Content uploaded by Laura Bernardi
Author content
All content in this area was uploaded by Laura Bernardi on Nov 23, 2016
Content may be subject to copyright.
Content uploaded by Laura Bernardi
Author content
All content in this area was uploaded by Laura Bernardi on Sep 15, 2015
Content may be subject to copyright.
PaVie Working Papers
2010Ͳ4
GenderroleͲset,familyorientations,andfertility
intentionsinSwitzerland
LauraBernardi,JeanͲMarieLeGoff
&ValérieͲAnneRyser
CenterforLifeCourseStudies
Centred>étudedesParcoursdeVie
ThePaVie Working Papers is aworkͲinͲprogress onlineseries.Each paper receives only limited
review.Authors areresponsible forthepresentation offacts andfortheopinionsexpressed
therein,which donotnecessarily reflect those ofthePaVie Center.
Institutd'Etudesdémographiques
etdesParcoursdeVie
CHͲ1211Genève
Laboratoired>étudedes
ParcoursdeVie
CHͲ1015Lausanne
EͲmail:pavie_wp@unil.ch
Website:http://www.unil.ch/labopavie
Bernardi,L.¹,LeGoff,J.ͲM.¹&Ryser,V.ͲA.² (2010).GenderRoleͲSet,Family
Orientations,andFertilityIntentionsinSwitzerland.PaVie Working Papers,
4.Lausanne:PavieCenter.
Abstract
Thispaperinvestigateswomen>sattitudestowardspaidemploymentandfamilyin
relationtotheshareofpaidanddomesticworkloadswithinacoupleandinrelationto
theintentiontohaveachild.Weuselongitudinaldatafromthe SwissHouseholdPanel
thatquestionsbothpartnersseparatelyonfamilyandworkvalues,fertilityintentions,
andgenderdivisionoftasks.Ourmultilevelestimationsconfirm theimportantroleof
subjectiveevaluationsofthecouple>sroleͲset:beingsatisfiedwiththecouple>sroleͲset
favorsfertilityintentionsforwomenwhoarealreadymothers.However,forchildless
women,aspirationstoeconomicindependenceandbeingemployedhavean
independentandnegativeeffectonfertilityintentions.
Keywords:Fertility intentions,Gender
Authors< affiliation:University ofLausanne¹ andSwissCentreofExpertiseinSocial
Sciences(FORS)²
Correspondence to:laura.bernardi@unil.ch
Laboratoired>étudedes
ParcoursdeVie
CHͲ1015Lausanne
EͲmail:pavie_wp@unil.ch
Website:http://www.unil.ch/labopavie
CenterforLifeCourseStudies
Centred>étudedesParcoursdeVie
Institutd'Etudesdémographiques
etdesParcoursdeVie
CHͲ1211Genève
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 1
Gender Role-Set, Family Orientations, and Fertility Intentions in Switzerland
Contemporary low levels of fertility in Switzerland as well as in many other European
regions raised questions about the determinants of shrinking family sizes and the rise in
childlessness. Most studies in post-industrial societies indicate that if the family size which
people desire, expect, or consider ideal as declared in surveys were realized, fertility rates would
be up to levels at around replacement. However, in most cases both at the macro and at the micro
level there is a gap between the expected number of children and the actual number of children
born (Le Goff, Sauvain-Dugerdil, Rossier, & Coenen-Huther, 2005). It has been claimed,
paraphrasing the more famous expression “unmet need for contraception”, that we face a sort of
“unmet need” for children which would justify both attention by family researchers and
intervention thorough child-friendly family policies (Chesnais, 1998). Identifying the
determinants of the intentions-realizations gap in fertility is therefore crucial. One key aspect is
the gendered nature of parenthood and the different gender norms ruling representations and
practices of appropriate parental roles for men and women.
The terms “stalled revolution” and “second shift” (Hochschild & Machung, 1989) are by
now accepted and powerful synonyms for a situation in which women are increasingly active in
the labor market while men prolong their absence from the domestic sphere. Similar meanings
are suggested by expressions like “dual-burden,” “double-burden,” and “double-day” (Baxter,
2002; Bittman, 1999; Bittman & Matheson, 1996; Harrington, 1998; Meissner, Humphreys,
Meis, & Scheu, 1975; Pahl, 1984; Shelton, 1992). While these terms describe an actual division
of tasks between partners, the value and normative dimension surrounding the stalled revolution
are implicit in the half cynical term “supermom” (Hochschild & Machung, 1989; Shaevitz,
1984), indicating attitudes and norms which expect women to successfully fulfil their aspirations
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 2
in the domestic and labor market without asking for help from her partner, or from society at
large. Gendered practices are not necessarily perceived as unfair, but are rather talked about and
explained in discourses of partners’ free choice and gendered preferences or abilities for given
tasks. In turn, these preferences and abilities are justified in terms of socialization and attitudes to
given gender roles or in terms of the different biological predisposition of men and women in
parenting (Blain, 1994).
The stalled revolution and the double burden have been indicated as two of the causes for
the emergence and persistence of low fertility and increased childlessness in the late twentieth
century. McDonald’s (2000) seminal paper on gender systems and family dynamics argues that
those gender systems in which equality is expected and supported in the public sphere of the
market and the law, but not in the domestic sphere of family relations and responsibilities, are
likely to experience low fertility. McDonald’s prototypes for such gender systems are
contemporary Southern European countries like Italy and Spain, where high expectations on
women’s time in the domestic sphere would be avoided by delaying or forgoing additional
family responsibilities represented by children.
While McDonald’s arguments are macro, his explanation relies on the micro level: A
woman who competes in the public sphere on a basis of gender equal treatment and who expects
at the same time to be mostly responsible for house and child care will perceive high costs-
opportunities in having a first child or an additional child. On the one hand, an unequal treatment
of men and women in the labor market granting women some sort of privileged status to
compensate for their extra labor in the domestic sphere is not viable because it would be
perceived as unfair in modern democratic societies, which equality for all citizens. On the other
hand, the need to choose part time employment or to opt for a temporary or permanent exit from
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 3
the labor market would reduce women’s life-long earnings and career opportunities. The double
burden under these conditions becomes a too disadvantageous option for women and would have
discouraging effect on fertility.
Several studies apply McDonald’s analysis at the micro level by studying the relation
between couples’ role-set and fertility (Cooke, 2003; Mills, Mencarini, Tanturri, & Begall, 2008;
Olah, 2003; Tazi-Preve, Bichlbauer & Goujon, 2004; Torr & Short, 2004). Indeed most of this
empirical evidence shows a negative effect of the unequal role-set on fertility for both second
and third births. The perception of an unfair division of domestic labor within the couple has
been associated with a greater likelihood of depression and divorce, negative opinions of marital
quality, and overall satisfaction (Coltrane, 2000). However, such studies raise the issue of
whether the share of domestic tasks alone does accurately tell the whole story. In other words,
fertility decisions may be related not only to the amount of domestic work, but also to the overall
workload share (total time spent on paid and unpaid work taken together) between men and
women in each couple. Such shares are broadly similar in most industrialized countries (Bianchi,
Robinson, & Milkie., 2006; Bittman & Wajcman, 2000; Greenstein, 2000; Robinson & Godbey,
1997; Shelton & Firestone, 1989).
Switzerland: Inconsistent Public and Private Gender Spheres
Contemporary Switzerland fits well in McDonald’s category of countries where gender
spheres are inconsistent. On the one hand, equality between men and women at the institutional
level is granted. At each political level (e.g., confederation, canton, and municipality), as well as
at the level of other types of institutions (e.g., universities), there are gender equality offices in
charge of promoting women’s professional careers and often men’s participation in domestic
work. Social beliefs and norms favor gender equality and particularly childless couples value it
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 4
(Le Goff, Levy, Sapin, & Camenisch, 2009; Levy, Joye, Guye, & Kaufmann, 1997). On the other
hand, gender practices show a different picture of the labor market, one much more gender
biased than that portrayed in the official regulations. Despite the promotion of gender equality at
the institutional level, the development of part time jobs has contributed to keeping women in
lower paying jobs with less responsibility. Then, several studies have shown that inequalities
between men and women are part of everyday experience, especially after the transition to
parenthood. Most women reduce the time they spend working once they become mothers (Levy,
2006; Le Goff et al., 2009). According to the state regimes of time policy proposed by Anxo et al
(2006), Swiss women’s patterns of labor market participation can be labelled as maternal part
time work. Data from the Swiss census 2000 show that part time working schedules became the
norm for mothers in Switzerland. While at the end of the 1980s the pattern was rather
characterized by the discontinuation of participation in the labor force by mothers of young
children, during the 1990s part time employment became the preferred alternative. Only 20% of
them go back to full time employment after a period of interruption, while a little more than half
of the mothers continue working part time (Levy, 2006; Le Goff et al., 2005; Widmer &
Ritschard, 2009). The lack of explicit life course policies supporting mothers’ full participation
in the labor force translates therefore into a life course regime (an empirically dominant model of
practices) in which women interrupt their careers when they become mothers. By contrast, men‘s
participation in the labor market, mostly full time, is insensitive both to parental status and to the
number of children men have. The decrease in the time that women spend working in the labor
market after the transition to parenthood corresponds to an equivalent increase in the time they
devote to domestic work (Henchoz & Wernli, to be published) and to the emergence of a more
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 5
traditional gender role-set even among those couples who before the birth of the child declared
being in favor of equal sharing (Le Goff et al., 2009).
In the Swiss context, giving birth represent a turning point in a woman’s life course. The
origin of this phenomenon is to be found in the gendered nature of cultural and institutional
integration of women in Swiss society. The theory of gendered master status postulates that
family and work represent the two major spheres of social integration for individuals, and that
these differ for men and women (Krüger & Levy, 2000, 2001). The priority integration for
women is the domestic sphere (family); this does not mean that women are excluded from the
labor market (work), but that their professional path is subordinate to family life. For men,
professional integration has the priority, while family life is subordinate to it. Several recent
studies demonstrate that women’s and men’s professional trajectories remain quite different
(Levy, Gauthier, & Widmer, 2006). These studies specify that, compared to those of men,
women’s professional trajectories are more heterogeneous and sensitive to characteristics such as
education level, number of children, and cohort. These different studies show that women’s
working patterns are constructed and negotiated within the limits represented by family life,
while men’s family involvement is limited by the demands of the professional life. Given that
motherhood represents such a turning point in women’s professional life course and more in
general in their social integration patterns, it is crucial to take into account the function of the
gender role-set within the couple, as well as take into account how each partner’s gender
orientation within the couple affects his or her intentions to have a child.
Fertility Intentions and the Gendered Workload
Childbearing behavior may be predicted by declared intentions to have a child. For this
reason, research on fertility intentions attracted the attention of social psychologists and
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 6
demographers (Miller & Pasta, 1995; Philipov, Thévenon, Klobas, Bernardi, & Liefbroer, 2009)
as important antecedents of fertility. Yet, the empirical literature highlights that there are often
large discrepancies between declared intentions and realized behavior due to the instability of
intentions, to external and unpredicted factors intervening between the intention formulation and
its foreseen realization. However, the gap in intention-realization is substantially smaller when
intentions refer to a precise and relatively short time interval (2-3 years) rather than the whole
life course, and when they are formulated with respect to the next child only rather than to the
overall fertility (final number of children) of a couple. Fertility intention is often found to be
correlated to individual socio-demographic characteristics similar to those that matter for
behavior (e.g., age, parity, marital and employment status), for ideational factors like norms and
values (e.g., religious affiliation and practices, family and gender values), and for institutional
opportunity structures (e.g., childcare availability and social networks support).
Despite such a blooming interest in fertility intentions, the effect of paid and unpaid labor
shares between man and woman in a couple on their fertility intentions is rarely addressed (Mills
et al., 2008). The main question to answer would be: Are couples in which the woman takes on
the primary responsibilities of the domestic sphere and share equally with her partner the paid
work load less likely to intend or have a child than couples where both partners take equal
responsibilities in both spheres? Rizzi, Judd, White, Bernardi, and Kertzer (2008) address this
question by analyzing couples’ role-sets based on partners’ share of domestic and paid work
jointly and their relations to women’s fertility intentions, controlling for measures of familistic
values in the Italian context. The starting point of Rizzi et al. is the identification of a typology of
couples’ role-sets, built on the basis of the number of hours that each partner devotes to either
domestic tasks or paid work and the gaps between the partners’ share in domestic work hours and
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 7
in (paid) labor work hours. Results show that a traditional role-set—in which the woman carries
out most of the domestic work and the man most of the paid labor—is predominant in Italy, even
among working women. The researchers observe no association between the way in which
partners arrange their share of paid and unpaid work and women’s intentions to have a first child.
In contrast, there is evidence of the negative effects of role-sets in which women do most of the
domestic work and as much paid work as men on the intention to have a second child; the
opposite is true among couples sharing equally in domestic and paid work.
An analysis of the effect of workload share on fertility cannot ignore the important
mediating role that perceived satisfaction with the gender division of tasks may have in shaping
individual subjective well-being. Subjective well-being and the perceived quality of the
relationships have been pointed out in the theoretical literature as crucial, but again little
empirical research examines its effects on fertility jointly with the actual workload share. A few
exceptions are the study by Benin and Agostinelli (1988) who use U.S. data from the 1980s to
show that an important determinant of satisfaction with partners’ workload share is its perceived
fairness. Women were not happy with just “minimal participation” (p 350) from their partner,
disconfirming the idea that just a symbolic participation would be enough to reduce status
distinction between partners and therefore be sufficient to satisfy a partner. However, fair
arrangements corresponded to very different combinations of shares depending on whether
equity was considered as: a) an equal share of household tasks independently of the share of paid
work; b) an equal share in household tasks which are typically female (housework chores or
care) independently from other tasks (technical, administrative); c) a share in which “people
want to maximize their own rewards” (p 350) maximization” depending on exogenously given
preferences (women prefer caring tasks, husband’s success contributes prestige and status to the
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 8
family, preference for not having arguments on how tasks shall be performed when they are
shared).
The most recent study on fertility intentions in Switzerland based on Family and Fertility
Survey data of 1994 (Coenen-Huther, 2005) shows that intentions to have a child within 24
months decreased with the age of women and men, and with parity. Religion also plays a role.
Catholics, whatever their level of engagement with religious practices, were more likely to
intend to have a child than Protestants. Women who identified as not belonging to any religion
more often declared not having intention to have a child. The most frequent reasons for which
childless women declared not intending to have a child are related to the difficulty to conciliate
family and professional lives, worries associated with parenting itself, and problems related to
time scarcity. All these factors are concerned with the responsibility involved in parenting in
general and mothering in particular, and not so much with the economic costs of children per se.
However, mothers of one child or more who do not intend to have another child mention more
frequently economic costs than scarce time budget. Difficulties conciliating family and
professional lives and worries about parenting do not discriminate between mothers and non-
mothers. In the next sections we are going to examine the role of couples’ gender role-set, of
partners’ satisfaction with it, and of attitudes towards family and gender on fertility intentions in
Switzerland in the early twenty-first century.
Data and Measures
We realized this study using the data collected in the Living in Switzerland project. This
project is conducted by the Swiss Household Panel (SHP), which is based at the Swiss Centre of
Expertise in the Social Sciences (FORS), located at the University of Lausanne. The project is
devoted to analyzing changing living conditions in Switzerland and is funded by the Swiss
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 9
National Science Foundation (SNF). For our purposes, the SHP data are interesting because they
are longitudinal data and contain the relevant information on both partners of a couple. Two
random samples of households are followed yearly, and all household members older than 14
years are interviewed separately. The first cohort has been followed since 1999, while the second
since 2004. Because questions about fertility intentions have been asked only since 2002, we use
a subsample of the first cohort starting in 2002 only. About 2,513 women of the two cohorts
were asked at least once about their fertility intentions. However, for the purposes of our article,
we selected only women who are living in couples (married or cohabiting) and aged between 18
and 45, for whom we also have their partners’ interview data. Then analysis is conducted on a
subsample of about 3,058 observations across waves.
Dependent Variable
Our dependant variable is the intention to have a child (or another child) in the 24 months
following the interview (reference period). Fertility intentions are measured using a three point
scale: (1) women who intend to have a child; (2) women who do not know if they want a child;
(3) women who do not intend to have a child. A descriptive analysis shows that at each wave of
the SHP, the majority of childless women declare plans to have a child within this period, while
the majority of women with already one child do not plan to have another child (Table 1). In
both subsamples, women who do not know if they want a child are rare. For this reason, we
computed a dichotomous independent variable in which these answers are regrouped with
positive intentions to have a child.
Partners’ Workload Variables
Gender gap index. The SHP presents rich information on couples’ domestic and work
balance: the numbers of hours weekly devoted by men and women to domestic work and to
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 10
professional work respectively.1 The average time dedicated to domestic work is around 18 hours
a week for women’s observations, but with strong differences between mothers and non mothers.
Not surprisingly, the number of hours dedicated by childless women to domestic work is smaller
(almost 9 hours) compared to the number of hours dedicated to domestic work by women with at
least one child (21 hours). By contrast, there is no difference between men of corresponding
types, for all of whom implication in domestic work reaches a little more than 5 hours. The
amount of time spent by childless women on paid work is larger than that spent by mothers. The
number of hours of paid work these women work approaches the number of those of their
partners. In the case of men, there are no differences according to whether they are fathers or not
(see Table 1).
To go beyond a first descriptive approach and in order to take stock of the workload rate
and household rate for men and women within couples, we computed two relative indexes. First,
we divided women’s number of domestic hours per week by their partner’s number of domestic
hours per week. When this indicator is higher than 1, it indicates that women do more housework
compared to men. Second, we divided men’s number of paid labor hours per week by their
partner’s number of paid labor hours per week. A value higher than 1 here means that men do
more paid work than their partners.
Satisfaction with domestic tasks. In order to measure women’s satisfaction with the
organization of housework within the couple, we have a three point scale from 0 (“not at all
satisfied”) to 2 (“very satisfied”)2. Descriptive results show that childless women are more
1These questions do not include time devoted to child care. In order to measure time devoted to housework, the SHP
asks people, on average, how many hours they spend on housework , including washing, cooking, or cleaning, in a
normal week. But there is an instruction that this question does not count care of the children.
2 The original question—“To what extent are you satisfied with the way the housework is shared out—washing,
cooking, cleaning—within your household, if 0 means "not at all satisfied" and 10 "completely satisfied?"—has
been recoded in a three point scale.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 11
satisfied than women with at least one child (see Table 1).
Gender attitude. We computed a three point indicator which measures gender attitude
toward equality between men and women. This indicator is composed of three items: (1) whether
women think that they are in general penalized compared to men; (2) whether women think that
they are personally penalized; and (3) whether women are in favor of measures to promote
equality between men and women. This scale, which is a sum of these three items, starts from 0
(“not at all penalized”) and goes to 10 (“strongly penalized”). The internal consistency of this
scale is satisfactory: the Cronbach’s alpha rises from 0.65 to 0.72 among waves,3 which
corresponds to the widely accepted social science satisfactory cut-off for the internal consistency
of a scale.
Familistic attitude. Two items are taken into account:
The first one considers, on a three point scale—from 0 (“completely disagree”) to 2
(“completely agree”)—whether women think that having a job preserves independence.
The second one measures, on a three point scale—from 0 (“completely disagree”) to 2
(“completely agree”)—whether women think that a child suffers if it has a working mother.
Control and Intermediate Variables:
We also control for socio-demographic variables like women’s education (high, middle,
or low), occupation (full time, part time, in training, looking for a job, and being a housewife),
and age. The control for age deserves a comment: There exist normative beliefs about the
appropriate age for a woman to bear a child, in particular pertaining to upper age limits. As a
consequence, we expect a non linear effect for age (intentions increasing and then decreasing
with age). We therefore add a quadratic effect for the age.
3 The Cronbach’s alpha is a measure of a psychometric scale’s internal consistency.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 12
Social support. Finally, we control for social support. With two indicators we measure
the amount of social support received by women. One item measures whether women receive
practical support from relatives on a three point scale from 0 (“not at all”) to 2 (“a great deal”).
Moreover, we also have one item in order to know whether women receive emotional support
from relatives on a three point scale from 0 (“not at all”) to 2 (“a great deal”).4
# about here Tab. 1 #
Results
We have at least one declared fertility intention for each woman in our six-year yearly
panel, from 2002 to 2007. In most cases, the information on intentions covers multiple points in
time. We perform a variety of nested hierarchical 2-level models in which the lower level
represents the survey wave and the higher level the individual woman. The estimated models
allow us to disentangle interindividual measures (level of women) and intraindividual measures
(level of waves) (Hox, 2002; Singer & Willett, 2003). We estimate fixed effects for the intercept
and the different covariates as well as a random effect for the intercept. The hypothesis is that the
intercept varies for each woman, according to unknown characteristics, while there are no
variations in the effect between different covariates. This hypothesis of a sole random effect on
the intercept is often made in the case of multilevel logistic regressions. Models were estimated
using HLM software, version 6 (Bryk & Raudenbush, 1992). The chosen method of estimation is
full maximum likelihood.
4 In the analyses presented in this paper, we will not take into consideration involvement in religion since this
variable does not have any impact on intention to have a child within 24 months.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 13
Note, however, that we will distinguish in our analyses two subsamples: childless women
(739 observations), and women with at least one child (2319 observations). This distinction is
guided by the fact that, in Switzerland, the transition to parenthood often corresponds to a
moment in which couples become non-egalitarian in their practice and norms after a period of
being egalitarian until this transition (Le Goff et al., 2009; Levy et al., 1997).
Descriptive Results
Before showing the results of the HLM model, we propose a role-sets typology presented
below (Table 2), inspired by the typology proposed by Rizzi et al. (2008). Our descriptive
analysis reveals the difference between men and women with respect to involvement in
housework. Yet, fertility intentions may be affected not only by the amount of housework, but
also by the amount of paid labor a woman does. Consequently, by comparing work organization
both inside and outside the home (and following Rizzi et al., 2008), we distinguish different role-
sets according to the amount of work done in each domain. This alternative of combining
information from the amount of paid work and the amount of domestic work for each partner in
the couple leads us to nine different categories. These categories range from a traditional role-set
to an egalitarian role-set with a wide range of combinations. The traditional role-set describes a
situation in which women do more domestic work in comparison to their partner. At the same
time, they do less paid labor compared to their partners. The super woman cluster is
characterized by women who do more domestic work in comparison to their partner, whereas
these women have the same amount of paid labor work compared to their partners. The ultra
woman cluster comprises women who put in extra hours in both paid and domestic work. By
contrast, the egalitarian role-set pertains to couples where both partners have the same amount
of domestic and paid work. The super man cluster is defined by men who experiment with extra
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 14
hours in paid labor but have the same amount of domestic work compared to their partner. The
ultra man cluster is composed of men with extra hours in both domestic and paid labor, whereas,
post modern super men do extra hours in domestic work only. The reversed traditional role-set
describes women who do more paid and do less domestic hours compared to their partner.
Finally, the post modern super women do extra hours in paid labor but not in domestic work.
Typologies of Role-Sets in the SHP Data
Not very surprisingly, the most frequent role-set for people living in Switzerland is the
traditional one in which the man carries a heavier burden in paid labor and the woman in
domestic work. The prevalence of this traditional role-set increases with the number of children.
Couples adopt a traditional role-set after the transition to parenthood, and more so when they
have more than one child.
The second most important division of tasks cluster is the ultra women role-set, in which
women are expected to participate in the labor market and in domestic work as well. The women
in the ultra women set do more hours in both domestic and paid labor compared to their partners.
The third most important cluster is, as presented by the authors, the ultra man set. The
egalitarian role-set is quite marginally represented. Moreover, a majority of women in this group
do not have any children and work full time. Finally our descriptive statistics show that several
role-sets, such as the reversed and postmodern role-sets, which are in fact extremely marginally
represented, go beyond equality and represent new tendencies and, maybe, future challenges in
the division of tasks within couples.
# about here Tab. 2 #
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 15
Multilevel Analysis
We estimated different models for each of the two subsamples: (1) childless women and
(2) women with at least one child. In the first model, we introduced indicators related to
familistic attitudes. In models 2, 3, 4, and 5, we progressively introduce the variables measuring
the partners’ workload, as well as other control and intermediate variables.
Table 3 presents results for childless women. The first model shows us that women who
agree with the idea that jobs preserve independence are less likely to develop the intention to
have a child within the period (ß = -0.129 ; p < .01). However, being in favor of gender equality
has a positive impact on the intention to have a child within 24 months.
The second model shows that being satisfied with the way household work is shared is
positively related with the intention to have a child, as is the case with the situation where
women do more housework compared to their partner. For each covariate, estimated coefficient
is only weakly significant (10% level) and becomes non-significant in other models in which are
added supplementary covariates. As expected, the third model shows a quadratic effect of age.
Intention to have a child first increases with age but flattens down at higher age. The fourth
model shows that there are no differences between women working full time or part time in their
intention. Such an absence of differences could mean that intention to have a child is not related
to the economic level of women. Being in training or receiving an education is strongly
negatively related to the intention to have a child. This result is consistent with the classic result
of life course studies showing that women rarely give birth to a first child while undergoing
education (Blossfeld & Huinink, 1991). Looking for a job has, however, a positive effect on the
intention to have a child. The fifth model demonstrated that familistic attitude, socio-
demographic variables, and social support do not have any influence on the intention to have a
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 16
child. Moreover, the two variables measuring emotional and practical support do not improve the
model fit (the introduction of the two social support variables does not decrease the maximum of
the log likelihood).
# about here Tab. 3 #
In the case of women who already have at least one child, Table 4 shows that familistic
attitudes, gender opinion and division of tasks within a couple have an influence on the intention
to have an extra child. The first model indicates that women who agree that a child suffers
because his mother is working are less likely to have another child. But the significance
disappears when all covariates are introduced in the model. Being in favor of gender equality has
a positive impact on the intention to have a child within 24 months, this effect being more and
more significant as other covariates are added. Satisfaction with household task has a positive
impact on the intention to have a child within 24 months.
The second model presents the fact that women who do more household work than their
partners are less likely to have another child. The third model presents that intention to have a
child within the period depends on the age of the individual. As with childless women, the
intention to have a child first increases and then decreases with age. The fourth model underlines
that working full time is negatively related to the intention to have another child; as well, that a
low level of education is associated to no intentions to have a child. By contrast, women with a
high level of education develop intentions to have a child within 24 months more often than
women with a middle level. The last model indicates that practical support is negatively
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 17
related—but only at the level of 10%—to the intention to have a child, whereas benefiting from
emotional support is positively related to this intention.
# about here Tab. 4 #
Conclusion
The longitudinal approach we have adopted in the analysis of the household panel
produced a number of interesting results. As expected, intentions to have children are strongly
related to age, which shall yet be interpreted as an indicator for a given phase of the life course.
Both for childless women and for mothers there is a moment in the life course during which the
intention to have a child seems to be more likely to appear. Three phases can then be described
beginning, first, with a period of their lives during which women do not want (yet) to have a
child, especially when they are in education or professional training. In a second phase, intending
to have a child becomes much more common; but, once this moment has passed, childbearing
intention weakens, independently of whether the woman has or not already given birth. There
seems to be a normative window for the appropriate childbearing timing—“not too young” and
”not too old”—which is relatively stable in Switzerland. Sauvain-Dugerdil (2005) on the basis of
Swiss FFS data collected in the 1990s5 found that among those women who did not intend to
have a child, the youngest often mentioned reasons related to insufficient housing conditions or
to the difficult conciliation between professional and family life, while the eldest, though still in
their reproductive years, mentioned reasons related to their age.
5 The Fertility and Family Survey was commissioned by the Swiss Federal Statistical Office in the early 1990s. This
survey enables Switzerland to take part in the international Fertility and Family Survey (FFS) project launched by
the United Nations Economic Commission for Europe.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 18
The separate analyses for childless women and mothers were useful to show that
childbearing intention depends partially on different factors in the two groups. Couples are likely
to adopt a more traditional role-set after the transition to parenthood in Switzerland (Le Goff et
al., 2009; Levy et al., 1997). This process is even reinforced when they have more than one
child. It is then likely that couples in a more traditional role-set do not intend another child
because they have already achieved their desired family size.
On the contrary, in the case of childless women, a positive intention to have a child
within two years is more influenced by holding egalitarian values than by actual conditions
(except for undergoing education). This is consistent with results in other contexts where an
unequal share of domestic tasks especially seems to associate with lower intentions to have a
child (Cooke, 2003; Mills et al., 2008). However, women who consider labor market
employment as synonymous with independence are less prone to have children. In a context
where conciliation of employment and family responsibilities is not easy given the poor child
care services, as is the case in Switzerland (Branger et al., 2008), childbearing is probably and
understandably experienced as a threat to independence. The insistence on the unique mother
role in children’s development and education, of the conservative parties and their resistance to
the development of alternative systems to mother’s care, seems to depress childbearing
intentions and contribute to fertility decline and higher ages at the transition to parenthood.
In sum, we are able to show with some confidence that more egalitarian couples are more
likely to intend a child, other things being equal, confirming to a certain extent McDonald’s
hypothesis (2000) at the micro level. When the couple is more gender equal in the private sphere
within a context in which gender equity is promoted in the public sphere as well, as in
Switzerland, then the effects on fertility intentions are positive. Both having egalitarian roles and
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 19
satisfaction with the division of household tasks work towards producing a higher likelihood to
intend to have a child. Important codeterminants of such intentions are, however, equality in time
devoted by partners to paid employment, the possibility to benefit of emotional support.
However, other factors play a substantial independent role. In the first place, the aspiration to
independence of some women is one important factor which is not necessarily related to holding
egalitarian values or to a more equal workload share. Then, the economic situation of the couple,
especially in the presence of one or more children, heavily condition intentions.
Our analysis has been limited to fertility intentions as this is a powerful predictor of
fertility behavior (Philipov et al., 2009). Our next step is to investigate the possibility of changes
in fertility intentions during the life course. In particular, we are interested in identifying the
causes for changes in those cases in which a woman of a given parity switches from intending to
bear a child to not intending it any longer, sometimes later, or vice versa. Do they depend on
changes in attitudes, on changes in material conditions and activities, or on changes in a couple’s
role-set? We started exploring our longitudinal data in this direction; however, in order to have
sufficient fertility intentions we need to take into account more waves of SHP. Given the
existence of a fertility intention-behavior gap, there exists a second promising venue of
investigation: We can study the relationship between fertility intentions and their subsequent
realization in a longitudinal perspective. The identification of the causes for the fertility gap is
currently one of the most debated empirical issues in the demography of the family (Philipov et
al., 2009). For our interests, the question becomes whether a couple’s role-set is not only relevant
for intentions but also plays a role as determinant of the probability that women will bear an
intended child.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 20
Bibliography
Anxo, D., Fagan, C., Cebrian, I., & Moreno, G. (2005). Patterns of labour market integration in
Europe—a life course perspective on time policies. Socio-Economic Review, 5, 233-260.
Baxter, J. (2002). Patterns of change and stability in the gender division of household labor in
Australia, 1996-1997. Journal of Sociology, 38, 399-424.
Benin, M. H., & Agostinelli, J. (1988). Husbands' and wives' satisfaction with the division of
labor. Journal of Marriage and the Family, 50, 349-361.
Bianchi, S. M., Robinson, J. P., & Milkie, M. A. (2006). Changing rhythms of American family
life. New York: American Sociological Association.
Bittman, M. (1999). Now that the future has arrived: A retrospective of Gershuny’s theory of
social innovation. Paper presented at the Social Policy Research Centre.
Bittman, M., & Matheson, G. (1996). All else confusion: What time use surveys show about
changes in gender equity. Sydney: University of N.S.W.
Bittman, M., & Wajcman, J. (2000). The rush hour: The character of leisure time and gender
equity. Social Forces, 79(1), 165-189.
Blain, J. (1994). Discourse of agency and domestic labor. Journal of Family Issues, 15(4), 515-
549.
Blossfeld, H.-P., & Huinink, J. (1991). Human capital investments or norms of role transition?
How women's schooling affects the process of family formation. American Journal of
Sociology, 97, 143-168.
Branger, K., Crettaz, E., Oetliker, U., Robatti Mancini, V., Rochat, S., Roulet, F., et al. (2008).
Les familles en Suisse. Neuchâtel: Office fédéral de la statistique.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 21
Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data
analysis methods. Newbury Park: Sage Publications.
Chesnais, J.-C. (1998). Below-replacement fertility in the European Union (EU-15): Facts and
policies 1960-1997. Review of Population and Social Policy, 7, 83-101.
Coenen-Huther, J. (2005). Le souhait d'enfant. Un idéal situé. In J.-M. Le Goff, C. Sauvain-
Dugerdil, C. Rossier, & J. Coenen-Huther (Eds.), Maternité et parcours de vie (pp. 85-
136). Berne: Peter Lang.
Coltrane, S. (2000). Research on household labor: Modeling and measuring the social
embeddedness of routine family work. Journal of Marriage and the Family, 62(4), 1208-
1233.
Cooke, L. P. (2003). The South revisited: The division of labor and family outcomes in Italy and
Spain. Luxembourg: IRISS Working Paper Series.
Greenstein, T. N. (2000). Economic dependence, gender, and the division of labor in the home:
A replication and extension. Journal of Marriage and the Family, 62(2), 322-335.
Harrington, M. (1998). The care equation. The American Prospect, 9, 4-94.
Henchoz, C., & Wernli, B. (to be published). Cycle de vie et travaux ménagers en Suisse.
L’investissement ménager des hommes et des femmes lors de différentes étapes de la
construction de la famille. Revue suisse de sociologie: Numéro special.
Hochschild, A., & Machung, A. (1989). The second shift. New York: Viking.
Hox, J. J. (2002). Multilevel analysis. Techniques and applications. Mahwah, New Jersey:
Lawrence Erlbaum Associates.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 22
Krüger, H., & Levy, R. (2000). Masterstatus, Familie und Geschlecht. Vergessene
Verknüpfungslogiken zwischen Institutionen des Lebenslaufs. Berliner Journal für
Soziologie, 3, 379-401.
Krüger, H., & Levy, R. (2001). Linking life course, work, and the family: Theorizing a not
visible nexus between women and men. Canadian Journal of Sociology, 26(2), 145-166.
Le Goff, J.-M., Levy, R., Sapin, M., & Camenisch, M. (2009). Devenir parent. Changer de vie.
In M. Oris, E. Widmer, A. de Ribaupierre, D. Joye, D. Spini, G. Labouvie-Vief, & J.-M.
Falter (Eds.), Transition dans le parcours de vie et construction des inégalités (pp. 233-
252). Lausanne: Presses Polytechniques et Universitaires Romandes.
Le Goff, J.-M., Sauvain-Dugerdil, C., Rossier, C., & Coenen-Huther, J. (2005). Maternité et
parcours de vie. Berne: Peter Lang.
Levy R. (2006), Particulière, singulière ou ordinaire? La régulation suisse des parcours de vie
sexués. In T. Eberle & K. Imhof (Eds.), Sonderfall Schweiz. Zurich: Seismo. (pp. 224-
226)
Levy, R., Gauthier, J.-A., & Widmer, E. (2006). Entre contraintes institutionnelles et
domestiques : Les parcours de vie masculins et féminins en Suisse. Canadian Journal of
Sociology/Cahiers canadiens de sociologie, 31(4), 461-489.
Levy, R., Joye, D., Guye, O., & Kaufmann, V. (1997). Tous égaux? De la stratification aux
représentations. Zürich: Seismo.
McDonald, P. (2000). Gender equity, social institutions and the future of fertility. Journal of
Population Research, 17, 1-16.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 23
Meissner, M., Humphreys, E., Meis, S., & Scheu, W. (1975). No exit for wives: Sexual division
of labor and the cumulation of household demands. Canadian Review of Society and
Anthropology, 12, 424-439.
Miller, W. B., & Pasta, S. L. J. (1995). Behavioral intentions: Which ones predict fertility
behavior in married couples? Journal of Applied Social Psychology, 25, 530-555.
Mills, M., Mencarini, L., Tanturri, M. L., & Begall, K. (2008). Gender equity and fertility
intentions in Italy and the Netherlands. Demographic Research, 18, 1-26.
Olah, L. S. (2003). Gendering fertility: Second births in Sweden and Hungary. Population
Research and Policy Review, 22(2), 171-200.
Pahl, R. E. (1984). Divisions of Labor. Oxford.
Philipov, D., Thévenon, O., Klobas, J., Bernardi, L., & Liefbroer, A. (2009). Reproductive-
decision making in a macro-micro perspective (REPRO). Paper presented at the
European Demographic Research Papers.
Rizzi, E., Judd, M., White, M., Bernardi, L., & Kertzer, D. (2008). Familistic attitudes, dual
burden and fertility intentions in Italy. Paper presented at the European Population
Conference.
Robinson, J. P., & Godbey, G. (1997). Time for life: The surprising ways Americans use their
time. State College, PA: Penn State Press.
Sauvain-Dugerdil, C. (2005). La place de l'enfant dans les projets de vie: Temporalité et
ambivalence. In J.-M. Le Goff, C. Sauvain-Dugerdil, C. Rossier, & J. Coenen-Huther
(Eds.), Maternité et parcours de vie (pp. 281-318). Berne: Peter Lang.
Shaevitz, M., Hansen. (1984). The superwoman syndrome. Boston.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 24
Shelton, B. A. (1992). Women, men and time: Gender differences in paid work, housework and
leisure. New York: Greenwood Press.
Shelton, B. A., & Firestone, J. (1989). Household labor time and the gender gap in earnings.
Gender & Society, 3(1), 105-112.
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and
event occurrence. Oxford: University Press.
Tazi-Preve, I. M., Bichlbauer, D., & Goujon, A. (2004). Gender trouble and its impact on
fertility intentions. Yearbook of Population Research in Finland, 40, 5-24.
Torr, B., Miller W. B., & Short, S., E. (2004). Second births and the second shift: A research
note on gender equity and fertility. Population and Development Review, 30(1), 109-130.
Widmer, E., & Ritschard, G. (2009). The de-standardization of the life course: Are men and
women equal? Advances in Life Course Research, 14(2), 28-39.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 25
Table 1: Sample characteristics (women aged between 18 to 45 years old, in couple)
WOMEN AGED BETWEEN 18 TO 45 YEARS OLD IN COUPLE
2,513 WOMEN WERE INTERVIEWED AT LEAST
ONCE
VARIABLE S
3,058 OBSERVATIONS 739 OBSERVATIONS
CHILDLESS WOMEN
2,318 OBSERVATIONS
WOMEN WITH AT LEAST
ONE CHILD
DEPENDENT VARIABLE
Intentions to have a child in next 24 months
No 76.4 % 40.6 % 83.3 %
Don’t know 2.5 % 4.5 % 1.9 %
Yes 21.1 % 54.9 % 14.8 %
PARTNERS ’ WORKLOAD VARIABLES
Satisfaction with organization of domestic
work
No - Little 22.4 % 16.9 % 24.2 %
Somewhat 39.5 % 36.4 5 % 40.6 %
Yes-Very satisfied 38 % 46.6 % 35.2 %
Women domestic hours 17.98 hours / week 8.78 hours / week 20.97 hours / week
Men domestic hours per week 5.71 hours / week 5.12 hours / week 5.90 hours / week
Women labor hours 25.87 hours / week 37.54 hours / week 20.88 hours / week
Men labor hours 44.95 hours / week 44.30 hours / week 45.15 hours / week
CONTROL AND INTERMEDIATE VARIABLES
Women’s age groups
Less than 30 years old 15.3 % 41.5 % 6.9 %
30-34 years old 24.0 % 28.0 % 22.8 %
35-39 years old 33.4 % 16.4 % 38.9 %
40-44 years old 27.2 % 14.1 % 31.4 %
Women’s education
Low education 7.3 % 4.1 % 8.4 %
Middle education 70.3 % 62.5 % 72.9 %
High education 22.3 % 33.4 % 18.8 %
Women’s occupational status
Occupied full time 19.9 % 60.7 % 7 %
Occupied part time 49.7 % 29.7 % 56 %
Housewife 28 % 2.5 % 36.1 %
Looking for a job 1 % 1.4 % 0.8 %
Training 1.5 % 5.7 % 0.2 %
Practical support
Not at all 15.3 % 11.5 % 16.5 %
A little 45.4 % 40.9 % 46.9 %
A great deal 39.2 % 47.6 % 36.6 %
Emotional support
Not at all 8.8 % 6.5 % 9.5 %
A little 44.3 % 39.9 % 45.7 %
A great deal 46.9 % 53.5 % 44.8 %
Number of children younger than 17 living
in the household
0 24.2 % 100 %
1 18.7 % 24.6 %
2 37.6 % 49.6 %
3 15.9 % 20.9 %
4 and more 3.6 % 29.4 %
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 26
Table 2: Sample characteristics in function of the workload and household rate
Household Workload
Number of
observations
Intention to have a child
within 24 months
(Valid percent)
Number of kids
(Valid percent)
Women Education
(Valid percent)
Women Occupation
(Valid percent)
Traditional
role-set W > M W < M 2047
17.3 yes
2.3 do not know
80.4 no
13.7 no kids
19.4 one kid
43.4 two kids
23.4 more than three kids
6.6 low
75.3 middle
18 high
7.3 full time
55 part time
36.2 housewife
1.6 training / jobless
Egalitarian
role-set W = M W = M 14
21.4 yes
7.1 do not know
71.4 no
64.3 no kids
7.1 one kid
21.4 two kids
7.1 more than three kids
78.6 middle
21.4 high
64.3 full time
28.6 part time
7.1 housewife
Ultra women W > M W > M 346
31.5 yes
3.5 do not know
65 no
50.9 no kids
15.9 one kid
22.32 two kids
11.0 more than three
8.4 low
63 middle
28.6 high
56.4 full time
35.8 part time
6.1 housewife
1.8 training / jobless
Super women W > M W = M 104
30.8 yes
1.9 do not know
67.3 no
39.4 no kids
22.1 one kid
22.1 two kids
17.4 more than three kids
18.3 low
58.7 middle
23.1 high
43.7 full time
19.4 part time
33 housewife
3.8 training / jobless
Post modern
superwomen W = M W > M 37
43.2 yes
2.7 do not know
54.1 no
86.5 no kids
5.4 one kid
8.1 two kids
2.7 low
51.4 middle
45.9 high
67.6 full time
27 part time
5.4 training
Reversed
traditional
role-set
W < M W > M 161
23.6 yes
4.3 do not know
72 no
55.3 no kids
16.1 one kid
17.4 two kids
11.2 more than three kids
5.6 low
59 middle
35.4 high
60.9 full time
32.9 part time
0.6 housewife
5.6 training
Post modern
super men W < M W = M 34 29.4 yes
70.6 no
52.9 no kids
20.6 one kid
17.6 two kids
8.8 more than three kids
17.6 low
38.2 middle
44.1 high
58.8 full time
23.5 part time
2.9 housewife
14.7 training / jobless
Super men W = M W < M 77
37.7 yes
1.3 do not know
61 no
29.9 no kids
19.5 one kid
32.5 two kids
18.2 more than three kids
3.9 low
58.4 middle
37.7 high
18.7 full time
54.7 part time
20 housewife
6.7 training / jobless
Ultra man W < M W < M 234
23.1 yes
2.6 do not know
74.4 no
29.9 no kids
18.8 one kid
39.7 two kids
11.5 more than three
9 low
61.5 middle
29.5 high
22.1 full time
55.8 part time
17.3 housewife
4.8 training / jobless
Note. W=women; M=men.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 27
Table 3: Childless Women: Results of logit models; unit-specific model; coefficient and
odds ratio (round bracket).
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept -0.370
(0.691)
-0.540
(0.582)
-1.022
0.359
-0.981
(0.375)
-2.333*
(0.097)
Job preserve
independence
-0.129**
(0.878)
-0.125*
(0.882)
-0.114*
(0.892)
-0.109*
(0.897)
-0.119*
(0.887)
Child suffers with working
mother
-0.030
(0.970)
-0.037
(0.963)
-0.012
(0.988)
-0.021
(0.979)
-0.018
(0.982)
Gender opinion 0.138**
(1.150)
0.141*
(1.151)
0.135*
(1.145)
0.150**
(1.161)
0.153**
(1.165)
Satisfaction with
household tasks
0.090
(1.095)
0.106+
(1.111)
0.119+
(1.126)
0.120+
(1.127)
0.103
(1.108)
Household rate 0.150+
(1.162)
0.188+
(1.206)
0.157
(1.170)
0.153
(1.165)
Workload rate -0.029
(0.971)
0.019
(1.018)
0.031
(1.032)
0.039
(1.039)
Age 1.918***
(6.810)
1.879***
(6.544)
1.851***
(6.368)
Age square -0.030***
(0.970)
-0.030***
(0.974)
-0.029***
(0.971)
Occupation full time / ref.
part time 0.005
(1.005)
-0.019
(0.981)
Occupation housewife /
ref. part time 0.767
(2.154)
0.700
(2.014)
Occupation training / ref.
part time -3.365**
(0.034)
-3.562**
(0.028)
Occupation looking for a
job / ref. part time 1.889**
(6.610)
1.829
(6.230)
Education low /
ref. middle 0.032
(0.968)
0.167
(1.181)
Education high /
ref. middle -0.145
(0.865)
-0.122
(0.885)
Practical support 0.057
(1.059)
Emotional support 0.126
(1.134)
Random effect
Standard deviation
Variance component
Chi-square
1.504***
2.262
651.599
1.515***
2.295
652.050
1.425***
2.032
550.628
1.416***
2.005
528.381
1.992***
1.411
524.442
Log-Likelihood -963.027 -962.425 -956.264 -940.669 -941.696
Note. + p < .01 ;* p < .05 ; ** p < .01 ; *** p < .001. Mode of estimates: full maximum
likelihood. N = 377; observations = 706.
GENDER, FAMILY, AND FERTILITY IN SWITZERLAND 28
Table 4: Women with at least one child: Childless Women: Results of logit models; unit-
specific model; coefficient and odds ratio (round bracket).
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept -2.78***
(0.062)
-2.577***
(0.075)
-3.043***
(0.048)
-3.079***
(0.045)
-3.417***
(0.033)
Job preserve
independence
-0.002
(0.997)
-0.009
(0.990)
-0.009
(0.990)
-0.014
(0.986)
-0.016
(0.984)
Child suffers with working
mother
-0.049*
(0.951)
-0.039+
(0.961)
-0.062*
(0.940)
-0.041
(0.959)
-0.041
(0.959)
Gender opinion 0.066+
(1.069)
0.062+
(1.064)
0.098*
(1.102)
0.077*
(1.080)
0.079*
(1.082)
Satisfaction with
household task
0.120**
(1.127)
0.122**
(1.129)
0.103*
(1.108)
0.095*
(1.100)
0.090*
(1.095)
Household rate -0.057
(0.944)
-0.030
(0.970)
-0.018
(0.982)
-0.020
(0.980)
Workload rate -0.046+
(0.954)
-0.052+
(0.949)
-0.070*
(0.932)
-0.068*
(0.933)
Age 0.731**
(2.079)
0.642**
(1.901)
0.654**
(1.924)
Age square -0.015***
(0.985)
-0.014***
(0.986)
-0.014***
(0.986)
Occupation full time / ref.
part time -0.634+
(0.530)
-0.664+
(0.515)
Occupation housewife /
ref. part time 0.110
(1.117)
0.075
(1.077)
Occupation training / ref.
part time 1.633
(5.120)
1.606
(4.981)
Occupation looking for a
job / ref. part time -0.497
(0.608)
-0.617
(0.539)
Education low /
ref. middle -0.742+
(0.476)
-0.736+
(0.479)
Education high /
ref. middle 0.901***
(2.463)
0.924***
(2.520)
Practical support -0.075+
(0.927)
Emotional support 0.116*
(1.122)
Random effect
Standard deviation
Variance component
Chi-square
1.387***
1.924
1229.511
1.405***
1.974
1239.992
1.374***
1.887
1011.647
1.378***
1.899
997.184
1.383***
1.913
999.023
Log-Likelihood -873.054 -870.934 -769.971 -757.194 -755.925
Note. + p < .01 ;* p < .05 ; ** p < .01 ; *** p < .001. Mode of estimates: full maximum
likelihood. N = 858; observations = 2197.
