Divorce and Adult Children’s Perceptions of Family Obligations

Divorce and Adult Children’s Perceptions of
Family Obligations
Belinda Wijckmans & Jan Van Bavel
Belinda Wijckmans
Interface Demography, Vrije Universiteit Brussel
Pleinlaan 2
BE-1050 Brussel
Belgium
Tel: +226 75 53 46 20
Email: bwijckma@gmail.com
Jan Van Bavel
FaPOS Leuven / Family and Population Studies Leuven
Faculty of Social Sciences, University of Leuven
Parkstraat 45 bus 3601
BE-3000 Leuven
Belgium
Tel : +32 16 32 30 48
Email : Jan.VanBavel@soc.kuleuven.be
ABSTRACT - Research has indicated that divorce often weakens the ties between adult children and their parents,
affecting the exchange of support in both directions. Most studies of this issue have focused on actual transfers between
generations. Much less work has been done about the associated cultural values, norms, and attitudes. Using data from
the Netherlands Kinship Panel Study (NKPS) we investigated how divorce and re-partnering in either the parent’s or
adult child’s generation are related to filial, parental and general family obligations. Contrary to what might be
expected, the results show that divorce is positively associated with norms of family obligations, even after controlling
for the exchange of support between the generations and the perceived quality of the relationship. Furthermore, there is
a link between the type of family obligation and the generation in which the divorce occurred: the respondent’s own
divorce history is more strongly and positively correlated to his or her attitudes towards filial obligations, whereas the
parents’ divorce history is only positively related to the respondent’s parental obligations.
KEY WORDS: Intergenerational solidarity, family obligations, intergenerational relations, divorce, remarriage,
relationship characteristics, the Netherlands
Paper accepted for publication in Journal of Comparative Family Studies (forthcoming in 2013).
1
Divorce and Adult Children’s Perceptions of Family Obligations
INTRODUCTION
Family members are an important source of informal support to each other (Brandt, Haberkern and Szydlik, 2009;
Grundy, 2008). Many parents continue to play a supportive role in the lives of their adult children after they have left
the parental home and started families themselves. Conversely, many adult children provide a degree of care and
assistance to their ageing parents (Grundy and Henretta, 2006). At the same time, family relationships have become
increasingly complex due to rising divorce rates and subsequent (step)family formation, potentially threatening
intergenerational exchange of family support (De Graaf and Kalmijn, 2006).
On the one hand, research has indicated that divorce, either in the parent or the adult child generation, tends to weaken
ties between adult children and their parents affecting the exchange of support. That is, support between generations
seems to be increasingly defined by particular individual circumstances and dependant on the quality of the relationship
(Coleman, Ganong and Cable, 1997; Ganong and Coleman, 2006; Hilton and Kopera-Frye, 2007; Lye, 1996; Pezzin
and Steinberg Schone, 1999; Van Gaalen and Dykstra, 2006). On the other hand, scholars have suggested that some
family connections may represent a latent kind of social support network that may remain inactive over long stretches
of time when all is going well but that may be activated when a family crisis occurs (Bengtson 2001). Seen from this
perspective, divorce may put the nuclear family under pressure while at the same time activating wider intergenerational
family relations.
This paper focuses on attitudes about family obligations after divorce rather than on actual transfers of support within
families. Most research to date has focused on the demography of intergenerational relationships or on behaviours like
exchange of resources. Normative and attitudinal dimensions of intergenerational relations merit equal scholarly
attention, however (Rossi and Rossi, 1990; Bengtson 2001), for several reasons. Studying attitudes and norms
underpinning intergenerational support can give important insights into behaviour and may also help to explain how
feelings of mutual responsibility are distributed within families (Ganong and Coleman, 1999; Ikkink, van Tilburg and
Knipscheer, 1999). The distribution of such feelings will affect which nodes in the latent matrix of kin connections
(Bengtson 2001) may be activated in times of family crisis. From a policy perspective, studying attitudes will give
insight into the extent to which policy measures do or do not match expectations in the population. It may give clues
about how to connect policy measures with stated preferences: the relationship between attitudes and expectations on
the one hand and the actual state of affairs on the other hand will affect to what extent people will feel satisfied with
proposed policy measures (Van Bavel et al., 2010).
Using the Netherlands Kinship Panel Study (Dykstra et al., 2005) we examine the relationship between attitudes about
family obligations and divorce and re-partnering in either the parent or adult child generation from the adult child’s
perspective. How are the parents’ or adult child's divorce and re-partnering related to the adult child’s attitudes towards
family obligations? Does the association, if any, between divorce and re-partnering on the one hand and family
obligations on the other hand persist after controlling for the quality of the relationship and the degree of support
exchanged between parents and their adult children?
2
DIVORCE AND FAMILY OBLIGATIONS
Family obligations are culturally prescribed normative expectations based on kinship (Gans and Silverstein, 2006;
Ganong and Colman, 1999; Ganong and Coleman, 2005; Lye, 1996; Rossi and Rossi, 1990; Stein et al., 1998). Most
research on the effect of divorce and remarriage on family support and obligations draws on exchange theory. Actual
exchange of support and normative obligations are seen as mutually reinforcing: parents are normatively expected to
take care of their children and in return the children, when adult, feel obliged to reciprocate this care by helping their
elderly parents (Ganong et al., 1998; Ganong and Coleman, 2006; Stein et al., 1998, Ribar and Wilhelm, 2006).
There is no agreement in the literature on the relationship between divorce and family obligations. This may be partly
due to the different frameworks employed by researchers. Firstly, there is diversity in defining and conceptualizing
family obligations. Family obligations can be defined either as generalized norms about responsibilities between
generations based on kinship or alternatively as individual beliefs and attitudes about responsibilities toward specific
family members (Ganong and Coleman, 2005; Finch and Mason, 1991). In general, intergenerational research has not
drawn clear lines between these approaches. Researchers often use different terms for the same concepts (e.g.,
intergenerational obligations, expectations, responsibilities, normative solidarity, etc.) or apply measures designed for
general norms to the examination of personal obligations and attitudes or vice versa (Ganong and Coleman, 2005).
Moreover, some studies also include societal level intergenerational attitudes, focusing on the macro-social
relationships across cohorts rather than on the generational ties within families (Leeson, 2005; Logan and Spitze, 1995;
Ward, 2001).
Secondly, it also depends on which viewpoint is adopted - the parent’s or the adult child’s - and the marital transitions
of which generation one focuses on. For example, as Stuifbergen and colleagues (2008) point out, from the adult child’s
perspective a parental divorce might lead to more pressure, having to divide time and support between two households
instead of one, resulting in less support given to each individual parent. From the parent’s point of view a divorce might
indeed lead to receiving less support from each of their adult children but also to a wider support network through repartnering and higher order fertility. Also, adult children of divorced parents are often perceived as receiving less
support from their parents compared to children of the never-divorced. However, when support exchanges between
households are considered (i.e., when the support of both divorced parents’ individual households is combined and
compared to the support given from the single household of the still-married), the differences seem to disappear (Amato
et al., 1995). Few studies have compared the different perspectives of parents and their adult children (Mandemakers
and Dykstra, 2008). Finally, marital status is often not the primary focus of the studies involved or it is reduced to
specific group comparisons, for example between the ever- and never-divorced or between the ever- and never-married
(Liefbroer and Mulder, 2006). The findings therefore are often difficult to compare.
This study will focus on one dimension of the conceptual framework of family solidarity presented by Bengtson (2001),
namely the dimension of what he calls “normative solidarity”. This dimension, which we designate as “family
obligations” for convenience, includes “expectations regarding filial obligations and parental obligations, as well as
norms about the importance of familistic values” (Bengtson, 2001: 8). We look at attitudes held by adults and
distinguish between attitudes about obligations towards ageing parents (filial obligations), attitudes about obligations of
ageing parents towards adult children (parental obligations) as well as obligations towards family in general (familism,
cf. Bengtson 1991; Silverstein et al. 2006). We will look at the implications of a divorce of the parents of the
3
investigated adults (“parental divorce”) as well as the implications of divorce of these adults themselves (“adult child’s
divorce” below) for each of these three types of family obligations.
Family obligations and support following parental divorce
Numerous studies about the influence of parental divorce on intergenerational relationships have been conducted with
the main focus on filial obligations towards ageing parents. The decline of involvement in the children’s lives by their
parents after divorce has been repeatedly documented to affect the parent-child’s relationship negatively, even when
they are co-resident (Amato and Booth, 1994; Cooney and Uhlenberg, 1990; Furstenberg, Hoffman and Shrestha, 1995;
Lin, 2008). The reduction of parent-child contact or strained relationships in (early) childhood, or even when the child
has reached adulthood, might lead to a reduced tendency to help ageing parents later, fathers often being more
disadvantaged than mothers (Aquilino, 1994; Cooney, 1994; Daatland, 2007; De Graaf and Fokkema, 2007; Ganong
and Coleman, 2006; Gans and Silverstein, 2006; Kalmijn, 2007; Lawton, Silverstein and Bengtson, 1994; Tomassini,
Glaser and Stuchbury, 2007). A new partner might alleviate some (time and financial) constraints, but might also bring
new challenges. Research on new partnership formation has yielded ambiguous results (Amato, Rezac and Booth, 1995;
Cooney and Uhlenberg, 1990; Ganong and Coleman, 2005; Pezzin and Steinberg Schone, 1999). The gender of both
parent and adult child, and the custody arrangement during childhood may be important factors in determining the
support exchanged in later life.
However, not all studies found a relationship between parental divorce and adult children’s attitudes towards supporting
the older generation (e.g., Gans and Silverstein, 2006), nor do all studies agree on a general decline of support
exchanged between parents and children following parental divorce (Amato, Rezac and Booth, 1995). Ganong and
colleagues (1998; Ganong and Coleman, 1999) found that although people tend to agree with the general notion of filial
obligation, irrespective of whether or not parents are divorced, when such obligations are reformulated into specific
tasks there is much less consensus on what is obligatory. Ongoing contact and closeness after the divorce seem to be
important precursors of filial obligations.
Little is known about how parental obligations are affected by the divorce of parents. Some studies examine the
continuing responsibilities of parents towards younger children after divorce, including financial child support
(obligations) and custody arrangements (Ganong and Coleman, 1999). Other studies focus on the actual transfers of
support between divorced parents and their adult offspring (cf. supra). It is often suggested that divorced parents give
less support to their adult children, but there is little detail available as to in what circumstances they feel less obliged to
help, or to what extent adult children expect help from their divorced parents.
Family obligations and support following an adult child’s divorce
Research about the consequences of adult children’s divorce has focused mainly on the support received from parents
and has yielded mixed results. Parents may be an important source of support in times of crisis. Some studies found that
divorced children received larger amounts of support from their parents compared with still-married adult children
(Dykstra, 1997; Sarkisian and Gerstel, 2008). In contrast, other studies suggest that divorced children perceived
receiving less support from their parents, and reported higher levels of strain in the relationship than married children
(Umberson, 1992).
The gender of the divorcing child and the presence of (young) children are believed to be important in the post-divorce
relationship with parents. For example, Kaufman and Uhlenberg (1998) found that a daughter’s divorce has a strong
4
negative effect on the relationship with her parents, while there was no such effect for a divorced son. Yet, other studies
came to very different, even opposite conclusions (Spitze et al., 1994). The presence of young children has been linked
to reinforced reciprocal support (Spitze and Logan, 1991; Spitze et al., 1994). Still, when lone custodial mothers and
fathers were compared in a study in the US, lone divorced mothers reported the lowest relationship quality with their
parents and in-laws, and lower overall support levels than lone fathers or married mothers and fathers. It appeared that a
divorced mother's relationship with her parents contained more conflict and ambivalence perhaps as a result of higher
levels of contact between them (Hilton and Kopera-Frye, 2007).
For the support provided to parents or the norms and attitudes towards parental care, it seems that there is little or no
relationship with the marital status of adult children, or that it might be very short-lived (Gans and Silverstein, 2006;
Killian and Ganong, 2002; Lye, 1997; Logan and Spitze, 1995; Stuifbergen, Van Delden and Dykstra, 2008). However,
suggestions have been made that divorced children might be less supportive to their parents. In general, they have
reduced resources and may in fact need support themselves, potentially diminishing their sensitivity to the need of their
parents (Connidis, 2001; Ganong and Coleman, 1999). Hence, an adult child’s divorce has been associated with weaker
filial obligations.
Research questions
We discern three perspectives towards the implications of divorce for family obligations. The first and most dominant
perspective is that divorce weakens norms of family obligation. Divorce may complicate family relationships,
especially when new partners, in-laws and stepchildren become involved, making the existing norms less clear-cut and
difficult to apply to the changed family structure (Coleman, Ganong and Cable, 1997 ; Ganong and Coleman, 1999,
2006). In contrast, some scholars adhering to the continuity perspective, argue that families and the norms existing
within them are resilient to changes in the family structure or the marital status of its members (Bengtson, 2001;
Ganong and Coleman, 1999; Hans, Ganong and Coleman, 2009; Rossi and Rossi, 1990; Spitze et al., 1994). From this
second perspective, divorce would not affect existing feelings of family obligation. A third perspective is that divorce
might enhance rather than weaken feelings of family obligations. This hypothesis is unconventional and has hardly been
encountered in the existing literature. However, it can be founded on the theory that family connections, particularly
filial and (grand)parental relationships, make up a latent support network that is activated especially in times of crisis
(Bengtson 2001).
So our basic research question is the following: is divorce associated with weaker feelings of family obligations, as can
be expected based on the dominant perspective in the literature, or does it rather stimulate stronger feelings of family
obligations, or is there no association at all?
This research question is relevant for filial obligations and parental obligations as well as general family obligations.
However, we expect that the strongest connection between divorce and family obligations showing up in our data, if
any, will relate to filial obligations since we use survey data gathered from the middle generation. Filial obligations
express what adult children should do for their ageing parents, so they are directly relevant for the lives of middle
generation adults and most likely to be affected by their own specific circumstances.
Finally, we explore the role of repartnering: does repartnering mitigate or enhance the effects of divorce, if any? We
have no strong theoretical arguments to expect a positive or a negative association. On the one hand, a new partner
might lead to more tension and ambivalence in the family, resulting in weaker feelings of obligation. On the other hand,
5
finding a new partner might ease time and financial constraints, allowing (previous) patterns of mutual support to
persist and leading to stronger norms of obligation.
DATA AND METHOD
The data used in our analyses come from the first wave of the Netherlands’ Kinship Panel Study (NKPS). The NKPS is
a large scale survey among more than 8000 individuals, aged 18 to 79 years (Dykstra et al., 2005; see also Merz et al.,
2009). The focus of this contribution is on the adult child or simply “the respondent” below. We select a subsample
containing respondents whose parents have ever been married and at least one of them is still alive and not living in the
respondent’s household. Parents and adult children who are cohabiting are a small, exceptional subgroup in the
Netherlands, with a particular type of contact, relationship quality and exchange of support, and potentially reflecting
specific attitudinal outcomes. The questions we use about the exchange of emotional and instrumental support were not
asked to them. Therefore we decided to exclude them from our analyses. This amounts to a total of 4.305 cases. Sample
design weights (Dykstra et al., 2005) were applied in all analyses, which were run separately for each type of obligation.
Measurement of family obligations
We applied exploratory factor analysis to 11 items concerning family obligations. This yielded three clearly
distinguishable factors: general family obligations, filial or upward obligations and parental or downward obligations.
Table 1 gives the full list of items and their factor loadings. We use ordinary least squares regression analysis to explain
the variance of the standardized factor scores for each of the three types of family obligations.
[Table 1 about here]
Respondent’s and parent’s divorce history
Our central explanatory variables consist of the divorce and subsequent re-partnering histories of both the respondents
and their parents. In contrast to some other studies we only consider legal divorce after formal marriage. Since the goal
of this study is to examine attitudinal outcomes we believe that separation from an out-of-wedlock union might reflect
different attitudinal dynamics. For the respondents we distinguish between six categories: (a) still in a first marriage; (b)
never married, currently cohabiting; (c) never married, currently living alone; (d) divorced, currently in a higher order
union (including both remarriage and unmarried cohabitation); (e) divorced, currently living alone and, (f) never
divorced, widowed. The first group is the reference category.
If the parents of a respondent divorce, the subsequent partnership histories of the mother and of the father might differ.
Hence, for the parents we have created two sets of variables distinguishing between the relationship history of the
respondent’s father and mother separately. Respondents were asked if their parents ever got divorced and if so, if they
ever remarried or cohabited with someone else. We identify four categories: (a) parent still in a first marriage; (b) parent
divorced, ever re-partnered; (c) parent divorced, never re-partnered; and (d) parent never divorced but widowed. In
accordance with the variables measuring the relationship history of the adult child, we have considered adding the
parents' current living situation to their past life course. It is possible, for example, that one or both parents did repartner after the divorce but that they are currently single again. However, by doing so the subsamples became too
6
small. Since earlier research has argued that parental divorce and subsequent re-partnering loosen family ties we believe
that these variables suffice to investigate our main research questions.
The effects of the marital status of both the father and the mother cannot be estimated simultaneously with the effects of
relationship quality or support exchange due to issues of perfect multicollinearity. For example, when we have a score
for relationship quality with the mother, the father can never be widowed (because no questions were asked about the
relationship quality with a deceased mother). So, in order to be able to estimate the effects of the parental marital
history while at the same time controlling for relationship quality and support exchange, we ran our regressions
separately: first with the relationship history of the father as one of the explanatory variables (see header “Father” in
Table 3), then with the relationship history of the mother (header “Mother” in Table 3). The variables about relationship
quality, parental age, and support exchange (to be discussed next) relate to the same parent, as indicated by the column
heading.
Control variables
The relationship quality with each parent as perceived by the adult child is derived from the question: “Taking
everything together, how would you describe your relation with your mother/your father?”. We have created a dummy
variable indicating a good or very good relationship as opposed to a bad or reasonable one.
Next, we constructed variables to reflect the exchange of different types of support. Questions were asked both about
receiving the following types of support from each parent: (a) help with household tasks (”In the last three months, did
you receive help from your mother (c.q. your father) with housework, such as preparing meals, cleaning, fetching
groceries, doing the laundry?”); (b) help with odd jobs (“In the last three months, did you receive help from your
mother (c.q. your father) with practical matters such as chores in and around the house, lending things, transportation,
moving things?”); (c) showing interest in the other person’s life (“Has your mother (c.q. your father) shown an interest
in your personal life in the last 3 months?”) and (d) giving advice (“Did you get counsel or good advice from your
mother (c.q. your father) in the last 3 months?”). Exactly the same questions were asked about giving rather than
receiving the same types of support during the last 3 months. Answering options were “none”, “once or twice”, and
“several times”. We have applied principal component analysis in order to check the dimensionality of this set of items,
treating the scores as interval level variables for simplicity. The correlations between these items exhibited three
components: (a) exchange of socio-emotional support (i.e., showing interest and exchanging advice – people who give a
lot of socio-emotional support to parents tend to receive a lot of support as well, and they score high on this
component); (b) providing instrumental support (including household chores and odd jobs) and (c) receiving
instrumental support (including household chores and odd jobs). The internal consistencies of all these variables were
not great but still reasonably good, the Cronbach’s alpha all lying between .541 and .725. The components were only
weakly correlated after oblique rotation. This is not so surprising as socio-emotional support is more based on, or an
indication of, what is known as “affectual solidarity” (Bengtson, 2001), whereas instrumental support is primarily
triggered by need. We include scores for all three components as control variables in our models of family obligations.
Note that we have also estimated models with indicators for the exchange of financial support included. Since these
yielded almost identical results, we will not report them in this paper for reasons of parsimony, but the results are
available from the authors upon request.
Unfortunately, we do not have information about the health status of the respondents’ parents. Therefore, we have used
age as a proxy. We have created a dummy variable indicating whether or not the mother and/or father are over 75 years
7
old. The chances of being in good health are expected to be rapidly declining once over 75 (Stuifbergen, Van Delden
and Dykstra, 2008).
Other background variables we control for are the respondent’s age (5 categories: <30 years, 30-39, 40-49, 50-59, and
60 and more), gender, educational level (3 categories: up to lower secondary education, up to intermediate vocational
education, higher educated), the total hours per week spent in paid employment (included as a simple continuous
variable; categorizing in order to check for linearity yielded the same (non-significant) results), a dummy indicating
whether or not the respondent has children, and a set of dummies indicating whether the respondent has any brothers or
sisters, or both. Table 2 provides an overview of the basic characteristics of the sample and all variables used.
[Table 2 about here]
RESULTS
Regression results are shown in Table 3. Looking at the overall picture, we can determine some remarkable differences
in the respondents’ attitudinal outcome concerning filial and parental obligations. Yet, this is far less the case for
general family obligations. This might indicate that when family obligations are expressed in more general terms,
without identifying specific family members or relationships, they refer to more general norms of conduct between
wider kin and might be more de-contextualized or less dependent on personal circumstances. Conversely, when
expressed in terms of specific relationships (e.g., parents and adult children), these general norms might be more
affected by the individual’s own personal situation.
Respondent’s and parent’s divorce history and family obligations
Examining the respondent’s own partnership status, the most striking result is that divorced respondents without a new
partner generally have more positive attitudes towards all types of obligation compared to the ones still in their first
marriage. Divorcees who are currently living with a new partner also have stronger attitudes of filial obligations than
the never-divorced, as far as judged from the respondent-mother subsample. The other estimates of association between
divorce and attitudes towards family obligations tend to go in the positive direction as well, but they are not statistically
significant at the 5% level.
Although unexpected, the finding that divorce is associated with stronger normative family obligations is very robust. It
persists in all model reformulations that we have tested. Apart from a positive rather than a negative divorce-effect,
living alone without ever being married is also related to more positive attitudes towards filial and parental obligations
compared to the married. Further analyses revealed however that this seemed to be only the case for women (not
shown), whereas the divorce effect holds for men as well as for women. These effects are controlled for by the support
exchanged between the respondents and their parents, taking into account the fact that respondents without a partner,
i.c. never-married women and divorced men and women, might be more in need of support or possibly receive more
support from their parents compared to respondents with a partner. Nevertheless, the absence of a (new) partner might
lead to attaching more importance to kinship and family as a potential source of all kinds of support. Never-married
respondents who are living with their partner and widowed respondents do not significantly differ from the married in
their attitudes towards family.
[Table 3 about here]
8
The parents’ divorce histories are not related to more general and filial obligations, but they seem important for parental
obligations. Respondents whose parents ever divorced seem to have higher expectations of parents in general, especially
when their parents ever re-partnered, than respondents whose parents are still in their first marriage. Again, this is a
remarkable finding, not only because the effects are positive rather than negative but also because we expected to find
that the relationship between the parents’ divorce and family obligations, if any existed, would be more pronounced for
filial than for parental obligations. The same conclusions can be drawn for respondents with a widowed parent.
Particularly having a widowed father is more positively associated with parental obligation compared to having stillmarried parents. One possible explanation, drawing on exchange theory, could be that because adult children are more
likely to support their father after widowhood, they also have higher normative expectations that their father must help
them whenever they are in need. By including the relationship quality and support measures, the model controls for the
possibly lower relationship quality and reduced exchange of support between respondents and their divorced parents.
Yet, a clear and positive relationship between the parents’ divorce and the respondent’s attitudes about parental
obligations remains.
Interestingly, our results show that the types of obligations that are affected (filial or parental) are tied to the generation
in which the divorce occurred: the respondent’s own divorce history is more strongly and positively correlated to filial
obligations (Figure 1), whereas the parents’ divorce history seems to be only positively related to the respondent’s
attitudes towards parental obligations (Figure 2, see also Table 3).
[ Figure 1 about here]
[Figure 2 about here]
Control variables and family obligations
We included indicators for the quality of the relationship with the respective parents in our models in order to control
for the fact that this might be related both to marital status and to family obligations, potentially leading to a spurious
effect of divorce on family obligations. However, with respect to the association between marital status and relationship
quality with the mother, our data do not show a statistically significant difference between divorced people and people
still in their first marriage, with 85% reporting a good or very good relationship in both groups (X²=0.8, df=1, p>0.38).
With respect to the relationship with the father, in contrast, divorced people report a good or very good relationship less
often (78% compared to 84% in the still-married group, X²=14.6, df=1, p<0.01). The relationship quality with the
parent is also less often assessed as good or very good when the parent is divorced, although the difference between
divorced and married mothers (81% and 89% good relationship, respectively, X²=44.0, df=1, p<0.01) is much smaller
than the difference between divorced and married fathers (65% and 87% good relationship, respectively, X²=168.5,
df=1, p<0.01).
Respondents who have a good relationship with their mother show significantly more positive attitudes towards all
types of obligations compared to respondents who expressed a lower relationship quality with their mother. For the
relationship quality with the father we do not find such clear difference between respondents who have a (very) good
and those who have either a reasonable or a bad relationship. However, again there seem to be opposing associations for
men and women. Having a (very) good relationship with either parent is positively associated with feelings of family
9
obligations for men but negatively for women (not shown). So overall, men’s attitudinal outcomes seem to be more
conditioned by, or conditioning for, the current quality of the relationship with their parent(s), than women’s.
Verbally expressed attitudes about family obligations might also be affected by actual exchanges of support. In turn,
actual support exchange is likely to be related to divorce. This is clearly the case in our data. Divorced respondents
report significantly more exchange of socio-emotional support than respondents who still are in their first marriage with
both their fathers (t=3.03, df=2520, p<0.01) and their mothers (t=5.05, df=3341, p<0.01); overall there is more
exchange of socio-emotional support with mothers. When the parents are divorced, , there is less exchange of socioemotional support with both fathers (t=-5.8, df=897, p<0.01) as well as mothers (t=-2.1, df=3644, p<0.05). For practical
support, a distinction is made between giving and receiving. Divorced respondents tend to give less instrumental
support (although the differences between divorced and married respondents are not statistically significant for support
to fathers) but they receive more from their parents, from fathers (t=6.6, df=2416, p<0.01) as well as mothers (t=8.9,
df=3098, p<0.01). Divorced mothers, in turn, are reported to receive more instrumental support than still married
mothers (t=16.0, df=3405, p<0.01), while the difference is not statistically significant for divorced fathers (t=1.3,
df=910, p>0.19). Respondents receive less instrumental support from divorced fathers as well as mothers (both t-tests
statistically significant with p<0.01).
All models in Table 3 control for this pattern of actual intergenerational support in order to avoid estimating spurious
effects of divorce on family obligations. Table 3 reports the net association between actual support and attitudes of
family obligations as well. Existing support patterns seem to matter most for the respondent’s sense of filial obligations.
Exchanging socio-emotional or instrumental support with the father and/or receiving instrumental support from the
mother seem to be positively related to filial obligations. In other words, respondents who actually exchange support
with their parents, also express stronger norms that adult children should provide support for their parents. As to general
family obligations and parental obligations, most effects are not statistically significant.
Other important predictors of attitudes towards family obligations are gender, age and educational level. Overall,
women express weaker normative obligations than men. This is in line with some studies in other, mainly European
countries (Daatland and Herlofson, 2003). It appears contradictory to their actual behaviour but it might be an indication
of their more realistic view of caring tasks and the sacrifices they entail (Gans and Silverstein, 2006). The correlations
with age indicate that the young express more positive attitudes towards family obligations. Respondents in the oldest
age category, who are most likely to have adult children themselves, have the weakest attitudinal outcome on filial
obligation. In line with other research we can interpret these results as being favourable towards the next generation
(i.e., not wanting to burden them), rather than an unwillingness to care for the previous generation (Bromley and
Blieszner, 1997; Logan and Spitze, 1995; Lye, 1997; Ward, 2001). Overall, the youngest respondents have the strongest
attitudes about all kinds of family obligations but they are also less likely to be actually having to provide practical
support or be in need of such support themselves – perhaps with the exception of financial support -, which might make
them underestimate the implications of this responsibility. The parent’s age was included as a proxy for health status,
with age over 75 years indicating a higher likelihood of deteriorated health status. However, we hardly found any
statistically significant effect. Without a direct measure of the parents’ health status, no clear conclusions can be drawn
about this. Finally, support norms are negatively related to the respondents’ educational level: the higher the level of
education received, the weaker the norm of obligation. This finding can be related to a stronger emphasis on individual
autonomy by the highly educated and the extended opportunities to use formal care (Liefbroer and Mulder, 2006).
10
LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH
Clearly, this study has a number of limitations and our results call for further research. One of the difficulties in
studying family obligations lies in the general terms in which they are described. With our data, it is impossible to tell
whether respondents interpret items about family obligations with respect to some general ideal state of affairs or rather
with their own potential course of action in mind. As has been shown by other studies, obligations do not always seem
to motivate specific behaviour. For example, some types of support may still remain even if feelings of obligation
decline (or vice versa), indicating that other motivations besides feeling ‘obliged’ may be behind the support behaviour
(Stuifbergen, Van Delden and Dykstra, 2008). Especially in countries with well-developed welfare systems, like the
Netherlands, where the role of primary care-giving is often shared with state-provided services, parent-child
relationships might be more defined in terms of affection (Lowenstein and Daatland, 2006; Merz et al., 2009).
Secondly, it is important to bear in mind that all respondents in our sample have at least one parent still alive. Moreover,
some respondents may already have adult children of their own. Hence, it is difficult to disentangle the direction,
upward or downward, in which these attitudes must be interpreted. The same obligations can be interpreted either as
expectations one has to live up to towards others or as certain behaviour one is entitled to expect from others. For
example, on the one hand respondents who have (adult) children might have higher expectations of receiving support
from parents compared to respondents without children. On the other hand, being a parent themselves they might have
stronger feelings towards providing support to children. In both instances we would find stronger feelings of parental
obligations in respondents who have children but the rationale is different. In future studies, controlling for whether
respondents have adult rather than young children could shed a bit more light on the matter.
Thirdly, as divorce is often replicated down generations, it is important to study the effect of marital dissolution in
subsequent generations. For example, one study in the Netherlands found that when the child(ren) as well as the parents
had gone through a divorce the actual exchange of support between the generations was at its lowest (Dykstra, 1997). It
is important to gain more insight in the cumulative effect of these family dynamics. To that end, future studies should
try to investigate the interactions between the partnership histories of respondents and their parents and how these
cumulatively relate to attitudes toward family obligations.
Finally, given the cross-sectional nature of the data, the high level of endogeneity of the variables involved, and the
many mechanisms potentially leading to feedback and reverse causality, we cannot estimate the "true causal effect", as
it would be conceptualized in a counterfactual perspective (Morgan and Winship, 2007). For instance, it could be that
those with stronger feelings of filial or parental obligations are also more likely to divorce (i.e., that the attitudes were
preceding the divorce rather than being affected by it). But whatever the causal mechanisms involved, the pattern of
association is relevant in its own right. For example, the relationship between attitudes and expectations on the one hand
and actual behaviour on the other hand will affect how satisfied people will be with the actual state of affairs, whatever
the origin of the attitudes and expectations.
CONCLUSION AND DISCUSSION
In this paper we investigated whether divorce is associated with attitudes towards family obligations. We distinguished
between a divorce either in the parent or in the adult child's generation. We also distinguished between three types of
normative obligations: general family obligations, filial obligations and parental obligations. The dominant perspective
in the literature implies a negative association between divorce and all types of normative obligations. However, given
11
that we focused on the adult child's perspective, we expected to find this association to be stronger for filial obligations.
Researchers and policy makers have been, and remain, concerned about the role that family members play in the
provision of support for each other. Concerns have been expressed about the deleterious effect of divorce on the parentchild relationship, which might lead to an unwillingness to provide support to one another and potentially to more
people having to rely on social welfare. However, we found no negative correlation between divorce and norms of
family obligation in the Netherlands, nor did we find a non-effect, as would be expected from the continuity perspective
which argues that family obligations are resilient to changes in family structure.
The key finding of this paper is that divorced people in general tend to express stronger norms of family obligation, at
least in the Netherlands. Having divorced parents was found to correlate positively with family obligations as well. So,
on the normative attitudinal level, our findings lend support to the perspective that argues that family connections make
up a latent support network that is activated particularly in times of crisis. In addition, the norms to provide support to
parents (filial obligations) seem not to be guided by a parental divorce in itself, but rather by the way the parent-child
relationship persists or evolves after the divorce, as shown in the effects of the current relationship characteristics.
Conversely, respondents with divorced parents do have higher expectations to receive support from parents (parental
obligations), regardless of the perceived quality of the relationship or the amount of support exchanged with their
parents. So, our analyses reveal that the predominant type of obligation is tied to the generation in which the divorce
occurred: the respondent’s own divorce history is more strongly and positively related to his or her feelings of filial
obligation, whereas the parents’ divorce history seems to be only positively related to the respondent’s feelings of
parental obligation.
Living in a higher order union is also positively related to the respondent's attitudes towards family obligation, although
the effects are less clear. As argued before, finding a new partner after divorce may have opposing effects on support
norms. On the one hand, it could create tension or conflict in families, weakening family obligation. On the other hand,
by giving the opportunity to pick up again or continue (previous) patterns of support exchange, it might strengthen
family obligations. When the divorce history of the parents is considered, we find that respondents whose parents ever
re-partnered have stronger norms of parental obligation compared with respondents with still-married parents.
Besides the strong association with the respondent’s socio-demographic characteristics, such as gender, age and
educational level, we also found evidence that the relationship quality and the support exchanged between respondents
and their parents are positively associated with family obligations. This is in line with other research. In other words,
normative obligations are (partly) conditioned by, or conditioning for, the current parent-child relationship
characteristics. This seemed to be especially true for men's filial obligations, and the actual support exchange is more
strongly related to the respondent's relationship with the father than with the mother. Nevertheless, the positive
association between divorce and family obligations persist after controlling for these variables.
Based on our results one could hypothesize that having experienced loss through divorce, either from a parent or a
partner, and having experienced the hardships a divorce might bring with it, may result in stronger family norms,
regardless of the behaviour individuals show. One interpretation may be that people cherish blood relationships as
relatively unconditional, especially the relationship between parents and children, when they experience that one cannot
always count on the more "contractual" types of relationships, like marriage. The latter are freely chosen but this may
also make them more fragile. In-laws may come and go, but blood relatives remain family and people might argue that
they should be able to count on each other. Future research will have to see whether our findings can be replicated in
other countries.
12
Acknowledgements
The authors are very thankful to two anonymous reviewers for providing some very useful suggestions for improvement
of a previous version of this paper. Also, this study has benefited from many stimulating comments made by the
participants of the Multilinks EU-FP7 project. We thank all of them, and in particular Pearl Dykstra, the coordinator of
that project.
13
REFERENCES
Amato, P. R., and A. Booth, (1994). Parental marital quality, parental divorce and relations with parents, Journal of
Marriage and the Family, 56(2), 356-365.
Amato, P. R., S. J. Rezac, and A. Booth, (1995). Helping between parents and young adult offspring: the role of
parental marital quality, divorce, and remarriage, Journal of Marriage and the Family, 57(2), 363-74.
Aquilino, W. S. (1994). Later life parental divorce and widowhood: impact on young adults' assessment of parent-child
relations, Journal of Marriage and the Family, 56(4), 908-922.
Bengtson, V. L. and Roberts, R. E. L. (1991). Intergenerational Solidarity in Aging Families : An Example of Formal
Theory Construction. Journal of Marriage and Family, 53(4), 856–870.
Bengtson, V. L. (2001). Beyond the nuclear family: the increasing importance of multigenerational bonds, Journal of
Marriage and the Family, 63(1), 1-16.
Brandt, M., K. Haberkern, And M. Szydlik, (2009). Intergenerational help and care in Europe, European Sociological
Review, 25(5), 585 – 601.
Bromley, M. C. and R. Blieszner, (1997). Planning for long-term care: filial behavior and relationship quality of adult
children with independent parent, Family Relations, 46(2), 155-162.
Coleman, M., L. Ganong, L. and S. M. Cable, (1997). Beliefs about women's intergenerational family obligations to
provide support before and after divorce and remarriage, Journal of Marriage and Family, 59(1), 165-176.
Connidis, I.A. (2001). Family ties and aging. CA: Sage, Thousand Oaks.
Cooney, T. M. (1994). Young adults' relations with parents: the influence of recent parental divorce, Journal of
Marriage and the Family, 56(1), 45-56.
Cooney, T. M. and P. Uhlenberg, (1990). The role of divorce in men's relations with their adult children after mid-life,
Journal of Marriage and Family, 52(3), 677-688.
Daatland, S.O. (2007). Marital history and intergenerational solidarity: the impact of divorce and unmarried
cohabitation, Journal of Social Issues, 63(4), 809-825.
Daatland, S. O. and K. Herlofson, (2003). ‘Lost solidarity’ or ‘changed solidarity’: a comparative European view of
normative family solidarity, Ageing & Society, 23, 537 – 560.
De Graaf, P. M. and M. Kalmijn, (2006). Change and stability in the social determinants of divorce: a comparison of
marriage cohorts in the Netherlands, European Sociological Review, 22(5), 561 – 572.
De Graaf, P. M. and T. Fokkema, (2007). Contacts between divorced and non-divorced parents and their adult children
in the Netherlands: an investment perspective, European Sociological Review, 23(2), 263 – 277.
Dykstra, P. A. (1997). The consequences of divorce for intergenerational exchanges within families, The Netherlands'
Journal of Social Sciences, 33(2), 77-93.
Dykstra, P. A., M. Kalmijn, T. C. M. Knijn, A.E. Komter, A.C. Liefbroer, and C.H. Mulder, (2005). Codebook of the
Netherlands Kinship Panel Study, a multi-actor, multi-method panel study on solidarity in family relationships, Wave 1.
NKPS Working Paper No. 4. The Hague: Netherlands Interdisciplinary Demographic Institute.
Finch, J. and J. Mason, (1991). Obligations of kinship in contemporary Britain: is there normative agreement? The
British Journal of Sociology, 42(3), 345 – 367.
Furstenberg Jr., F.F., S.D. Hoffman, and L. Shrestha, (1995). The effect of divorce on intergenerational transfers: new
evidence, Demography, 32(3), 319-333.
Ganong, L., M. Coleman, A. Kusgen McDaniel, and T. Killian, (1998). Attitudes regarding obligations to assist an older
parent or stepparent following later-life remarriage, Journal of Marriage and the Family, 60(3), 595-610.
14
Ganong, L. and M. Coleman, (1999). Changing families, changing responsibilities. Family obligations following
divorce and remarriage. London: Lawrence Erlbaum associates.
Ganong, L. and M. Coleman, (2005). Measuring intergenerational obligations, Journal of Marriage and Family, 67(4),
1003-1011.
Ganong, L. and M. Coleman, (2006). Patterns of exchange and intergenerational responsibilities after divorce and
remarriage, Journal of Aging Studies, 20, 265-278.
Gans, D. and M. Silverstein, (2006). Norms of filial responsibility for aging parents across time and generations,
Journal of Marriage and Family, 68(4), 961-976.
Grundy, E. (2008). The challenges of ageing: prospects for the family support of older people in 21st century Europe. In
J. Surkyn, P. Deboosere, and J. Van Bavel, (eds.), Demographic Challenges for the 21st Century. A State of the Art in
Demography (pp. 223-246). Brussel: VUBPress.
Grundy, E. and J.C. Henretta, (2006). Between elderly parents and adult children: a new look at the intergenerational
care provided by the 'sandwich generation', Ageing & Society, 26(5), 707-722.
Hans, J., L. Ganong, and M. Coleman, (2009). Financial responsibilities toward older parents and stepparents following
divorce and remarriage, Journal of Family and Economic Issues, 30(1), 55-66.
Hilton, J. M. and K. Kopera-Frye, (2007). Differences in resources provided by grandparents in single and married
parent families, Journal of Divorce and Remarriage, 47(1/2), 33-54.
Ikkink, K. K., T. van Tilburg, and K. Knipscheer, (1999). Perceived instrumental support exchanges in relationships
between elderly parents and their adult children: normative and structural explanations, Journal of Marriage and
Family, 61, 831-844.
Kalmijn, M. (2007). Gender differences in the effects of divorce, widowhood and remarriage on intergenerational
support: Does marriage protect fathers? Social Forces, 85(3), 1079-1104.
Kaufman, G. and P. Uhlenberg, (1998). Effects of life course transitions on the quality of relationships between adult
children and their parents, Journal of Marriage and the Family, 60(4), 924-938.
Keizer, R., P.A. Dykstra, and A.R. Poortman, (2011). Childlessness and norms of familial responsibilities in the
Netherlands, Journal of Contemporary Family Studies, 42(4), 421-438.
Killian, T. and L.H. Ganong, (2002). Ideology, context, and obligations to assist older persons, Journal of Marriage and
Family, 64(4), 1080 – 1088.
Lawton, L., M. Silverstein, and V.L. Bengtson, (1994). Affection, social contact, and geographic distance between adult
children and their parents, Journal of Marriage and the Family, 56, 57-68.
Leeson, G. W. (2005). Changing Patterns of Contact with and Attitudes to the Family in Denmark, Journal of
Intergenerational Relationships, 3(3), 25 - 45.
Liefbroer, A. C. and C.H. Mulder, (2006). Family obligations. In P.A. Dykstra, M. Kalmijn, T.C.M. Knijn, A.E.
Komter, A.C. Liefbroer, and C.H. Mulder, (eds.), Family solidarity in the Netherlands (pp. 123-145). Amsterdam:
Dutch University Press.
Lin, I. F. (2008). Consequences of parental divorce for adult children’s support of their frail parents, Journal of
Marriage and Family, 70(1), 113-128.
Logan, J. R. and G.D. Spitze, (1995). Self-interest and altruism in intergenerational relations, Demography, 32(3), 353364.
Lowenstein, A. and S.O. Daatland, (2006). Filial norms and family support in a comparative cross-national context:
evidence from the OASIS study, Ageing & Society, 26, 203 – 223.
Lye, D. N. (1996). Adult child-parent relationships, Annual Review of Sociology, 22, 79-102.
15
Mandemakers, J. J. and P. A. Dykstra (2008). Discrepancies in Parent's and Adult Child's Reports of Support and
Contact, Journal of Marriage and Family, 70(2), 495-506.
Merz, E.-M., N.S. Consedine, H.-J. Schulze, and C. Schuengel, (2009). Wellbeing of adult children and ageing parents:
associations with intergenerational support and relationship quality, Ageing & Society, 29(5), 783-802.
Morgan, S.L. and C. Winship, (2007). Counterfactuals and Causal Inference. Methods and Principles for Social
Research. Cambridge: Cambridge University Press.
Pezzin, L. E. and B. Steinberg Schone, (1999). Parental marital disruption and intergenerational transfers: an analysis of
lone elderly parents and their children, Demography, 36(3), 287-297.
Ribar, D. C. and M.O. Wilhelm, (2006). Exchange, role modeling and the intergenerational transmission of elder
support attitudes: evidence from three generations of Mexican-Americans, Journal of Socio-Economics, 35, 514-531.
Rossi, A. and P. Rossi, (1990). Of human bonding. Parent-child relationships across the life course. New York: Aldine
de Gruyter.
Sarkisian, N. and N. Gerstel, (2008). Till marriage do us part: adult children’s relationships with their parents, Journal
of Marriage and Family, 70, 360 – 376.
Silverstein, M., Gans, D., & Yang, F. M. (2006). Intergenerational Support to Aging Parents: The Role of Norms and
Needs. Journal of Family Issues, 27(8), 1068–1084.
Spitze, G. and J.R. Logan, (1991). Employment and filial relations: is there a conflict? Sociological Forum, 6, 681-697.
Spitze, G., J.R. Logan, G. Deane, and S. Zerger, (1994). Adult children's divorce and intergenerational relationships,
Journal of Marriage and the Family, 56(2), 279-293.
Stein, C.H., V.A. Wemmerus, M. Ward, M.E. Gaines, A.L. Freeberg, and T.C. Jewell, (1998). "Because they're my
parents": an intergenerational study of felt obligation and parental caregiving, Journal of Marriage and Family, 60,
611-622.
Stuifbergen, M. C., J.J.M. Van Delden, and P.A. Dykstra, (2008). The implications of today’s family structures for
support giving to older parents, Ageing & Society, 28(3), 413 – 434.
Tomassini, C., K. Glaser, and R. Stuchbury, (2007). Family disruption and support in later life: a comparative study
between the United Kingdom and Italy, Journal of Social Issues, 63(4), 845-863.
Umberson, D. (1992). Relationships between adult children and their parents: psychological consequences for both
generations, Journal of Marriage and the Family, 54, 664-674.
Van Bavel, J., P. Dykstra, B. Wijckmans, and A.C. Liefbroer. 2010. Demographic change and family obligations. Paper
delivered for the EU-FP7 Multilinks project, How Demographic Changes Shape Intergenerational Solidarity, WellBeing, and Social Integration: A Multilinks Framework. www.multilinks-project.eu.
Van Gaalen, R. I. and P.A. Dykstra, (2006). Solidarity and conflict between adult children and parents: a latent class
analysis, Journal of Marriage and Family, 68(4), 947-960.
Ward, R. A. (2001). Linkages between family and societal-level intergenerational attitudes, Research on Aging, 23(2),
179-208.
16
Table 1: Factor loadings for items expressing family obligations
Factorsa, b
Variables
One should always be able to count on
family
Family members should be ready to
support one another, even if they don't
like each other
If one is troubled, family should be
there to provide support
Family members must help each other,
in good times and bad
Children should look after their sick
parents
In old age, parents must be able to live
in with their children
Children who live close to their
parents should visit them at least once
a week
Children should take unpaid leave to
look after their sick parents
Parents should support their adult
children if they need it
Parents should help their adult
children financially if they need it
Parents should provide lodging to their
adult children if they need it
a
b
General family
Filial
Parental
obligations
obligations
obligations
(Cronbach’s α = .87) (Cronbach’s α = .75) (Cronbach’s α = .80)
.773
.309
.390
.664
.356
.424
.872
.368
.454
.889
.346
.432
.434
.718
.366
.230
.703
.242
.411
.573
.383
.218
.668
.232
.473
.323
.706
.350
.305
.812
.386
.308
.743
Rotation method: Oblimin with Kaiser Normalization
Structure matrix
17
Table 2: Basic descriptive statistics of the sample*
Respondent characteristics
Gender: woman
Age: <30
30-39
40-49
50-59
60+
Partnership status:
- married
- cohabiting
- living alone
- widowed
Siblings:
- no siblings
- only brothers
- only sisters
- both brothers and sisters
Has any children?
Level of education:
- up to lower secondary
- up to intermediate vocational
- higher education
Number of working hours per week
3862
2308
692
1352
1116
591
111
Never divorced
Percentage
or Mean (SD)
100.0%
59.7%
17.9%
35.0%
28.9%
15.3%
2.9%
2613
646
548
55
67.7%
16.7%
14.2%
1.4%
169
97
183
0
37.6%
21.6%
40.8%
0.0%
176
953
856
1877
2618
4.5%
24.7%
22.2%
48.6%
67.8%
21
108
86
234
4.7%
24.1%
19.1%
52.1%
999
1377
1486
3862
25.9%
35.6%
38.5%
27.2 (18.3)
154
151
144
449
34.3%
33.6%
32.1%
24.8 (18.0)
Father characteristics
Marital history of father:
- still married with mother
- divorced, ever re-partnered
- divorced, never re-partnered
- never divorced, widowed
Father aged over 75 years?
Relationship quality (very) good?
Exchange of emotional support (component score)
Providing instrumental support (component score)
Receiving instrumental support (component score)
2618
100%
242
100%
2037
237
106
238
540
2159
2568
2568
2568
77.8%
9.1%
4.0%
9.1%
20.6%
82.4%
0.01 (0.98)
0.00 (0.99)
-0.03 (0.98)
164
26
9
43
112
177
235
235
235
67.8%
10.7%
3.7%
17.8%
45.7%
72.5%
-0.16 (1.09)
0.07 (1.09)
-0.13 (0.99)
3603
2038
201
214
1150
1043
3094
3553
3553
3553
100.0%
56.6%
5.6%
5.9%
31.9%
29.0%
85.9%
-0.01 (0.98)
0.01 (0.99)
-0.04 (0.98)
399
163
21
32
183
173
326
397
397
397
100.0%
40.8%
5.3%
8.0%
45.9%
43.4%
81.8%
0.03 (1.03)
0.06 (1.08)
-0.17 (0.94)
N
Mother characteristics
Marital history of mother:
- still married with father
- divorced, ever re-partnered
- divorced, never re-partnered
- never divorced, widowed
Mother aged over 75 years?
Relationship quality (very) good?
Exchange of emotional support (component score)
Providing instrumental support (component score)
Receiving instrumental support (component score)
Ever divorced
N
Percentage
or Mean (SD)
449
100.0%
292
65.1%
13
2.9%
98
21.9%
204
45.5%
112
25.0%
21
4.8%
* Unweighted numbers and percentages
18
Table 3: Ordinary least squares regression analyses of family obligations
General family obligations
Filial obligations
Father
Mother
Father
Mother
b (SE)
Sig.
b (SE)
Sig.
b (SE)
Sig.
b (SE)
Constant
.334 (.129) **
.408 (.107)
***
.413 (.118)
***
.337 (.095)
Woman
-.102
*
-.144 (.035)
***
-.178 (.038) *** -.240 (.031)
Respondent's age (< 30 = ref.)
(.041)
30-39
-.337
*** -.344 (.048)
***
-.136 (.045)
**
-.121 (.042)
40-49
-.559
*** -.509 (.057)
***
-.293 (.057) *** -.241 (.050)
(.049)
50-59
-.563
*** -.435 (.075)
***
-.413 (.086) *** -.330 (.066)
(.062)
60+
-.214
-.384 (.111)
***
-.510 (.212)
*
-.425 (.099)
(.095)
Education (Up to lower general secondary =
(.233)
Up to intermediate vocational
-.256
*** -.219 (.039)
***
-.169 (.043) *** -.157 (.035)
ref.)
Higher education
-.422
*** -.386 (.040)
***
-.283 (.044) *** -.235 (.035)
(.047)
Number of working hours/week
.002
(.001)
.000 (.001)
.001 (.001)
-.001 (.001)
(.049)
Respondent has children
.117 (.051)
*
.034 (.045)
.088 (.047)
.027 (.040)
Siblings (No siblings = ref.)
Only brothers
.086 (.092)
-.015 (.074)
-.120 (.084)
-.097 (.066)
Only sisters
.070 (.094)
-.054 (.075)
-.139 (.086)
-.147 (.066)
Both (brothers and sisters)
.129 (.091)
.005 (.071)
-.002 (.083)
-.023 (.063)
Respondent's partnership status *
Never married, currently cohabiting
-.027
-.022 (.050)
-.039 (.050)
.022 (.044)
Never married, living alone
.003
(.064)
-.020 (.058)
.139 (.059)
*
.140 (.052)
(.055)
Divorced, currently in higher order union
-.060
.011 (.063)
.083 (.073)
.138 (.056)
Divorced, currently living alone
.174
(.099)
.202 (.075)
**
.205 (.090)
*
.283 (.066)
(.080)
Never divorced, currently widowed
-.170
-.126 (.132)
-.159 (.192)
-.031 (.117)
Parent's marital history *
(.210)
Father/mother divorced, ever re-partnered
-.096
-.079 (.069)
.058 (.060)
.058 (.061)
Father/mother divorced, never re-partnered .037
(.097)
-.080 (.065)
.037 (.088)
.058 (.058)
(.065)
Father/mother never divorced, widowed
.059 (.060)
.024 (.038)
.030 (.055)
-.013 (.034)
Current parent-child relationship :
Very good/good relationship quality
.104 (.052)
*
.245 (.046)
***
.081 (.048)
.209 (.041)
Need: Father/mother over 75
.007 (.057)
-.043 (.048)
-.097 (.052)
-.059 (.043)
Support: Exchange socio-emotional support
.074 (.020) *** .022 (.017)
.068 (.018)
***
.038 (.015)
Providing instrumental support
.024 (.018)
-.002 (.015)
.062 (.016)
***
.016 (.014)
Receiving instrumental support
-.012
.009 (.017)
.043 (.018)
*
.058 (.015)
(.019)
R²
.074
.068
.075
.080
N
2801
3943
2801
3943
* ‘Still in first marriage’ is the reference category for both the respondent’s partnership status and the parent’s marital history
Sig.
***
***
Parental obligations
Father
Mother
b (SE)
Sig.
b (SE)
Sig.
.491 (.122) *** .295 (.100)
**
-.235 (.039) *** -.236 (.033) ***
**
***
***
***
-.379 (.046)
-.476 (.059)
-.324 (.089)
-.099 (.220)
***
***
***
-.376 (.044)
-.430 (.053)
-.219 (.070)
-.064 (.104)
***
***
**
***
***
-.225 (.044)
-.318 (.046)
-.001 (.001)
.203 (.048)
***
***
-.178 (.036)
-.240 (.037)
-.001 (.001)
.181 (.042)
***
***
*
-.194 (.087)
-.206 (.088)
-.152 (.086)
*
*
-.197 (.069)
-.236 (.070)
-.196 (.066)
**
***
**
**
*
***
.103 (.052)
.177 (.061)
.133 (.075)
.177 (.093)
-.291 (.198)
*
**
.074 (.047)
.161 (.055)
.086 (.059)
.207 (.070)
-.199 (.123)
.216 (.062)
.127 (.091)
.189 (.057)
***
***
*
***
-.026 (.049)
-.018 (.054)
.031 (.019)
.027 (.017)
.005 (.018)
.085
2801
***
***
***
**
**
.206 (.064)
.174 (.061)
.055 (.036)
***
**
.155 (.043)
-.015 (.045)
-.012 (.016)
.007 (.014)
.022 (.016)
.075
3943
***
19
Figure 1: Effect of respondent’s partnership status on his or her feelings of filial obligations compared with the
reference group (i.e., respondents still in first marriage)*
* Lines represent 95% confidence intervals; net effects after controlling for the other variables in the model for filial
obligations with mother's marital history (cf. Table 3).
20
Figure 2: Effect of the parent’s marital history and the respondent’s feelings of parental obligations compared
with the reference group (i.e., respondents with parents still in first marriage)*
* Lines represent 95% confidence intervals; net effects after controlling for the other variables in the model for parental
obligations with mother's marital history (cf. Table 3).
21