dents’ aspired occupation in the overall occupational distribution and hence
accounts for the vertical dimension of occupational aspirations, which we use to
differentiate between the ambition and the autonomy effects of children’s moti-
vation and self-esteem (see below).
Variables
Outcome variable Children’s favored occupation was identified by an open
question of the form “What job would you like to do once you leave school
or finish your full-time education?” This was coded to three-digit SOC90
occupational codes. The proportion of women or men typically employed in
each of these occupational codes was calculated using the LFS, as explained
above, and matched to the occupational choice. While there was a degree
of clustering of children’s occupational choices, overall the 1,868 boys for
whom we have valid responses identified 122 occupations and the 1,880
girls selected 153 occupations.
8
The top 20 choices for each sex are listed in
table 2.
The average proportion of women in children’s aspired occupations was 42
percent (58 percent for girls and 23 percent for boys). The LFS adult population
experienced an average of 46 percent women across occupations (71 percent
for women, 25 percent for men). Real-life occupations are therefore somewhat
more segregated for women on average than aspired occupations are for girls.
Figure 2 shows the kernel densities for the proportion of men and women in
boys’ and girls’ aspired occupations.
We operationalized sex-typed occupational aspirations as those occupational
choices falling in the top 30 percent of the sex-specific occupational distribution
of women (for girls) or men (for boys). Thus, we constrained around 30 percent
of girls’ and boys’ occupational choices to be “sex-typical.” As figure 2 shows,
this corresponds roughly to the second peak in each of the bimodal distribu-
tions. We carried out robustness checks for alternative specifications, including
continuous and multinomial specifications. While our findings were robust to
Nurse or Mechanic? 11
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Table 2. Top 20 Preferred Occupations for Girls and Boys (Those Chosen by More Than 30), by Descending Order of Popularity, and Actual Jobs of
Mothers and Fathers by Prevalence
Girls Boys Mothers Fathers
Actors, stage managers, etc. Athletes, sports officials, etc. Sales assistants Drivers of road goods vehicles
Hairdressers Motor mechanics Cleaners, domestics Production, works managers
Primary and nursery
education teachers
Armed forces Care assistants and attendants Service industry managers, etc.
Solicitors Police officers Educational assistants Other managers and administrators
Vets Artists, graphic designers, etc. Nurses Metal work, maintenance fitters
Artists, graphic designers, etc. Computer analysts, programmers Clerks Carpenters and joiners
Nursery nurses Architects Account clerks, bookkeepers Storekeepers and warehousepersons
Beauticians Plumbers, heating engineers Other child-care occupations Gardeners, groundspersons
Nurses Aircraft flight deck officers Community and youth workers Marketing and sales managers
Authors, writers, journalists Actors, stage managers, etc. Service industry managers Motor mechanics, etc.
Police officers Carpenters and joiners Primary, nursery teachers Builders, building contractors
Travel and flight attendants Chefs, cooks Other secretarial personnel Cab drivers and chauffeurs
Medical practitioners Secondary education teachers Filing and record clerks Building/contract managers
Secondary education teachers Authors, writers, journalists Other financial, etc., managers Farm owners and managers, etc.
University teachers Medical practitioners Secondary education teachers Other construction trades
Other child-care occupations Solicitors Retail cash and checkout operators Electricians
Clothing designers Electricians Bar staff All other laborers
Biological scientists Builders, building contractors Receptionists Computer systems, etc., managers
Other health professionals Musicians Counter clerks and cashiers Police officers
Psychologists Chartered and certified accountants Catering assistants Plumbers, heating engineers
Source: British Household Panel Survey waves 4–18.
12 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Figure 2. Kernel density distributions of the proportion of own sex in aspired occupations
0
0.5
1
1.5
Density, boys
Density, girls
2
2.5
0 0.2 0.4 0.6 0.8 1
Proportion of men in aspired occupation
0
0
.5
1
1.5
2
0
0
.2
0
.4
0
.6
0
.8 1
Proportion of women in aspired occupation
Source: British Household Panel Survey waves 4–18 and UK Labour Force Survey pooled quarters 1994–2000.
Nurse or Mechanic? 13
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
these alternatives,
9
we selected the binary specification as most clearly evidenc-
ing influences on sex-typicality of children’s aspirations.
Parental variables Parental resources are measured by parental educational
attainment using a dominance approach, whereby we use whichever parent’s
education is higher. For children with an absent father, mother’s educational
attainment is used. Educational attainment is measured using a set of discrete
categories: university degree and above; A levels (typically obtained at age 18)
and above but less than university; O levels or CSEs (typically obtained at age
16); and less than this or none. We employ a dummy for absent father to reflect
the diminution of parental resources that this implies.
Occupational imitation is measured straightforwardly using dummies to
reflect whether there is a direct match between boys’ (girls’) aspired occupation
and the current occupation of their father (mother).
We include several measures for parental behavior. The level of sex segre-
gation of both mother’s and father’s (last or actual) occupation is measured
using a three-category variable that differentiates among sex-atypical, inter-
mediate, and sex-typical occupations. The respective cutoff points for these
categories were the top 30 percent, the middle 40 percent, and the bottom 30
percent of the sex-specific distributions. Alternative specifications of segrega-
tion measures were explored but did not alter the overall findings. Behavior
within the home is captured by two measures. First, we compute the difference
between the number of hours of housework contributed by mothers and the
number of hours contributed by fathers (self-reported). This measure captures
variation in housework requirements and preferences at the household level.
Thus, a positive value indicates additional hours carried out by the mother (a
gender-typical distribution), and a negative value indicates additional hours
carried out by the father. Finally, we compute a variable that measures mothers’
labor-market attachment by calculating the average incidence over the waves
at which they were observed as out of the labor force through looking after
home and family.
Children’s psychological attributes The construction of psychological attributes
follows the general principle of maximizing information and reducing error
by using repeat measurements across waves and multiple items when available
for each construct. Children’s motivation is measured as school motivation
using responses to the following two questions: “How much does it mean
to you to do well at school?” and “How important do you think it is for
you to get your GCSE/Standard Grades exams?”
10
Each of these questions
had four possible options, ranging from “a great deal/very important” to
“very little/not very important.” For each item, we first generate a (reversed)
within-person average score over waves. The first item is asked every wave
in the youth panel except the first, while the second appears only in the last
seven waves. For respondents with observations on both items, we summed
the two and then generated a z-score (i.e., normalized to a mean of zero and
a standard deviation of one), while for respondents who had no observations
14 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
for the latter item, we created a z-score based on repeat responses to the
former item alone.
Self-esteem is measured using the Rosenberg self-esteem items included in the
BHPS. Out of the original complete 10-question battery, four questions were
asked at every wave of the youth panel, and a further two were asked from wave
9 onward.
11
We thus use a restricted selection of items, which is not unusual
in the specialized literature (e.g., Tafarodi and Swann 1995; see also Robins,
Hendin, and Trzesniewski 2001). While there is ongoing debate as to whether
the Rosenberg scale represents one or two constructs (Gray-Little, Williams,
and Hancock 1997; Schmitt and Allik 2005; Tafarodi and Swann 1995), we
assume for the purposes of this paper a global construct in line with the original
claims. Our rationale is both conceptual and practical. Because feelings of self-
competence obviously increase people’s feelings of self-worth, and because high
self-worth enhances the capacity for efficacious action (Tafarodi and Swann
1995), both dimensions are highly correlated in practice.
12
This provides the
main justification for treating both dimensions as part of the same construct, as
in Rosenberg’s original formulation (1965).
We utilize repeat measurements across waves in order to reduce measurement
error and take the average score for each item. For those who were observed at
least once from wave 9 onward, we then use the sum score of the averages for
all six items. For those who were only ever asked four questions (less than a
quarter of the total sample), we calculate the sum score across those four items.
In order to render the measurement equivalent between the two, we standardize
the scores for both the four-question and six-question responses.
For our operationalization of agency (motivation score and self-esteem score),
it was important to be able to distinguish the net effect of the two concepts, rec-
ognizing that as measured they showed some overlap (correlation of 0.2). To
better interpret the specific contribution of each of these partially collinear vari-
ables, we orthogonalized the standardized scores (Sribney 1995). This resulted
in only minimal adjustment but enabled us to distinguish the contribution of
self-esteem from that of motivation.
The final test for our theoretical predictions regarding the role of personality
consists of differentiating empirically between the ambition and the autonomy
effects of children’s motivation and self-esteem. This we do by introducing the
log average wages of each aspired occupation as a measure of occupational
hierarchy in a second model.
13
The logic of this test is simple: If the effect of any
given personality indicator on the degree of sex-typicality in children’s occupa-
tional preferences disappears after controlling for the average wages in aspired
occupations, we should conclude that all the impact of this estimated psycho-
logical attribute is due to its effect on children’s occupational ambition, in that
they aspire to higher-paid jobs. If, on the other hand, the effect persists, we
should conclude that this given attribute decreases sex-typicality at all levels
of the aspired occupational distribution, which would be consistent with an
autonomy effect. Log average wages should also mediate any effect of parental
resources on girls’ sex-typing since, as discussed, having higher-educated par-
ents will tend to increase children’s occupational ambition, resulting in a higher
Nurse or Mechanic? 15
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
propensity to select top-rank occupations, which are typically male dominated
(and hence sex-atypical for girls).
In addition to parental and psychological variables, models include age of
child, which is the age at which their job aspirations were last measured with
a valid response, number of siblings, since the number of siblings is associated
with levels of parental investment and hence occupational ambition, and dum-
mies for the presence of older male or female siblings, who might be expected
to influence occupational aspirations. We also include dummies for the wave at
which the child is observed. The descriptive statistics for all variables used in the
analyses, separated by sex, can be found in table 3.
The Model
We estimate a series of logit regression models, fitted to our nationally repre-
sentative sample of British children aged 11 to 15. We explore those groups of
factors hypothesized as shaping children’s chances of aspiring to a sex-typical
occupation. We conduct separate analyses for boys and girls. Thus, equation
(1.1) is the model for girls and (1.2) is the model for boys. Y
g
is the probability of
a girl’s aspired occupation being sex-typical; that is, with the density of women
in the occupation being in the top 30 percent of girls’ choices; and Y
b
is the
probability of a boy’s aspired occupation falling in those for which the density
of men is in the top 30 percent. V is a vector of independent family and child
characteristics, and W is a set of controls for the wave.
log
11
Y
Y
g
g
1 −
=+ ++=…
{}
=…
{}
αβ γ
vv w w
V Wev1Vw 111;;
(1.1)
log
22
Y
Y
b
b
1 −
=+ ++=…
{}
=…
{}
αβ γ
vv w w
V Wev1Vw 111;;
(1.2)
Vector V includes variables for family structure, parental resources, parental
behaviors, and children’s psychological attributes. As explained above, our final
test consists of introducing the log average wage in children’s aspired occupa-
tions, as a means to control for the vertical dimension of children’s occupational
preferences, and thereby to disentangle the role of ambition from that of auton-
omy (equations [2.1] and [2.2]).
log
11
Y
Y
g
g
1 −
=+ ++
()
+=…
{}
=…
αβ γ
vv w w
VWLn occwageev1V
w1
;;
111
{}
(2.1)
16 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Table 3. Descriptive Statistics of Estimation Sample
Girls Boys
mean sd min max count mean sd min max count
Aspire sex-typical occupation 0.29 0.45 0 1 1,667 0.29 0.45 0 1 1,693
Wave 12.55 4.51 5 18 1,667 12.37 4.56 5 18 1,693
Age 14.01 1.39 11 16 1,667 13.89 1.40 11 16 1,693
Absent father 0.23 0.42 0 1 1,667 0.22 0.42 0 1 1,693
Total number of siblings 1.03 0.98 0 7 1,667 1.10 0.98 0 7 1,693
Older brother 0.28 0.45 0 1 1,667 0.28 0.45 0 1 1,693
Older sister 0.24 0.43 0 1 1,667 0.24 0.43 0 1 1,693
Parental educational level 2.57 0.96 0 4 1,667 2.63 0.95 0 4 1,693
Homo-lineal occupational match 0.01 0.11 0 1 1,667 0.02 0.16 0 1 1,693
Mother occupational segregation 1.99 0.75 1 3 1,667 2.04 0.75 1 3 1,693
Father occupational segregation 2.00 0.77 1 3 1,667 2.01 0.77 1 3 1,693
Prop time mother housewife 0.23
0.30 0 1 1,667 0.24 0.30 0 1 1,693
Housework inequality 12.73 13.32 −34 89 1,667 12.77 13.27 −53 70 1,693
School motivation −0.15 0.86 −1.61 4.38 1,667 0.05 0.99 −1.92 6.21 1,693
Self esteem −0.17 0.97 −3.64 2.12 1,667 0.25 0.91 −4.39 2.12 1,693
Log wage in aspired occupation 2.22 0.47 1.30 3.17 1,667 2.31 0.36 1.30 3.17 1,693
Achieved sex-typical occupation
as adult
0.27 0.45 0 1 567 0.29 0.46 0 1 621
Note: Sample statistics are weighted; Ns are unweighted.
Nurse or Mechanic? 17
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
log
22
Y
Y
b
b
1 −
=+ ++
()
+=…
{}
=
αβ γ
vv w w
V WLnoccwageev1V
w1
;;
……
{}
11
(2.2)
To ascertain if there is a relationship between sex-typicality in aspired and
achieved occupations, we initially estimate logit models of the probability of
being in a sex-typed occupation in adulthood, for both young men and young
women, exploring the impact of childhood sex-typical aspirations, and incorpo-
rating basic controls. We also estimate linear regression models for the impact
of sex-typical preferences in childhood on achieved wages in early adulthood.
In all analyses, the data were weighted, using the cross-sectional weight for
the wave at which children’s occupational aspirations were measured, to account
for nonresponse in that wave and of the differential weightings for the additional
samples. Additionally, standard errors were adjusted for repeat observations in
households; that is, where there was more than one child respondent per family,
though in practice there were few such cases in our sample.
As a robustness test, we also fitted ordinary least squares regression models
on the proportion female/male in children’s aspired occupations and multino-
mial regression models on a three-category variable including high, medium,
and low values of sex-typicality. Results were robust to these alternative specifi-
cations, which are available on request.
Results
Our analysis first addresses the crucial question of whether children’s early sex-
typical occupational aspirations have real consequences in adult life. Table 4
illustrates the influence of early preferences on two important outcomes—
occupational segregation and wages—for the approximately 1,200 children
who can be followed into their early adult occupational outcomes. Even though
by this stage only a mere six percent of them work in the exact occupation that
they aspired to as children, we find that both girls and boys with sex-typed
preferences are significantly more likely to end up in sex-segregated occupa-
tions as adults. The effects are strong. This suggests that early sex-typical pref-
erences, as manifested during childhood, express some underlying tendency for
gender-typical behavior that leaves visible traces later in life. We also find that
women who aspired to sex-typical occupations as children are likely to have
lower wages in their first significant jobs. This is hardly surprising, given that
female-dominated occupations are known to pay lower average wages. In sum,
early preference formation appears to have identifiable consequences for gender
segregation and consequently for women’s earnings in adult life. Given the rel-
evance of children’s occupational aspirations for their subsequent occupational
outcomes, we turn to the factors that help us understand how these early pref-
erences are formed.
Table 5 below shows the results of estimating the logit models for the
probability that children aspire to sex-typical occupations. Model 1 includes
18 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
children’s socio-demographic characteristics, parental influences, and children’s
personality attributes, as well as a range of controls for family structure. Model
2 adds the average log wages of children’s aspired occupation as a means to
control for occupational hierarchy. Hence, model 1 corresponds with equations
1.1 and 1.2 and model 2 corresponds with equations 2.1 and 2.2 above. Several
important findings are worth reporting.
First, our models provide some evidence that homo-lineal occupational imita-
tion might be a transmitter of sex-typicality, but effects are significant only for
boys. Boys whose occupational aspirations exactly match the occupations of
their fathers are more sex-typical than boys who do not imitate. Yet it must be
noted that only a mere three percent of boys in our sample actually have aspira-
tions that match their fathers’ occupations. This means that occupational imita-
tion plays only a very minor role in the formation of sex-typical occupational
preferences.
Our findings regarding behavioral sex-role learning are largely consistent
with our expectations. In particular, we find very strong evidence of behavioral
sex-role learning from parental occupations for both boys and girls. Model
1 shows that daughters whose mothers are (or were last) employed in sex-
atypical occupations (i.e., male dominated) have a lower probability of aspiring
to sex-typical occupations than observationally equivalent girls whose mothers
are employed in integrated and sex-typical occupations. This effect is net of
direct occupational imitation and parental education. If this correlation were
Table 4. Children’s Gendered Aspirations and Adult Outcomes among Currently Employed
Young Adults, Logistic and Linear Regression Estimates
Probability of sex-typed
adult occupation Adult wage
Girls Boys Girls Boys
Adult age −0.0593 −0.146*** 0.0495*** 0.0319***
(0.0442) (0.0360) (0.00715) (0.00499)
Gender-typical aspirations 0.841*** 1.288*** −0.116** −0.0376
(0.213) (0.207) (0.0383) (0.0283)
Matched child-adult
occupation
1.739*** 0.469 0.0387 0.145**
(0.424) (0.298) (0.0853) (0.0447)
Constant −1.035
†
−0.184 1.585*** 1.700***
(0.583) (0.522) (0.105) (0.0763)
Observations 567 621 567 620
Adjusted R
2
0.09 0.12 0.24 0.12
Note: Additional controls for wave not shown. Weighted estimates. Standard errors adjusted
for clustering in households. Standard errors in parentheses.
*** p
< 0.001 ** p < 0.01
†
p < 0.10
Nurse or Mechanic? 19
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
an artifact of the intergenerational transmission of occupational ambition from
mothers to daughters, it should disappear once we control for the occupa-
tional ranking of daughters’ aspirations. Yet model 2 shows that the associa-
tion between mothers’ and daughters’ sex-atypical preferences holds even after
controlling for the average wages of girls’ aspired occupations. Girls whose
mothers are employed in sex-atypical jobs are thus more likely to aspire to
sex-atypical occupations across the aspired occupational distribution (e.g., to
surgeons as much as to mechanics). Similarly, we find that boys whose fathers
are employed in typically masculine jobs are themselves more likely to aspire
to sex-typical occupations than boys whose fathers are employed in integrated
and female-dominated occupations. This effect for boys is also robust to direct
occupational imitation and wage controls (see model 2). We find no significant
effect of mothers’ occupational sex-typicality on their sons’ aspirations, nor do
we find any effect of fathers’ occupational sex-typicality on their daughters’.
The evidence is therefore highly consistent with a process of homo-lineal sex-
role learning from parental occupations (H2a), whereby girls learn to be sex-
atypical from their sex-atypical mothers while boys learn to be sex-typical from
their sex-typical fathers.
When looking within the household, we also find that a traditional distribu-
tion of housework tasks between spouses, revealed in a positive coefficient on
housework inequality, seems to reinforce children’s sex-typical occupational
aspirations, although in this case effects are observed only for boys. This is
an interesting finding, as it suggests that parental behavior in the domestic
sphere can have sex-role learning effects on children’s occupational prefer-
ences (H2b), although the effects seem sex specific. Finally, model 1 shows
that, net of other behavioral variables, having a mother with high domestic
(low labor-market) attachment has no significant impact on children’s occu-
pational preferences, which contradicts our expectations (H2c), although it
must be noted that the sign of the coefficient is in the expected (i.e., positive)
direction.
Consonant with our expectations, we are also able to identify an ambition
effect of parental education on the probability that daughters have sex-typical
occupational aspirations (H3a). Girls of parents with higher levels of education
have a significantly lower probability of aspiring to sex-typical occupations than
girls from low-educated parents. As expected, this is due entirely to the effect
that parental SES has on daughters’ occupational ambition. Hence, when we
introduce average wages in the aspired occupation, the effects of parental educa-
tion on girls’ sex-typicality disappear (see model 2). In other words, girls from
more privileged backgrounds tend to aspire to better-paid occupations, which
are on average less sex-typical (since there are few women in the better-paid
jobs).
Interestingly, we also find a significant negative effect of parental educa-
tion on sons’ probability of sex-typical aspirations, but in this case effects are
observed only for sons of parents with tertiary education (when compared to
sons of uneducated parents). This difference between the degree of sex-typicality
of boys coming from the two extremes of the parental educational distribution
20 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
is not fully accounted for by controlling for average wages in sons’ aspired
occupations (see model 2). This suggests that highly educated backgrounds
could reduce boys’ levels of sex-typicality through channels other than occupa-
tional ambition.
Model 1 also tests for personality effects. As explained above, these effects
are measured using repeated multi-item measures of children’s motivation and
self-esteem. Crucially, both personality indicators seem to have a direct influence
on the degree of sex-typing of children’s occupational aspirations. Girls—but
not so clearly boys—with high levels of motivation and both girls and boys with
high levels of self-esteem report less sex-typical occupational preferences. These
results are symmetric in that if we construct our dependent variable as low,
rather than high, sex-typicality, we find consistent results. We have hypothesized
that motivation and self-esteem could influence sex-typicality through two dis-
tinctive mechanisms: ambition and autonomy, though with motivation being
particularly associated with ambition. Both motivation and self-esteem are, as
we would expect, positively and significantly correlated with the wages of the
aspired occupation, consistent with an ambition mechanism.
14
By introducing
average wages as a control for the hierarchy of children’s occupational aspira-
tions, model 2 tests this ambition effect controlling for background character-
istics and allows us to investigate whether there is evidence for the autonomy
mechanism.
We note that the negative effect of both school motivation and self-esteem
on girls’ occupational sex-typicality disappears when occupational hierarchy is
accounted for. This suggests that both self-esteem and motivation increase girls’
occupational ambition, which by itself decreases occupational sex-typicality
(H4a). Again, the results are consistent if we regress on low rather than high sex-
typicality. Yet, given the high negative correlation between average wages and
proportion of women in aspired occupations, we cannot tell whether self-esteem
has, as expected, an extra independent effect on girls’ levels of sex-typicality.
Daughters who aim for high-paid occupations are at the same time ambitious
and sex-atypical, and this makes it particularly hard to separate autonomy from
ambition effects for girls.
It is boys who provide the best grounds for testing the autonomy mechanism—
that is, the idea that self-esteem boosts children’s capacity to act against the
existing social norms regarding sex-typical behavior. Boys can choose male-
dominated occupations at both ends of the occupational distribution, and this
implies that ambition and autonomy effects are not necessarily confounded for
them. Crucially, model 2 shows that the negative effect of self-esteem on sex-
typical aspirations for boys is fully resistant to controlling for the average wage
of the aspired occupation, our measure of the occupational ranking of chil-
dren’s aspirations. This indicates that self-esteem reduces boys’ occupational
sex- typicality at all levels of the aspired occupational hierarchy, a finding that is
fully consistent with the autonomy mechanism. Our interpretation is that boys
with high self-esteem are better predisposed to exercise their individual agency
and hence more capable of acting independently of those social influences that
promote sex-typical behavior (H4b).
Nurse or Mechanic? 21
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Table 5. Logistic Regression Estimates for Probability of Aspiring to a Highly Sex-Typed
Occupation
Girls,
model 1
Girls, with
wage control
Boys,
model 1
Boys, with
wage control
Age 0.103
†
0.197** 0.212*** 0.195***
(0.0543) (0.0655) (0.0512) (0.0527)
Absent father 0.00983 0.137 0.327 0.351
†
(0.198) (0.246) (0.201) (0.203)
Parental qualifications (Ref = none)
Higher −0.871*** −0.00920 −0.747** −0.532*
(0.234) (0.291) (0.238) (0.241)
Upper secondary −0.348
†
0.170 −0.408* −0.309
(0.197) (0.242) (0.206) (0.207)
Lower secondary −0.531** −0.293 −0.133 −0.0631
(0.191) (0.227) (0.192) (0.192)
Child’s occupation
matches same-sex parent
0.528 0.383 0.917* 0.642
†
(0.434) (0.666) (0.377) (0.373)
Mother’s occ. gender typicality (ref = intermediate)
Gender atypical −0.447** −0.577** −0.0827 −0.0335
(0.151) (0.186) (0.155) (0.157)
Gender typical −0.00879 −0.00834 0.117 0.119
(0.144) (0.179) (0.145) (0.146)
Father’s occ. gender typicality (ref = intermediate)
Gender atypical 0.125 0.333 −0.0278 0.0166
(0.182) (0.231) (0.196) (0.203)
Gender typical −0.0645 −0.0320 0.676*** 0.697***
(0.183) (0.225) (0.183) (0.190)
Average occasions
mother was housewife
0.382
†
−0.193 0.176 0.0621
(0.220) (0.265) (0.234) (0.236)
Housework inequality −0.00195 0.0000367 0.0125* 0.0126*
(0.00535) (0.00555) (0.00507) (0.00523)
Motivation score −0.276*** −0.0407 −0.103
†
−0.0192
(0.0675) (0.0895) (0.0612) (0.0623)
Self-esteem score −0.175** −0.0402 −0.191** −0.148*
(0.0633) (0.0778) (0.0682) (0.0692)
Log of hourly wage in
aspired occupation
−3.304*** −1.353***
(0.160) (0.232)
Constant −0.806* 5.634*** −1.187*** 1.877**
(0.325) (0.495) (0.332) (0.584)
Observations 1,667 1,667 1,693 1,693
Pseudo R
2
0.05 0.31 0.07 0.10
Note: Controls for wave and family composition included. Estimates weighted and standard
errors adjusted for clustering in the household. Standard errors in parentheses.
*** p
< 0.001 ** p < 0.01 * p < 0.05
†
p < 0.10
22 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Discussion
Occupational sex segregation is an enduring feature of Western labor mar-
kets that has been strongly implicated in the perpetuation of gender inequality.
Analyzing the factors that influence the formation of sex-typical occupational
preferences is therefore critical for illuminating our understanding of gender
stratification. It is clear that gendered occupational choices begin early, before
girls and boys have any experience of the labor market. Moreover, these early
choices have real consequences in later life.
This study set out to shed light on the factors that shape the degree of sex-
typing in early occupational preferences. We exploited a data set that allowed us
to measure both children’s aspirations prior to labor-market contact and their
parents’ coterminous characteristics, as well to follow the children into their
early occupational outcomes. We investigated different channels of parental
influence on children’s occupational aspirations that are relevant for the trans-
mission of sex-typical preferences. At the same time, we have allowed for the role
of individual agency in the process of preference formation. In order to avoid the
risk of over-individualization, we have defended a restricted definition of agency
that is anchored in observable psychological attributes. This definition turns
a hitherto intangible concept into one that is both theoretically grounded and
empirically testable. Our analytical strategy has allowed us to estimate simulta-
neously the relative impact of parental influences and individual psychological
characteristics on the development of sex-typical occupational aspirations in
what constitutes an innovative approach to the study of preference formation.
We have identified several distinctive channels of parental influence, includ-
ing two distinctive mechanisms linking parental behavior to children’s occu-
pational preferences: occupational imitation and behavioral sex-role learning.
Our empirical models show that boys (but not girls) who imitate homo-linearly
are significantly more likely to have sex-typical aspirations. Yet very few young
children actually imitate, which suggests that this mechanism plays only a very
minor role in the intergenerational reproduction of sex-typed preferences.
Consonant with the social structural insights of sex-role theory, we have
found that the daughters of mothers who work in male-dominated jobs tend to
aspire to less sex-typical occupations themselves, while the sons of fathers who
work in traditionally male jobs display more sex-typical aspirations. Moreover,
boys—but not girls—living in families with a traditional division of housework
tend to aspire to more traditionally male occupations. Parents’ enactment of
gender roles, both inside and outside the household, thus seems to exert a sig-
nificant influence on the degree of sex-typicality of their children’s occupational
aspirations.
We further posited that parental socio-economic resources should affect the
degree of sex-typing in occupational preferences by influencing children’s occu-
pational ambition. Given the existence of vertical sex segregation, this ambi-
tion effect was expected to have consequences for sex-typicality only in the case
of daughters, since for them, aiming high on the occupational ladder typically
means aspiring to occupations where women do not predominate. Boys, on the
Nurse or Mechanic? 23
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
other hand, can find traditionally male occupations at both ends of the occu-
pational distribution, so higher occupational ambition does not automatically
imply lower sex-typicality. We have found that parental education does indeed
increase daughters’ occupational ambition, which in turn reduces the probabil-
ity of their having sex-typical occupational aspirations. Yet we have also found
that boys from highly educated backgrounds are less sex-typical than boys from
low-educated parents at all levels of the aspired occupational hierarchy. This
latter finding suggests that parental education could influence boys’ sex-typical
preferences through channels other than occupational ambition, most probably
the intergenerational transmission of certain values (e.g., egalitarianism) that are
linked to higher education.
15
Finally, we have found that psychological predispositions also have a sig-
nificant impact on children’s occupational preferences. Girls with high school
motivation and both girls and boys with high self-esteem are less likely to aspire
to gender-typical occupations, regardless of other family influences. Motivation
and self-esteem make girls more likely to aim higher on the occupational lad-
der, where female-dominated jobs are scarce. This is why, when we control for
the average wage of girls’ aspired occupations, both motivation and self-esteem
effects disappear. Vertical segregation makes it particularly hard for us to iden-
tify the exact mechanisms linking motivation and self-esteem to sex-typical pref-
erences in the case of girls.
We have found, however, that the effect of self-esteem on boys’ levels of sex-
typicality survives controls for wages in their aspired occupation. This means
that boys with high self-esteem are significantly less likely to aspire to tradi-
tionally male occupations at all levels of the occupational ladder. We interpret
this finding as indicating that boys with high self-esteem are better predisposed
to contradict the existing social norms regarding sex-typical behavior. This we
have called the autonomy effect of self-esteem. Autonomy is the core component
of agency, understood as the capacity to make independent choices.
To our knowledge, this study provides the first psychologically anchored test
of agency effects in the formation of children’s sex-typed occupational aspira-
tions. One interesting implication of this study is that any action directed to
increasing children’s motivation and self-esteem, if successful, is likely to reduce
occupational sex segregation in the future. Another obvious implication of this
study is that boys’ preferences also matter. Stressing that supply-side processes
leading to occupational sex segregation concern both genders might seem self-
evident, as obviously it takes both to make occupational sex segregation. Yet the
gender literature has traditionally paid much more attention to women’s choices
than to men’s. By focusing disproportionally on women’s preferences, research
on gender stratification could be missing out.
By stressing the importance of psychological predispositions in early prefer-
ence formation, this study also contributes to contemporary supply-side theo-
ries of gender stratification. In a number of influential papers, Correll (2001,
2004) has shown that social-psychological processes of status generalization in
achievement-oriented settings (e.g., schools) can lead to gender-biased assess-
ments about task-specific self-competence. Such biased assessments (e.g., beliefs
24 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
about mathematical competence) may in turn lead to sex-typed academic choices,
from which segregated occupational outcomes are likely to follow. We believe
a unique contribution of our approach is that it can help us better understand
why some individuals are more sensitive to status-generalization processes than
others, a question that contemporary supply-side models have not addressed to
date (see also Correll and Ridgeway 2003; Ridgeway 1997, 2011; Ridgeway
and Correll 2004).
Our final comment concerns what we cannot explain. Although our models
show that there is an interpretable structure in the distribution of preferences,
on the strength of the pseudo R-squared statistics, their overall contribution to
the explanation of segregation in occupational aspirations must be judged as
modest. This means that much still remains to be explained.
The impact of other socialization agents (e.g., peers, teachers, mass media)
and situational contexts (e.g., schools) unaccounted for in this study could play
an important role in explaining part of the variance currently represented by
children’s own sex (Corsaro and Fingerson 2003; Hitlin 2006). Similarly, recent
explanations suggest that in informing their occupational choices children could
learn from wider social signals besides their own family experiences (Polavieja
2012). Yet testing for these wider social influences seems particularly hard with
the existing data, since we lack direct measures of horizontal socialization that
are external to the family.
Given these constraints, perhaps the best way of approaching horizontal
influences, the impact of which is expected to affect all children at a given time,
is by looking at cohort shifts. Cohort shifts should be expected if there are soci-
etal changes that affect the socialization milieu in which all children are embed-
ded. Such shifts would include macro-level changes in the labor market and
domestic behavior—from which children can learn—as well as changes in gen-
der attitudes, values, and cultural representations. In all these realms, observed
trends in advanced Western societies have worked in favor of greater gender
equalization.
Partially consistent with horizontal pressures for gender equalization, our
data show a decline over time (net of other factors) in the tendency for girls to
prefer sex-typical occupations (see figure 3). This decline represents a reduction
of around four percent in the aspired proportion female from one decade to the
next.
16
Yet no cohort trend is found for boys.
17
Given the lack of convergence
from boys and the modest size of the effect for girls, we must conclude that, even
if horizontal socialization pressures for sex-typing are declining over time for
girls, it would take several generations before this was reflected in a shift from
the current picture of highly segregated aspirations.
Meanwhile, we believe this study has already shown that focusing on the
interplay between socialization influences and individual psychological predis-
position can yield important analytical payoffs. We have provided new insights
into the correlates of sex-typing in the occupational choices of children. Our
findings strongly suggest that both social influences and individual psychological
predispositions provide the essential cogs and wheels of preference formation.
Yet we still lack a clear understanding of how these pieces are assembled. To
Nurse or Mechanic? 25
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Figure 3. Cohort shifts in the sex-typicality of children’s occupational aspirations by sex, estimates from full models
0
2
4
6
8
Density, boys
0.4 0.5 0.6 0.7 0.8 0.9 1.0
Estimated proportion men in occupation
Earlier cohort Later cohort
0
2
4
6
8
Density, girls
0.3 0.4 0.5 0.6 0.7 0.8
Estimated proportion women in occupation
Earlier cohort Later cohort
Source: Distributions estimated from full regression models fitted separately by sex, British Household Panel Survey Waves 4–18.
Note: Cohorts are defined so as to split the sample into two equal halves. Earlier cohort comprises observations from waves 4 to 13 (1994–2003), later
cohort comprises observations from waves 14 to 18 (2004–2008).
26 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
advance our study of mechanisms further may entail exploring the formation of
explicit gendered aspirations and expectations even earlier in children’s lives, to
account for the crucial influence of peers and context, and to explore the effects
of other psychological attributes and traits possibly associated with children’s
capacity to make independent choices.
Notes
1. Self-esteem is argued to be composed of two dimensions: self-worth and self-compe-
tence (see, e.g., Tafarodi and Swann 1995; Cast and Burke 2002).
2. Tafarodi and Swann (2001, 655) define self-competence as the “valuative imprint of
self-efficacy on identity.”
3. Consistent with this interpretation, several empirical studies have failed to clearly
distinguish between global self-esteem and generalized self-efficacy (see Bernard
et al. 1996; Judge et al. 1998; Judge et al. 2002).
4. While our argument stems largely from discussions of the self-competence dimen-
sion of self-esteem, the distinction between the self-worth and self-competence
dimensions can easily break down conceptually and empirically (see further Data
and Methodology). Thus, our interpretation connecting self-esteem to agency should
also hold for a single global measure combining both dimensions.
5. Only about one-third of these were observed five times, which is the maximum
number of waves a respondent can be in the youth panel; around 15 percent were
observed for each of two, three, or four waves, and 19 percent were observed only
once.
6. We matched on SOC90 occupational codes, avoiding a series break at the change to
SOC2000 in the LFS in 2001.
7. For parents not currently in paid work, we used information on their last job.
8. To ensure that our findings were not driven by a few favored aspirant occupations of
boys and girls, we estimated an alternative specification of our models excluding the
favorite five occupations of both boys and girls. This did not alter our results.
9. Results are available on request.
10. The GCSE, General Certificate of Secondary Education, is an academic qualification
awarded in a specified subject, generally taken in a number of subjects by students
aged 14–16 in secondary education in England, Wales, and Northern Ireland. In
Scotland, the exam is called Standard Grades.
11. The four measures asked from wave 4 were (1) “I feel that I have a number of good
qualities”; (2) “All in all, I am inclined to feel that I am a failure”; (3) “I am able
to do things as well as most other people”; and (4) “I feel I do not have much to
be proud of.” The measures added from wave 9 were (5) “I certainly feel useless at
times”; and (6) “At times I think I am no good at all.” In the social psychology litera-
ture, items 1, 4, and 6 are considered part of the self-worth dimension, while items
2, 3, and 5 are considered part of self-competence. However, exploratory factor
analysis indicated that all items loaded on a single factor, and there was no support
for higher inter-item correlation among the items covering each individual dimen-
sion compared to those crossing dimensions.
12. The fact that our data set includes only a limited version of the full Rosenberg scale
is possibly the reason we cannot precisely identify each theoretical dimension.
13. Since average wages are highly endogenous to the outcome variable, we avoid any
interpretation of its coefficient in terms of “effects.”.
14. The correlations are around 0.17 for motivation and around 0.10 for self-esteem.
Nurse or Mechanic? 27
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
15. Lower-educated parents in our data can be shown to have significantly higher gender
traditionalism scores than higher-educated ones.
16. This effect was robust to splitting the period at different points. Results are available
on request.
17. It is also worth noting the lower average proportion of women in girls’ aspired occu-
pations (58 percent) compared to their mothers’ achieved occupations (71 percent),
whereas boys’ aspirations are little different from the average gender concentra-
tion experienced by their fathers, at around 23 percent. This is consistent with the
observed cohort shift for girls in aspirations.
About the Authors
Javier Polavieja is Banco de Santander Professor of Sociology at Universidad
Carlos III de Madrid. His main fields of research are labor-market stratifica-
tion, economic sociology, political sociology, and immigration research. He is
particularly interested in the interplay of social, institutional, and economic
processes leading to labor-market inequalities. Recent publications include arti-
cles in the American Journal of Sociology on sex differences in skill specializa-
tion and Labour Economics on the economic determinants of attitudes toward
immigrants.
Lucinda Platt is Professor of Social Policy and Sociology at the London School
of Economics and Political Science. She researches in the area of inequalities and
stratification, with a particular emphasis on longitudinal approaches. Specific
interests are in ethnic minority pay, mobility, and income; ethnic identity; child
poverty; and youth transitions and aspirations. Recent publications include
articles in Ethnic and Racial Studies on ethnic identity and Oxford Economic
Papers on pay gaps.
References
Arnett, Jeffrey J. 1995. “Broad and Narrow Socialization: The Family in the Context of a Cultural Theory.”
Journal of Marriage and Family 57:617–28.
Bandura, Albert. 1977. Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.
______. 1990. “Conclusion: Reflections on Nonability Determinants of Competence.” In Competence
Considered, edited by Robert J. Sternberg and John Kolligian Jr. New Haven, CT: Yale University Press.
______. 1997. Self-Efficacy: The Exercise of Control. New York: Freeman.
Baumeister, Roy F., Jennifer D. Campbell, Joachim I. Krueger, and Kathleen D. Vohs. 2003. “Does High-
Self-Esteem Cause Better Performance, Interpersonal Success, Happiness, or Healthier Lifestyles?”
Psychological Science in the Public Interest 4(1):1–44.
Bernard, Larry C., Steven Hutchison, Alexander Lavin, and Pamela Pennington. 1996. “Ego-Strength,
Hardiness, Self-Esteem, Self-Efficacy, Optimism, and Maladjustment: Health-Related Personality
Constructs and the ‘Big Five’ Model of Personality.” Assessment 3(2):115–31.
Bussey, Kay, and Albert Bandura. 1999. “Social-Cognitive Theory of Gender Development and Differ-
entiation.” Psychological Review 106:676–713.
Carneiro, Pedro, and James J. Heckman. 2005. “Human Capital Policy.” In Inequality in America: What
Role for Human Capital Policy? edited by B. M. Friedman. Cambridge, MA: MIT Press.
Cast, Alicia D., and Peter J. Burke. 2002. “A Theory of Self-Esteem.” Social Forces 80(3):1041–68.
28 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Chang, Mariko Lin. 2004. “Growing Pains: Cross-National Variation in Sex Segregation in Sixteen
Developing Countries.” American Sociological Review 69(1):114–37.
Correll, Shelley J. 2001. “Gender and the Career Choice Process: The Role of Biased Self-Assessments.”
American Journal of Sociology 106:1691–730.
______. 2004. “Constraints into Preferences: Gender, Status, and Emerging Career Aspirations.” American
Sociological Review 69:93–113.
Correll, Shelley J., and Cecilia L. Ridgeway. 2003. “Expectation States Theory.” In Handbook of Social
Psychology, edited by John Delamater. New York: Springer.
Corsaro, William A. 2005. The Sociology of Childhood. 2nd ed. Thousand Oaks, CA: Pine Forge Press.
Corsaro, William A., and Laura Fingerson. 2003. “Development and Socialization in Childhood.” In
Handbook of Social Psychology, edited by John Delamater. New York: Springer.
Crouter Ann C., Beth A. Manke, and Susan M. McHale. 1995. “The Family Context of Gender Intensification
in Early Adolescence.” Child Development 66:317–29.
Cunningham, Mick. 2001. “Parental Influences on the Gendered Division of Housework.” American
Sociological Review 66:184–203.
Eagly, Alice H. 1987. Sex Differences in Social Behavior: A Social Role Interpretation. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Eagly, Alice H., Wendy Wood, and Amanda B. Diekman. 2000. “Social Role Theory and Sex Differences
and Similarities: A Current Appraisal.” In The Developmental Social Psychology of Gender, edited by
Thomas Eckes, and Hanns M. Trautner. Hillsdale, NJ: Lawrence Erlbaum Associates.
Elder, Glen H. 1994. “Time, Human Agency, and Social Change: Perspectives on the Life-Course.” Social
Psychology Quarterly 57(1):4–15.
Gamoran, Adam. 1996. “Educational Stratification and Individual Careers.” In Generating Social
Stratification: Toward a New Research Agenda, edited by Alan C. Kerckhoff. Boulder, CO: Westview.
Gecas, Victor. 2003. “Self-Agency and the Life-Course.” In Handbook of the Life Course, edited by Jeylan
T. Mortimer and Michael J. Shanahan. New York: Springer.
Gray-Little, Bernadette, Valerie S. L. Williams, and Timothy D. Hancock. 1997. “An Item Response
Theory Analysis of the Rosenberg Self-Esteem Scale.” Personality and Social Psychology Bulletin 23:
443–51.
Hakim, Catherine. 1991. “Grateful Slaves and Self-Made Women: Fact and Fantasy in Women’s Work
Orientations.” European Sociological Review 7:101–21.
______. 2000. Work-Lifestyle Choices in the 21st Century: Preference Theory. Oxford: Oxford University
Press.
Harper, Barry, and Mohammad Haq. 2001. “Ambition, Discrimination, and Occupational Attainment: A
Study of a British Cohort.” Oxford Economic Papers 53:695–720.
Hays, Sharon 1994. “Structure and Agency and the Sticky Problem of Culture.” Sociological Theory
12:57–72.
Heckman, James J., Jora Stixrud, and Sergio Urzua. 2006. “The Effects of Cognitive and Noncognitive
Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24:411–82.
Hitlin, Steven. 2006. “Parental Influences on Children’s Values and Aspirations: Bridging Two Theories of
Social Class and Socialization.” Sociological Perspectives 49:25–46.
Hitlin, Steven, and Glen H. Elder. 2007. “Time, Self, and the Curiously Abstract Concept of Agency.”
Sociological Theory 25:170–91.
Jackson, Michelle. 2006. “Personality Traits and Occupational Attainment.” European Sociological Review
22:187–99.
Jacobs, Janis, Stephanie Lanza, Wayne Osgood, Jacquelynne S. Eccles, and Allan Wigfield. 2002.
“Changes in Children’s Self-Competence and Values: Gender and Domain Differences across Grades
One through Twelve.” Child Development 73(2):509–27.
Nurse or Mechanic? 29
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Judge, Timothy A., Edwin A. Locke, Cathy C. Durham, and Avraham N. Kluger. 1998. “Dispositional Effects
on Job and Life Satisfaction: The Role of Core Evaluations.” Journal of Applied Psychology 83(1):17–34.
Judge, Timothy A., Amir Erez, Joyce E. Bono, and Carl J. Thoresen. 2002. “Are Measures of Self-Esteem,
Neuroticism, Locus of Control and Generalized Self-Efficacy Indicators of a Common Core Construct?”
Journal of Personality and Social Psychology 83:693–710.
Kanazawa, Satoshi. 2001. “De Gustibus Est Disputandum.” Social Forces 79:1131–62.
Martin, Carol Lynn, Diane N. Ruble, and Joel Szkrybalo. 2002. “Cognitive Theories of Early Gender
Development.” Psychological Bulletin 128:903–33.
Maume, David J. 1999. “Occupational Segregation and the Career Mobility of White Men and Women.”
Social Forces 77:1433–59.
Office for National Statistics, Social and Vital Statistics Division. 1994–2000. Quarterly Labour Force
Survey, January − March 1994 to October − December 2000. Study Numbers: 5414, 5414, 5416, 5417,
5856, 5857, 5858, 5859, 5863, 5864, 5865, 5866, 5867, 5868, 5869, 5870, 5871, 5872, 5873, 5874,
5875, 5876, 5877, 58778, 5879, 5880, 5881, 5882 [computer files]. Colchester, UK: UK Data Archive
[distributor].
Okamoto, Dina, and Paula England. 1999. “Is There a Supply Side to Occupational Sex Segregation?”
Sociological Perspectives 42(4):557–82.
Polavieja, Javier G. 2009. “Domestic Supply, Job-Specialization, and Sex Differences in Pay.” Social
Indicators Research 93:587–605.
______. 2012. “Socially Embedded Investments: Explaining Gender Differences in Job-Specific Skills.”
American Journal of Sociology 118:592–634.
Reskin, Barbara F. 2003. “Including Mechanisms in our Models of Ascriptive Inequality.” American
Sociological Review 68:1–21.
Ridgeway, Cecilia L. 1997. “Interaction and the Conservation of Gender Inequality: Considering
Employment.” American Sociological Review 62:218–35.
______. 2011. Framed by Gender. New York: Oxford University Press.
Ridgeway, Cecilia L. and Shelley J. Correll. 2004. “Unpacking the Gender System: A Theoretical Perspective
on Gender Beliefs and Social Relations.” Gender & Society 18(4):510–31.
Robins, Richard W., Holly M. Hendin, and Kali H. Trzesniewski. 2001. “Measuring Global Self-Esteem:
Construct Validation of a Single-Item Measure and the Rosenberg Self-Esteem Scale.” Personality and
Social Psychology Bulletin 27(2):151–61.
Roscigno, Vincent J., Lisette M. Garcia, and Donna Bobbitt-Zeher. 2007. “Social Closure and Processes
of Race/Sex Employment Discrimination.” Annals of the American Academy of Political and Social
Science 609:16–48.
Rosenberg, Morris. 1965. Society and the Adolescent Self-Image. Princeton, NJ: Princeton University
Press.
Schmitt, David P., and Jüri Allik. 2005. “Simultaneous Administration of the Rosenberg Self-Esteem Scale
in 53 Nations: Exploring the Universal and Culture-Specific Features of Global Self-Esteem.” Journal of
Personality and Social Psychology 89(4):623–42.
Sribney, William M. 1995. “sg37: Orthogonal Polynomials.” In Stata Technical Bulletin 25:17–18. Reprinted
in Stata Technical Bulletin Reprints 5:96–98. College Station, TX: Stata Press.
Tafarodi, Romin W., and Alan B. Milne. 2002. “Decomposing Global Self-Esteem.” Journal of Personality
70:443–83.
Tafarodi, Romin W., and William B. Swann. 1995. “Self-Liking and Self-Competence as Dimensions of
Global Self-Esteem: Initial Validation of a Measure.” Journal of Personality Assessment 65(2):322–42.
______. 2001. “Two-Dimensional Self-Esteem: Theory and Measurement.” Personality and Individual
Differences 31:653–73.
Tomaskovic-Devey, Donald. 1993. “The Gender and Race Composition of Jobs and the Male/Female,
White/Black Pay Gaps.” Social Forces 72:45–76.
30 Social Forces
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
Tomaskovic-Devey, Donald, Kevin Stainback, Tiffany Taylor, Catherine Zimmer, Corre Robinson, and Tricia
McTague. 2006. “Documenting Desegregation: Segregation in American Workplaces by Race, Ethnicity,
and Sex, 1966–2003.” American Sociological Review 71(4):565–88.
Udry, Richard J. 2000. “Biological Limits of Gender Construction.” American Sociological Review
65:443–57.
University of Essex, Institute for Social and Economic Research. 2010. British Household Panel Survey:
Waves 1–18, 1991–2009 [computer file]. 7th ed. Colchester, UK: UK Data Archive [distributor], July.
SN: 5151.
van Putten, Anne, Perl A. Dykstra, and Joop J. Schippers. 2008. “Just Like Mom? The Intergenerational
Reproduction of Women’s Paid Work.” European Sociological Review 24(4):435–49.
Wigfield, Allan, and Jacquelynne S. Eccles. 2000. “Expectancy-Value Theory of Achievement Motivation.”
Contemporary Educational Psychology 25:68–81.
Nurse or Mechanic? 31
by guest on May 13, 2014http://sf.oxfordjournals.org/Downloaded from
- CitationsCitations1
- ReferencesReferences30
- "Parsonian approaches also depict cultural traditions as unified and largely coherent symbolic systems , widely shared by all members of a given social group. This conceptualization inevitably leads to essentialist views of national and ethnic cultures, as well as an over-socialized conception of human action (Gecas 2003; Hays 1994; Polavieja and Platt 2014 ). Contemporary motivational approaches have taken pains to distance themselves from the over-deterministic aspects of the Parsonian legacy, which has largely fallen out of favor in the field (see, e.g., Small et al. 2010; Smith 2003; Vaisey 2009). "
[Show abstract] [Hide abstract] ABSTRACT: We know that culture influences people’s behavior. Yet estimating the exact extent of this influence poses a formidable methodological challenge for the social sciences. This is because preferences and beliefs are endogenous, that is, they are shaped by individuals’ own experiences and affected by the same macro-structural conditions that constrain their actions. This study introduces a new method to overcome endogeneity problems in the estimation of cultural effects by using migrant populations. This innovative method uses imputed traits, generated from non-migrating equivalents observed at the country of origin, as instruments for immigrants’ own cultural traits measured at the country of destination. By construction, imputed traits are exogenous to immigrants’ host social environment. The predicted power of imputed traits over observed traits in instrumental-variable estimation captures the non-idiosyncratic component of preferences and beliefs that migrants and non-migrating equivalents share as members of the same national-origin group, that is, their culture. I use this innovative method to estimate the net exogenous impact of traditional values on female labor-force participation in Europe. I find that this impact is much larger than standard regression methods would suggest.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.
This publication is from a journal that may support self archiving.
Learn more