Father-Child Relations after Divorce in Four European Countries: Patterns and Determinants


Father-Child Relations after Divorce in Four European Countries: 
Patterns and Determinants

Matthijs Kalmijn

Abstract: Using nationally representative data on secondary school children in Eng-
land, Germany, the Netherlands, and Sweden, this study describes the relationships 
that children have with their fathers after divorce. Differences in the post-divorce 
relationship are explained in terms of demographic factors, socioeconomic factors, 
and contextual differences (between countries and between immigrants and na-
tives). The focus is on living arrangements after divorce, the amount of contact 
with the father, and the perceived quality of the relationship. Many children have 
at least weekly contact with their father but one in six children never see their fa-
ther at all. Fathers in high-status families are more highly involved in the child’s life 
after divorce than fathers in low-status families. A mother’s employment also has 
a positive effect on the post-divorce relationship with the father. Co-parenting is 
most common in Sweden. Post-divorce relationships are also strongest in Sweden 
and relatively weak in Germany. Immigrant children see their fathers less often after 
divorce than native children. In the second generation, children of mixed marriages 
in particular tend to have little contact with the father.

Keywords: Divorce · Fatherhood · Children · Parent-child relations · Immigrants

1 Introduction

One of the more important consequences of divorce is the effect it has on the fa-
ther-child relationship (Pryor/Rodgers 2001). It is no surprise then that much re-
search has been done on this topic in recent decades. Some studies focus on long-
term effects and compare the relationships that adult children have with their father 
depending on whether or not the parents divorced or separated when the children 
were young (Albertini/Garriga 2011; De Graaf/Fokkema 2007). Other studies focus 
on short-term effects and examine the relationship that divorced fathers have with 
their children when the children are still living at home, usually with the mother 

Comparative Population Studies
Vol. 40, 3 (2015): 251-276 (Date of release: 30.09.2015)

© Federal Institute for Population Research 2015  URL: www.comparativepopulationstudies.de
DOI: 10.12765/CPoS-2015-10en    URN: urn:nbn:de:bib-cpos-2015-10en1



•    Matthijs Kalmijn252

(Adamsons/Johnson 2013; Juby et al. 2007; Spruijt/Duindam 2009). In general, the 
degree to which divorced fathers remain involved in children’s lives after divorce is 
an important issue for understanding the well-being of both children and fathers, 
is relevant for the resources that fathers can pass on to their children, and has an 
impact on the degree to which fathers receive support from children when they are 
old and in need of support (Daatland 2007; Kalmijn 2007; Lin 2008; Seltzer/Bianchi 
2013). Because paternal involvement can be affected by legal custody and visiting 
arrangements and by rules about alimony, it is also of concern to policy makers and 
health practitioners.

Although there is consensus that a divorce weakens father-child relationships, 
there is considerable heterogeneity in this effect. Some divorced fathers maintain 
frequent contact and remain an infl uential father fi gure in the child’s life; other di-
vorced fathers lose contact or remarry and shift their investments to a new family 
(Cheadle et al. 2010; Manning/Smock 1999). In understanding these differences, 
authors have looked at several types of moderating variables. Most authors have fo-
cused on life course factors such as the child’s age at divorce, repartnering, residen-
tial moves, and employment changes (Aquilino 2006; Cooksey/Craig 1998; Jubyet 
al. 2007). Fewer authors have examined how the social context in which the divorce 
takes place moderates the impact of divorce. Examples of contexts that may be 
important are the country in which parents live (Kalmijn 2008), the time period in 
which the divorce occurs (Albertini/Garriga 2011), and the ethnic or racial group to 
which the parents belong (Kalmijn 2010; King et al. 2004). These contexts are rel-
evant because they are related to the cultural climate in which a divorce occurs, to 
the legal and informal rules about how to deal with divorce and single parenthood, 
and to the economic opportunities that families are facing. Other potentially rel-
evant moderators about which little is known are the socioeconomic characteristics 
of parents such as their education, occupational status, and employment. There is 
debate about whether children from high-status families are protected from the ad-
verse consequences of divorce (Bernardi/Radl 2014; Mandemakers/Kalmijn 2014), 
but it is not known if father-child relationships are more or less affected by a divorce 
in high-status families.

In this study, new data from 14-year-old children in four European countries are 
analysed to examine the determinants of the father-child relationship after divorce. 
The fi rst goal is to describe three aspects of the father-child relationship for ado-
lescents whose parents divorced or separated. The focus is on residential arrange-
ments (where does the child live?), the amount of contact with the father, and the 
perceived quality of the relationship. While these aspects have been described be-
fore, virtually all demographic and sociological studies are based on American or 
Canadian data. As a result, even the simplest facts such as the percentages of co-
parenting arrangements after divorce are unknown for most European countries. 
Moreover, there is much popular debate and concern about divorced fathers who 
never see their children (e.g. Zander 2012) but we do not have good data on how 
common this is. There are now several European studies on adult children and their 
divorced fathers (Albertini/Garriga 2011; Kalmijn 2008), but studies on children who 
are still living at home are scarce. Because we have nationally representative data, 



Father-Child Relations after Divorce in Four European Countries    • 253

this study is able to fi ll in this important gap in the literature. The second goal is to 
understand heterogeneity in father-child relationships after divorce. To meet this 
goal, we develop hypotheses about the infl uence of three types of characteristics 
on each aspect of the father-child relationship. More specifi cally, we focus on de-
mographic characteristics, socioeconomic characteristics, and contextual charac-
teristics (i.e. country differences and differences among immigrants and natives as 
well as among major immigrant groups). We test hypotheses about the infl uence of 
these characteristics in a series of multivariate regression models.

2 Hypotheses

Socioeconomic characteristics

Our fi rst set of hypotheses is about socioeconomic characteristics, in particular the 
socioeconomic status of the family and the employment status of the mother. There 
are several reasons why the socioeconomic status of the family may affect post-
divorce relationships. First, studies have shown that more highly educated fathers 
are more involved in child rearing tasks than fathers with lower levels of educa-
tion (Hook/Wolfe 2012; Sayer et al. 2004; Yeung et al. 2001). The more fathers are 
involved in their children’s lives, the more likely it is that the ties will remain strong 
after divorce. Fathers may be more attached to their children when they were highly 
involved, just as children may be more attached to their fathers. In the legal deci-
sion-making after divorce, these investments may also be recognized and this may 
give fathers more opportunity to retain contact or to obtain custody or co-parenting 
rights. A second argument is that fathers in high-status families may be more ef-
fective in working out post-divorce agreements with the mother (Ryan et al. 2008). 
Thirdly, higher-status fathers and mothers read newspapers and magazines more 
often (Kraaykamp 2002) and may therefore be more aware of public knowledge 
about the negative effects of divorce on children. For these reasons, we believe that 
parents in high-status families will pay more attention to the well-being of the child 
and this means, among other things, more balanced roles for the father and the 
mother after divorce. In sum, our hypothesis is that children in high-status families 
have stronger ties to their fathers than children in lower-status families. This would 
apply to all three outcomes that we study: co-parenting will be more likely, frequent 
contact will be more likely, and ties will be of higher quality.

The father-child relationship may also be affected by the employment status of 
the mother. There is considerable heterogeneity in female employment after di-
vorce, with many divorced mothers who are not working for pay (Van Damme 2010). 
There is debate about the possible effects of mothers’ employment on children’s 
well-being and school outcomes and the potentially mediating role of parent-child 
relationships in these effects (Waldfogel et al. 2002). Consensus has arisen that 
mothers’ employment does not have general negative effects on child outcomes 
(Waldfogel et al. 2002). There is also little evidence that maternal employment com-
petes with the time mothers spend with children, although there may be a quality/



•    Matthijs Kalmijn254

quantity trade-off among employed mothers (Bianchi 2000). It is not known, how-
ever, what infl uence employment has in single-parent families where working moth-
ers cannot rely on their partner to support them with child rearing tasks. We suspect 
that for single mothers, employment may have more negative outcomes for chil-
dren. More specifi cally, we expect that single mothers who work for pay have less 
time and energy for their children than single mothers who are at home most of the 
time. As a result, the children will be more likely to live (permanently or at least part 
of the time) with the father and will see their father more often as well if the mother 
works for pay. This will apply in particular to full-time working mothers but no infor-
mation is present in our data on the mothers’ working hours. Moreover, in families 
where the mother works, the father may already be more involved in child rearing 
tasks to begin with. Since early investments in a child have positive effects on the 
relationship with the child when the marriage breaks down (Juby et al. 2005), the 
effect of maternal employment could work in both ways: via the time that mothers 
have and via the investments that fathers make. 

Demographic characteristics

The literature has shown that there are no clear gender differences in the conse-
quences of divorce for children’s well-being except for a somewhat stronger effect 
of divorce on the problem behaviour of boys than on the problem behaviour of 
girls (Amato 2001). In addition, past studies fi nd inconsistent effects of gender on 
post-divorce contact (Cooney 1994; Swiss/Le Bourdais 2009; Manning et al. 2003; 
Cooksey/Craig 1998). Theoretically, one would expect that divorced fathers have 
stronger ties to sons than to daughters: Attachment is believed to be stronger to 
the same-sex parent, the development of a child’s gender identity is more diffi cult 
when a child is raised by an opposite-sex parent, and a parent may feel more com-
fortable in advising a same-sex child than an opposite-sex child (Powell/Downey 
1997). It is important to realise that this infl uence may be bi-directional: fathers may 
feel a stronger attachment to their son than to their daughter and hence take more 
initiative for contact, but sons may also feel a stronger attachment to their father 
than daughters and consequently visit more often. It is diffi cult to unravel these 
processes but we can develop some additional tests to clarify these things. If the 
effect of child gender on contact disappears when controlling for the quality of the 
tie with the father as perceived by the child, the gender difference probably stems 
from differences in what the children – sons and daughters – want. If the effect of 
child gender on contact remains when controlling for the perceived quality of the tie 
with the father, the gender difference also has to do with the preferences of fathers 
for sons as opposed to daughters. 

The second demographic variable is the number of siblings. Research on adult 
intergenerational relationships has found that adult children have contact with their 
parents less often when they have more siblings (Grundy/Read 2012). From the per-
spective of parents, there is a benefi t in having a large family: parents with many 
children have more contact with at least one of the children. At the dyad level, how-
ever, the effect of sibsize on contact frequency is negative. This can be interpreted 



Father-Child Relations after Divorce in Four European Countries    • 255

in terms of time constraints (when there are more children, parents have less time 
per child) and in terms of a “division of labour” within families (when one child visits 
often, the other child may not need to visit). In this study, we examine if such an 
effect also exists for children of divorced parents who live with their mother, some-
thing that has not been studied as far as we know. For this more specifi c case, it is 
not clear if the effect will operate. Young children who live with the mother may see 
their father simultaneously. This reduces the role of time constraints and may result 
in an insignifi cant “sibsize” effect on father-child contact. There may be a negative 
effect of sibsize on co-parenting if fathers fi nd it more diffi cult to care for a large 
family than for a small family

Whether or not the mother repartners is another important demographic vari-
able in the literature on divorce. In the present study, we have good information 
on whether the mother has a new partner but we only know this if the child lives 
with the mother. Previous research has not found consistent effects of the mother’s 
repartnering. A cross-sectional study fi nds a negative effect of the mother’s remar-
riage on visitation frequency with the father (Stephens 1996). A recent longitudinal 
study, however, shows that the entry of a stepfather in the mother’s household does 
not lead to less frequent contact or a poorer quality relationship with the (biologi-
cal) non-resident father (King 2009). This refutes the hypothesis that stepfathers are 
substitute parents. We test this notion again, this time using large-scale albeit cross-
sectional data in four European countries. Unfortunately, we do not have data on 
whether the father is remarried when the child is not living with him.

The national context: Origin and destination effects

There are two ways that the national context may play a role. First, there can be dif-
ferences between destination countries, i.e. between England, Germany, the Neth-
erlands, and Sweden. Second, there can be differences based on origin countries. 
Because schools with high numbers of immigrants were systematically oversam-
pled, we can examine differences between natives and (fi rst- and second-genera-
tion) immigrants, as well as between immigrant groups themselves. This allows us 
to examine “origin country” effects.

Starting with the destination effects, the question is: can we expect differences 
in post-divorce father-child relationships among the four countries? One hypothesis 
suggests that in countries with higher rates of divorce, there will be more frequent 
contact between the child and the divorced father. Divorce is more institutionalised 
in these countries and legal arrangements are more favourable for divorced fathers 
(Albertini/Garriga 2011; Kalmijn 2008). How do divorce rates vary among the four 
countries? To assess this, we present net divorce rates for 2002 and 2011 in Figure 1. 
Net divorce rates are not widely available and were constructed as the annual num-
ber of divorces divided by the population of married women in a year (in 1000s), 
based on data from the United Nations Statistics Division or the country-specifi c 
statistical websites (Offi ce for National Statistics, Statistisches Bundesamt, CBS 
Statline). For England and Wales, the early divorce rate is for 2001, the year in which 
the census was taken. The fi gure shows that divorce is most common in Sweden 



•    Matthijs Kalmijn256

and England and less common in Germany and the Netherlands. In England, the net 
divorce rate declined between 2001 and 2011 so that it is now more similar to that in 
Germany and the Netherlands. Our data were collected in 2010-2011 but the parents 
may have divorced at an earlier point in time. Hence, to compare divorce rates, we 
need to look at the entire decade. Based on this line of reasoning, we would expect 
to fi nd the strongest relationships between fathers and children in Sweden and Eng-
land, and weaker ties in Germany and the Netherlands.

The four destination countries also differ in their gender roles. Attitudes toward 
maternal employment are more positive in Sweden and the Netherlands than in Eng-
land and Germany (Kalmijn 2003). The percentage of mothers with young children 
(under 6) who are employed is also higher in Sweden (76.8) and the Netherlands 
(77.1) than in Germany (59.7) and the UK (59.1) (OECD website, http://ec.europa.
eu/eurostat/data/database, data for 2012). When looking at the division of paid and 
unpaid work, Sweden stands out as being the most egalitarian (Knudsen/Waerness 
2008; Gauthier/DeGusti 2012). Studies of the time fathers spend on childcare show 
that fathers are more involved in childcare in Sweden than in Germany and Eng-
land (Gauthier/DeGusti 2012). In sum, when gender roles are the dominant explana-
tion, we would expect Sweden to have the most involved fathers after divorce and 
Germany and England to have the least involved fathers. The position of the Neth-
erlands will be somewhere in between but is less certain since comparative time 
diary studies rarely include the Netherlands. Theories about gender roles and the 
institutionalisation of divorce thus imply a different order of destination countries 
with respect to father involvement. The comparison between the two high-divorce 
countries Sweden and England is especially informative in this respect. If gender 
roles are the dominant mechanism, Sweden will have the most positive post-di-

Fig. 1: Net divorce rates by country

0

2

4

6

8

10

12

14

16

Germany Netherlands Sweden England & Wales

10,7 9,8

14,6

11,110,8
9,4

13,7 12,9

2001/2002

2011

Divorces per 1,000 married women

Note: For England and Wales data refer to 2001; other countries refer to 2002.

Source: United Nations Statistics Division, Offi ce for National Statistics (UK), Statistisches 
Bundesamt (Germany), CBS Statline (Netherlands)



Father-Child Relations after Divorce in Four European Countries    • 257

vorce father-child relationships, whereas if the institutionalisation effect is domi-
nant, Sweden and England will be similar in this respect.

We now discuss possible differences between immigrants and natives and be-
tween the various immigrant groups that make up the European context. Like most 
other European analyses, we focus on non-western immigrants even though west-
ern immigrants are potentially interesting as well. So far, very few studies have 
analysed post-divorce father-child relationships for immigrants or for ethnic minor-
ity groups. Some studies in the US fi nd that black and Hispanic fathers are less in-
volved in their children’s lives after divorce than comparable white fathers (Cheadle 
et al. 2010). Other studies do not fi nd clear differences between major racial groups 
(Cooksey/Craig 1998; King et al. 2004; Manning/Smock 1999). There are more stud-
ies on ethnic and racial differences in child rearing practices but these focus only 
on intact families (Pels/Nijsten 2003; Yaman et al. 2010; Emmen et al. 2013). Why 
would we expect differences between immigrants and natives in the strength of 
the father-child relationship after divorce? First, we can look at gender roles in vari-
ous immigrant groups. In Muslim countries, there is considerably less support for 
gender equality than in Western countries (Norris/Inglehart 2002). In Latin American 
and Southern European countries, gender role attitudes are also more traditional 
but they deviate less from the Western European pattern than Muslim countries 
(Kalmijn 2003; Norris/Inglehart 2002). As a result, most immigrant fathers could be 
less involved in child rearing during marriage and if this is true, they consequently 
may have weaker ties to their children when they are divorced and the children are 
adolescents. 

Secondly, we can consider the institutionalisation hypothesis. In some of the 
origin countries – especially Muslim countries – divorce and single parenthood are 
strongly disapproved of (Norris/Inglehart 2002). One would expect that a normal 
continuation of the father-child relationship after divorce is more diffi cult when the 
community disapproves of divorce. In these cases, it is more likely that a divorce co-
incides with an abrupt discontinuation of contact between the two ex-partners and 
their respective families. Not all immigrant groups disapprove of divorce and single 
parenthood, however. Immigrants with African and Caribbean origins in particular 
are characterised by high levels of single parenthood and high tolerance of divorce 
(Kalmijn 2010). At the same time, however, it has been argued that in these groups, 
the role of the father in the family tends to be weaker while networks of female 
friends and kin are more strongly relied upon to provide support in single-mother 
families (Sarkisian/Gerstel 2004; Stack 1974). As a result of their more matrifocal 
orientation, we would expect that the children of African and Caribbean origins will 
also have a weak relationship to the father after divorce, although for a different 
reason than is assumed to be true for immigrants from Muslim countries. 

Thirdly, economic considerations may play a role. Immigrants from non-western 
origin countries on average have a weaker economic position in society. Their earn-
ings tend to be lower, they have less education on average, and they are unem-
ployed more often, even when controlling for the infl uence of education (Heath 
et al. 2008; Van Tubergen 2004). The weak economic position of immigrant men 
not only reduces their chances of marrying and increases their chances for divorce 



•    Matthijs Kalmijn258

(Edin/Kefalas 2005; Lichter et al. 1992; Oppenheimer 2003); it may also weaken the 
position of the father in the post-divorce family. When fathers have few economic 
resources, they will be less likely to pay child support and this may limit the degree 
to which they remain involved in the child’s life (Arditti/Keith 1993; Hofferth et al. 
2010). Economically disadvantaged fathers are also less able to provide a stable and 
comfortable place for their children to live, which would reduce the options for co-
parenting. It is therefore important to estimate the effect of immigrant background 
while controlling for the family’s socioeconomic status. While we have multiple in-
dicators of socioeconomic status in our data, we do not have data on the income 
situation of the absent father. As a result, the net effects of immigrant background 
should not entirely be interpreted from a cultural perspective; residual economic 
causes for the differences between natives and immigrants may still play a role.

3 Data and method

The CILS4EU data were collected in four countries. About 100 secondary schools 
were randomly chosen per country and two (randomly chosen) classes in each 
school were used (Kalter et al. 2012). The focus was on year 10 pupils in England, 
year 9 in Germany, year 3 of secondary schools in the Netherlands, and year 8 in 
Sweden. The children were about 14 years old in each country. Schools with high 
proportions of immigrants were systematically oversampled via a stratifi ed sam-
pling scheme to ensure a large enough sample of immigrants. Based on the percent-
age of immigrants in a school, four strata were constructed and random samples of 
schools were drawn within these strata. To solve the problem that schools may not 
be willing to participate, a replacement strategy was used where each school was 
matched to a replacement school that would be approached if the initial school did 
not respond. This is similar to the procedures used by other international school 
studies such as PISA and TIMMS. After replacement, response at the school level 
was 65.6 percent in England, 98.6 percent in Germany, 91.7 percent in the Nether-
lands, and 76.8 percent in Sweden (where no replacement procedure was used).

From the original data, we selected children whose parents were divorced or 
separated. Parents who never lived together were not included. When children had 
divorced parents, they were fi rst asked about their primary home, i.e., the home 
where they live most of the time. Next, they were asked about living in another 
home. It was assessed how often the child lived in this secondary home and with 
whom. We generated a variable indicating whether the child lives with the father 
and a variable whether the child lives with the mother. Living with a parent is de-
fi ned as living with that parent in the primary household or living with that parent in 
a secondary household for at least half of the time. Cross-classifying these variables 
yields four categories: living primarily with the mother, living primarily with the fa-
ther, living with both in two different households, and living with neither parent. In 
this defi nition, co-parenting means living with each of the two parents for half of 
the time. There are children who stay over once a week with the father but we do 



Father-Child Relations after Divorce in Four European Countries    • 259

not regard this as co-parenting even if the children regard this as their secondary 
household. This will be captured in the contact measure, however. 

Living arrangements were analysed with a multinomial logit model where the 
baseline category is living primarily with the mother. Secondly, we defi ned contact 
as the frequency of face-to-face contact with the biological father. This was recoded 
to missing for children who primarily live with their father (in which case there is 
no variation in contact). Categories were: every day, once or a couple of times per 
week, once or a couple of times per month, less often, and never. The variable was 
analysed with an ordinal logit model. Thirdly, we used a question about how well 
the respondent gets along with the father. The respondent could choose among the 
following options: very well, well, not so well, and not well at all. This variable was 
also analysed with an ordinal logit model. The question about quality was not asked 
if there had never been any contact with the father.

The following socioeconomic variables were used as independent variables: (a) 
a variable indicating whether the mother works for pay, (b) the family status, which 
is composed of four variables: father’s educational level (coded from 0 for no edu-
cation to 3 for tertiary), mother’s educational level (also coded from 0 to 3), father’s 
occupational status, and mother’s occupational status. Education was asked via 
three separate questions, one for each level (primary, secondary, tertiary). It was 
assumed that if a student answers “don’t know” to a level, the parent did not com-
plete that level. A student who answers “don’t know” to all three questions or does 
not answer for other reasons to any question is assigned to missing. The correla-
tion between the father’s and mother’s education, thus measured, was 0.53, which 
gives us confi dence in the operational defi nition. If a mother does not work and the 
child does not list mother’s past occupation, we impute the country-specifi c mean. 
Note that a dummy for employment status is included and most missings are due 
to non-working mothers. Occupations were coded in ISCO-08 (International Socio-
Economic Index of Occupations) and recoded to ISEI status scores (Ganzeboom 
et al. 1992). To create a family status scale, each of the individual status variables 
was standardised and the mean of the four variables was taken. The scale we con-
structed is the average of the valid items. We worked with an index in order to use 
as much information as possible and to reduce the number of missing cases in 
the regression model. Not all children know the socioeconomic characteristics of 
their parents. Missings were 6.9 percent for mother’s education, 15.3 percent for 
father’s education, 6.2 percent for mother’s occupation (for working mothers), and 
22.6 percent for father’s occupation. An alternative would be to use multiple impu-
tation while using individual status components in the models. This is not preferred 
because we have no specifi c hypotheses about which component of status is the 
most infl uential.

The following demographic variables were used: (a) the age of the child, (b) the 
sex of the child, (c) the number of siblings in the (primary) household, (d) whether 
(or not) the mother is living with a new partner. The repartnering variable was in-
cluded in a separate model where children who primarily live with the father and 
children who live with neither parent are excluded. In these cases, no information 
on repartnering is available. 



•    Matthijs Kalmijn260

The following contextual variables are used. First, we include variables for the 
country of destination. Second, we consider a variable that makes a distinction be-
tween “native” children (i.e. native-born children of two native-born parents) and 
“immigrant” children (i.e. the native-born children of immigrants and foreign-born 
children who came to the destination before or during the school ages). We make 
an additional distinction between 1st and 2nd generation children and we pay special 
attention to children of mixed parentage: children who have one native-born parent 
and one foreign-born parent. Finally, we consider immigrant groups, defi ned on the 
basis of the country of birth of the mother. If the mother was native-born, the coun-
try of birth of the father was used. We distinguish seven regions, following often 
used classifi cations of world regions: Southern Europe, Eastern Europe, the larger 
Middle East (including Northern Africa), Asia, Sub-Saharan Africa, Latin America, 
and the Caribbean. The means and standard deviations of the independent vari-
ables are presented in Table 1.

4 Findings

Before presenting the regression results, it is interesting to see how often children 
of divorced parents live with the mother, the father, or both. In Table 2, we pre-
sent these statistics separately for each of the four countries. The percentages are 
weighted to correct for the oversampling of schools with high numbers of immigrant 

Tab. 1: Descriptives of variables used in the analyses (N = 4524)

Mean Standard deviation Min Max

Mother employed 0.77 0.42 0 1

Family status index 0.00 1.00 -3.65 2.73

Daughter versus son 0.52 0.50 0 1

Age of child 14.48 0.67 13 18

Number of siblings 1.29 1.21 0 9

Mother new partner 0.31 0.46 0 1

Germany (vs England) 0.26 0.44 0 1

Sweden (vs England) 0.28 0.45 0 1

Netherlands (vs England) 0.22 0.41 0 1

Immigrant child (1st generation) 0.08 0.28 0 1

Immigrant child (2nd generation) 0.24 0.43 0 1

Both parents foreign 0.13 0.33 0 1

Only father foreign 0.08 0.27 0 1

Only mother foreign 0.04 0.19 0 1

Source: Own calculations based on CILS4EU, 2010/2011



Father-Child Relations after Divorce in Four European Countries    • 261

children. Co-parenting is most common in Sweden: about 36 percent of Swedish 
children have a co-parenting arrangement. In the Netherlands, 18 percent of the 
children live in two parental homes. This corresponds well with an earlier study for 
the Netherlands where it was found that 17 percent of 10-16 year-olds lived in a co-
parenting arrangement after divorce (Spruijt/Duindam 2009). Co-parenting is least 
common in England and Germany (about 10 percent). In all countries, however, the 
majority of children live primarily with their mother. A small minority of children live 
primarily with the father and even fewer children live without a parent.

For children who do not primarily live with their father, we examined how often 
they have been in contact with their father (Table 2). Overall, about 40 percent of the 
children have at least weekly contact with their father, which is a substantial num-
ber. There is much heterogeneity, however. About one in six (18 percent) respond-
ents say they never see their father. This is a substantial minority. The no-contact 
category is smallest in Sweden, larger in the Netherlands, and largest in England 
and Germany where about one in fi ve children never have contact with the father. 

Tab. 2: Descriptive analyses of father-child ties after divorce in four countries

England Germany Netherlands Sweden Total N

Living arrangement
With mother 74.3 70.5 70.6 51.6 70.7 3,091
Co-parenting 10.9 9.8 17.6 36.0 12.8 786
With father 9.5 12.9 6.4 7.6 10.6 389
With neither 5.3 6.8 5.4 4.8 6.0 271

Total 100.0 100.0 100.0 100.0 100.0 4,537

Contact with father (face-to-face)
Never (1) 18.9 20.7 14.2 8.0 18.4 678
Less often (2) 17.7 16.5 8.7 15.8 16.2 717
Monthly (3) 20.9 27.8 32.4 21.6 25.1 941
Weekly (4) 34.4 26.4 39.2 39.3 31.8 1,376
Daily (5) 8.1 8.6 5.5 15.2 8.6 415

Total 100.0 100.0 100.0 100.0 100.0 4,127

Perceived quality
Gets along:
Not well at all (1) 3.3 2.2 3.0 3.0 2.8 128
Not so well (2) 11.8 12.1 11.8 9.5 11.8 432
Well (3) 38.6 34.0 43.9 40.5 37.3 1,484
Very well (4) 46.3 51.6 41.3 47.1 48.2 1,753

Total 100.0 100.0 100.0 100.0 100.0 3,797

Note: Values refer to coding in ordinal logit models.

Source: Own calculations based on CILS4EU, 2010/2011



•    Matthijs Kalmijn262

Whether the relationships in these cases are hostile, estranged, or simply neutral 
and non-existent is not known, nor do we know how these children feel about the 
father that they never see. Note that the percentage of 18 is based on children who 
do not live with their father. When we consider all children of divorced parents, 
16 percent never see their father.

The large majority of children of divorced parents say they have a good or very 
good relationship with their father (Table 2). However, a signifi cant minority – about 
15 percent in each country – does not have a good relationship with the father (based 
on children who have contact). It is also of some interest to examine the association 
between contact and quality. Table 3 shows that there is a positive association, as 
one would expect. The correlation between contact and quality (when coded “as 
is”) is +.33, which is reasonable but not very strong. When we look at the table in 
detail, we see that there is a signifi cant group of children with a mismatch between 
quality and contact. Of children who qualify the relationship as very good, 9 percent 
see their father less than once a month. Of children who qualify the relationship as 
good, 19 percent see their father less than once a month. These numbers point to a 
possible source of deprivation in children’s post-divorce lives. 

Tab. 3: The association between contact with the father and perceived quality: 
Column percentages

Perceived quality (“gets along ...”)
Contact 
(face-to-face) Very well Well Not well Not well at all Total

Daily 459 240 46 6 751
26.2 16.2 10.7 4.7 19.8

Weekly 742 549 95 12 1,398
42.4 37.0 22.0 9.5 36.9

Monthly 401 416 105 17 939
22.9 28.1 24.3 13.4 24.8

Less often 149 277 186 92 704
8.5 18.7 43.1 72.4 18.6

Total 1,751 1,482 432 127 3,792
100 100 100 100 100

Note: Weighted numbers. If there has never been any contact, no question was asked 
about quality.

Source: Own calculations based on CILS4EU, 2010/2011



Father-Child Relations after Divorce in Four European Countries    • 263

Determinants of living arrangements

We now turn to an analysis of the determinants of living arrangements. In Table 4, 
we present the estimates of the multinomial regression model. Living primarily with 
the mother serves as the comparison group. In all regression models, the standard 
errors are corrected for the clustering of students in schools. Moreover, all models 
contain controls for the sample stratum but these effects are not presented in the 
tables.

The socioeconomic variables have signifi cant effects on living arrangements af-
ter divorce. In line with expectations, we fi nd that children from high-status families 
are more likely to have a co-parenting arrangement. There is no effect of socioeco-
nomic status on living primarily with the father. In addition we fi nd that children are 
more likely to have a co-parenting arrangement when the mother works for pay. 
This too is in line with our hypothesis. And again, we fi nd that the effect of employ-
ment on living primarily with the father is not as expected; the effect is negative 
rather than positive. In sum, the socioeconomic effects are as expected for co-par-
enting but they are not as expected for living primarily with the father. Perhaps the 
tendency to not live with the mother – which is a small minority – is dependent on 
more specifi c circumstances such as the mother’s health or personality.

We now discuss the demographic effects on living arrangements. Gender and 
age differences are substantial. Daughters are less likely to live in a co-parenting 

Tab. 4: Multinomial regression of living arrangements after divorce

Co-parenting With father Without parents
b p b p b p

Mother employed .431* (.00) -.820* (.00) -.222 (.15)
Family socioeconomic status .304* (.00) -.014 (.82) .053 (.43)

Daughter versus son -.270* (.00) -.383* (.00) -.294* (.02)
Age of child -.297* (.00) .220* (.01) .118 (.28)
Number of siblings .033 (.37) -.047 (.36) -.254* (.00)

Germany (vs England) .125 (.47) .276 (.13) -.569* (.01)
Sweden (vs England) 1.472* (.00) .378* (.04) -.086 (.65)
Netherlands (vs England) .530* (.00) -.191 (.34) .223 (.21)
Immigrant (1st generation) -2.386* (.00) -.335~ (.09) .300 (.13)
Immigrant (2nd generation) -.867* (.00) -.157 (.25) .319* (.04)
Constant -.041 (.74) .042 (.81) -.137 (.54)

N 4,524
Model Chi2 678.7*

Note: p-values in parentheses. Reference is living primarily with mother. See text for defi -
nition of co-parenting. 
~ p < 0.10, * p < 0.05

Source: Own calculations based on CILS4EU, 2010/2011



•    Matthijs Kalmijn264

arrangement and this applies even more to living primarily with the father. These 
effects are in line with expectations. Older children are less likely to live in a co-
parenting arrangement. We fi nd no effects of the number of siblings except on liv-
ing without parents, which is less common in large families. Presumably, arranging 
such a situation may be more complicated when there are many children.

Contextual differences are substantial as well. Co-parenting is most common 
in Sweden and more common in the Netherlands than in England and Germany. 
Living primarily with the father is more common in Sweden as well. There are also 
differences between natives and immigrants. Children of immigrants are much 
less likely to live with their father after divorce. This applies most strongly to co-
parenting. A clear generational difference is found as well, with second-generation 
children being more likely to have a co-parenting arrangement than fi rst-generation 
immigrants. The contrast between the generations for co-parenting is signifi cant 
(X2 = 13.9, p < .01). This is consistent with standard notions from assimilation the-
ory since co-parenting is a recently developed arrangement in western societies 
to which immigrant groups may adapt as they become more integrated in the host 
society. Finally, second-generation immigrants are more likely to live with neither 
parent after divorce. The more important role of the extended family (e.g. grand-
parents) in immigrant groups may play a role here. Since living with neither parent 
is often associated with poverty (Hynes/Dunifon 2007) and with poorer child well-
being compared to other non-standard families (Sun 2003), this effect may be an 
important disadvantage that immigrant children face while growing up. 

Determinants of contact and quality

The models for contact and quality are presented in Table 5 and 6 respectively. The 
models for contact exclude children who primarily live with their father. In these 
cases, contact is probably daily and does not vary. To examine if the determinants 
affect contact primarily through co-parenting, we also estimated the model again 
while excluding children with co-parenting. In this model, the outcome is contact 
with a non-residential father. The approach is the same for quality in Table 6, except 
that children who never have contact with the father were not asked about the qual-
ity of the tie so these are missing.

We fi rst see that children from high-status families have more frequent contact 
with their divorced father and perceive the quality of the tie to be better. These 
fi ndings are in line with our expectations. The quality of the relationship with the 
father is also better in high-status families that divorce. Contact is more strongly 
affected by status than quality. We also fi nd that when the mother works, the child 
sees the father more often and the quality of the tie with the father is better. This 
last effect is not as expected but may be a product of more frequent contact with 
the father when the mother works: more contact may improve the relationship. Both 
effects on contact – of socioeconomic status and mother’s employment – are still 
signifi cant when we limit the sample to children who primarily live with the mother 
(Model 5). Hence, the effects are not solely a product of more frequent co-parenting 
in high-status families.



Father-Child Relations after Divorce in Four European Countries    • 265

Sons are more likely to have contact with their divorced father than daughters 
and they also perceive the quality of the relationship to be better. What happens 
when we control the effect of child gender on contact for the effects of the quality of 
the relationship with the father and the mother? In a linear regression model for the 
contact scale, the gender effect is b = -.24 without controls and b = -.07 with con-
trols (p = .02). In other words, a large part of the gender difference in contact has to 
do with differences in the way sons and daughters assess the relationship with the 
father. However, even for children who assess the quality of the relationship with 
the father as similar, sons more often see their father than daughters. This suggests 
that the gender difference is in part also due to the preferences and behaviours of 
the father, but the children’s own role seems to be the more important one.

The other demographic variables have mixed effects. There is no effect of the 
number of siblings. Older children are less likely to see their father than younger 
children and more likely to assess the quality as poor. The most interesting fi nding 
is the effect of repartnering. When the mother lives with a new partner, children 
have less frequent contact with their divorced father than when the mother lives 
without a partner. The quality of the relationship with the father is not affected, how-
ever. This provides mixed evidence for the notion that stepfathers are substitutes. In 
terms of contact this may be true, but not in terms of what the father means to the 
child. The negative effect of the stepfather on contact with the biological father is 
the same when we limit the sample to children who primarily live with their mother 
(Model 5).

Are there contextual differences in contact and quality? Table 5 suggests that 
contact with the divorced father is most frequent in Sweden, least frequent in Ger-
many, with England and the Netherlands in between. These differences are partly 
due to differences in living arrangements. When we focus on children who primar-
ily live with their mother (Model 5), Sweden no longer stands out. In this model, 
Germany stands out as a country where divorced fathers have the least frequent 
contact with their children. Note that there are no differences in the quality of the 
father-child relationship among the countries. 

Are there differences in father-child relationships between immigrants and na-
tives, between immigrant generations, and between immigrant groups? In Model 1, 
we fi rst see that immigrant children are less likely to have frequent contact with their 
divorced father than natives. The effect is especially pronounced for the fi rst gen-
eration. This raises the question of whether the gap we fi nd is a phenomenon that is 
limited to the fi rst generation. When looking at the effect for the second generation, 
this appears not to be true. The effect is b = -.31 which shows that second-gener-
ation children have a 27 percent lower odds to have frequent (e.g. at least weekly) 
rather than less frequent (e.g. less than weekly) contact compared to native-born 
children of native parents.1 Similar effects are found for the quality of the relation-
ship. First and second-generation immigrant children evaluate the relationship with 
their divorced father more negatively than native born children of native origins.

1 Formally, this means that the odds of being in contact category j > k rather than in contact cat-
egory j ≤ k are 27 percent lower for second-generation immigrants.



•    Matthijs Kalmijn266

T
a

b
. 

5:
 

O
rd

e
re

d
 lo

g
it

 r
e

g
re

ss
io

n
 o

f 
fa

ce
-t

o
-f

a
ce

 c
o

n
ta

c
t 

w
it

h
 f

a
th

e
r 

a
ft

e
r 

d
iv

o
rc

e

(1
)

(2
)

(3
)

(4
)

(5
)

b
p

b
p

b
p

b
p

b
p

M
o

th
e

r 
e

m
p

lo
y
e

d
.2

0
2
*

(.
0
1)

.2
4

6
*

(.
0

0
)

.2
0

8
*

(.
0
1)

.2
5

9
*

(.
0

0
)

.2
4

5
*

(.
0
1)

F
a
m

ily
 s

o
ci

o
e

co
n

o
m

ic
 s

ta
tu

s
.2

13
*

(.
0

0
)

.2
2

4
*

(.
0

0
)

.2
19

*
(.

0
0

)
.2

0
8
*

(.
0

0
)

.1
9

0
*

(.
0

0
)

D
a
u

g
h

te
r 

v
e

rs
u

s 
so

n
-.

3
7

0
*

(.
0

0
)

-.
3

7
7
*

(.
0

0
)

-.
3

74
*

(.
0

0
)

-.
3

6
4
*

(.
0

0
)

-.
3

0
4
*

(.
0

0
)

A
g

e
 o

f 
ch

ild
-.
16

6
*

(.
0

0
)

-.
16

3
*

(.
0

0
)

-.
16

8
*

(.
0

0
)

-.
2
15

*
(.

0
0

)
-.
14

0
*

(.
0
1)

N
u

m
b

e
r 

o
f 

si
b

lin
g

s
-.

0
19

(.
4

5
)

.0
10

(.
7

0
)

-.
0

2
0

(.
4

2
)

-.
0

2
0

(.
4

3
)

.0
11

(.
7

2
)

M
o

th
e

r 
n

e
w

 p
a
rt

n
e

r
-.

2
8

7
*

(.
0

0
)

-.
3

5
3
*

(.
0

0
)

G
e

rm
a
n

y
 (

v
s 

E
n

g
la

n
d

)
-.

3
3

2
*

(.
0

0
)

-.
3

0
7
*

(.
0

0
)

-.
3

3
1
*

(.
0

0
)

-.
2

6
2
*

(.
0

0
)

-.
3

0
1
*

(.
0

0
)

S
w

e
d

e
n

 (
v

s 
E

n
g

la
n

d
)

.4
13

*
(.

0
0

)
.4

4
0
*

(.
0

0
)

.3
9

9
*

(.
0

0
)

.4
4

5
*

(.
0

0
)

.0
0

6
(.

9
5

)
N

e
th

e
rl

a
n

d
s 

(v
s 

E
n

g
la

n
d

)
.0

9
8

(.
2

8
)

.1
19

(.
2
1)

.0
8

2
(.

3
7

)
.1

51
~

(.
0

9
)

.0
61

(.
5

6
)

Im
m

ig
ra

n
t 

(1
st

 g
e

n
e

ra
ti

o
n

)
-1

.3
41

*
(.

0
0

)
-1

.3
2

9
*

(.
0

0
)

-1
.3

3
2
*

(.
0

0
)

-1
.0

6
8
*

(.
0

0
)

Im
m

ig
ra

n
t 

(2
n

d
 g

e
n

e
ra

ti
o

n
)

-.
3

0
5
*

(.
0

0
)

-.
4

0
0
*

(.
0

0
)

-.
2

74
*

(.
0

0
)

B
o

th
 p

a
re

n
ts

 f
o

re
ig

n
-.
14

8
(.1

8
)

O
n

ly
 f

a
th

e
r 

fo
re

ig
n

-.
6

0
8
*

(.
0

0
)

O
n

ly
 m

o
th

e
r 

fo
re

ig
n

-.
14

1
(.

3
2
)

S
o

u
th

e
rn

 E
u

ro
p

e
a
n

-.
5

0
2
*

(.
0
1)

E
a

st
e

rn
 E

u
ro

p
e

a
n

-.
6

0
3
*

(.
0

0
)

L
a

ti
n

 A
m

e
ri

c
a
n

-.
6

4
0
*

(.
0

0
)

A
si

a
n

-.
5

8
4
*

(.
0

0
)

L
a
rg

e
r 

M
id

d
le

 E
a

st
-.

2
5

6
*

(.
0

4
)

S
u

b
-S

a
h

a
ra

n
 A

fr
ic

a
n

-.
3

61
*

(.
0

2
)

C
a
ri

b
b

e
a
n

-.
4

3
9
*

(.
0

2
)

C
u

tp
o

in
t 

1
-4

.3
0

3
*

(.
0

0
)

-4
.2

9
8
*

(.
0

0
)

-4
.3

41
*

(.
0

0
)

-4
.8

8
8
*

(.
0

0
)

-3
.7

2
1
*

(.
0

0
)

C
u

tp
o

in
t 

2
-3

.2
5

4
*

(.
0

0
)

-3
.2

4
5
*

(.
0

0
)

-3
.2

9
1
*

(.
0

0
)

-3
.8

6
5
*

(.
0

0
)

-2
.6

0
8
*

(.
0

0
)

C
u

tp
o

in
t 

3
-2

.2
2

6
*

(.
0

0
)

-2
.1

9
3
*

(.
0

0
)

-2
.2

6
0
*

(.
0

0
)

-2
.8

5
6
*

(.
0

0
)

-1
.4

9
3
~

(.
0

6
)

C
u

tp
o

in
t 

4
-.

2
0

3
(.

7
9

)
-.

0
8

6
(.

9
1)

-.
2

3
4

(.
7

5
)

-.
8

4
3

(.
2

6
)

.7
18

(.
3

6
)

N
4

,1
14

3
,8

4
8

4
,1

14
4

,1
14

3
,0

6
6

M
o

d
e

l C
h

i2
3

9
1.

0
*

4
2

7.
2
*

4
0

6
.0

*
3
10

.9
*

2
5

2
.8

*

N
o

te
: 

p
-v

al
u

e
s 

in
 p

ar
e
n

th
e
se

s.
 R

e
sp

o
n

d
e
n

ts
 w

h
o

 d
o

 n
o

t 
p

ri
m

ar
ily

 l
iv

e
 w

it
h

 t
h

e
ir

 f
at

h
e
r.
 M

o
d

e
l 

(1
) 

al
so

 e
x
cl

u
d

e
s 

ch
ild

re
n

 l
iv

in
g

 w
it

h
 

n
e
it

h
e
r 

p
ar

e
n

t.
 M

o
d

e
l 

(5
) 

is
 b

as
e
d

 o
n

 c
h

ild
re

n
 p

ri
m

ar
ily

 l
iv

in
g

 w
it

h
 t

h
e
ir

 m
o

th
e
r.
 C

u
tp

o
in

ts
 a

re
 e

st
im

at
e
d

 b
y
 t

h
e
 o

rd
in

al
 l

o
g

it
 

m
o

d
e
ls

 a
n

d
 c

an
 b

e
 u

se
d

 t
o

 c
al

cu
la

te
 o

d
d

s 
ra

ti
o

s.
~

 p
 <

 0
.1

0
, 
* 

p
 <

 0
.0

5

S
o

u
rc

e
: 

O
w

n
 c

al
cu

la
ti

o
n

s 
b

as
e
d

 o
n

 C
IL

S
4
E

U
, 

2
0
1
0
/2

0
1
1



Father-Child Relations after Divorce in Four European Countries    • 267

T
a

b
. 

6:
 

O
rd

e
re

d
 lo

g
it

 r
e

g
re

ss
io

n
 o

f 
p

e
rc

e
iv

e
d

 q
u

a
lit

y
 o

f 
fa

th
e

r-
ch

ild
 r

e
la

ti
o

n
sh

ip
 a

ft
e

r 
d

iv
o

rc
e

(1
)

(2
)

(3
)

(4
)

b
p

b
p

b
p

b
p

M
o

th
e

r 
e

m
p

lo
y
e

d
.1

3
2

(.1
0

)
.1

9
0
*

(.
0

5
)

.1
3

5
~

(.
0

9
)

.1
3

7
~

(.
0

9
)

F
a
m

ily
 s

o
ci

o
e

co
n

o
m

ic
 s

ta
tu

s
.1

41
*

(.
0

0
)

.1
8

3
*

(.
0

0
)

.1
4

3
*

(.
0

0
)

.1
4

0
*

(.
0

0
)

D
a
u

g
h

te
r 

v
e

rs
u

s 
so

n
-.

5
6

9
*

(.
0

0
)

-.
5

7
8
*

(.
0

0
)

-.
5

6
9
*

(.
0

0
)

-.
5

71
*

(.
0

0
)

A
g

e
 o

f 
ch

ild
-.
12

4
*

(.
0

2
)

-.
13

8
*

(.
0

2
)

-.
12

6
*

(.
0

2
)

-.
13

2
*

(.
0

2
)

N
u

m
b

e
r 

o
f 

si
b

lin
g

s
.0

2
5

(.
3

4
)

.0
17

(.
5

9
)

.0
2

4
(.

3
5

)
.0

2
8

(.
2

9
)

M
o

th
e

r 
n

e
w

 p
a
rt

n
e

r
.1

13
(.1

2
)

G
e

rm
a
n

y
 (

v
s 

E
n

g
la

n
d

)
.1

0
2

(.
2

7
)

.0
8
1

(.
41

)
.1

0
3

(.
2

6
)

.1
0

5
(.

2
6

)
S

w
e

d
e

n
 (

v
s 

E
n

g
la

n
d

)
.1

0
0

(.
2

9
)

.0
3

9
(.

6
9

)
.0

9
5

(.
3
1)

.0
9

9
(.

3
0

)
N

e
th

e
rl

a
n

d
s 

(v
s 

E
n

g
la

n
d

)
-.

0
5

7
(.

5
6

)
-.

0
7

8
(.

4
5

)
-.

0
6

2
(.

5
3

)
-.

0
3

8
(.

7
0

)
Im

m
ig

ra
n

t 
(1

st
 g

e
n

e
ra

ti
o

n
)

-.
3
11

*
(.

0
2
)

-.
3

2
7
*

(.
0

3
)

-.
3

0
5
*

(.
0

2
)

Im
m

ig
ra

n
t 

(2
n

d
 g

e
n

e
ra

ti
o

n
)

-.
17

9
*

(.
0

3
)

-.
12

9
(.1

5
)

B
o

th
 p

a
re

n
ts

 f
o

re
ig

n
-.
12

3
(.

2
6

)
O

n
ly

 f
a

th
e

r 
fo

re
ig

n
-.

2
6

4
*

(.
0

4
)

O
n

ly
 m

o
th

e
r 

fo
re

ig
n

-.
17

5
(.

2
3

)
S

o
u

th
e

rn
 E

u
ro

p
e

a
n

-.
0

8
1

(.
7

2
)

E
a

st
e

rn
 E

u
ro

p
e

a
n

-.
0

4
0

(.
7

7
)

L
a

ti
n

 A
m

e
ri

c
a
n

-.
2

0
8

(.
4

3
)

A
si

a
n

-.
12

3
(.

4
7

)
L

a
rg

e
r 

M
id

d
le

 E
a

st
-.

2
4

5
~

(.
0

6
)

S
u

b
-S

a
h

a
ra

n
 A

fr
ic

a
n

-.
2

3
3

(.1
7

)
C

a
ri

b
b

e
a
n

-.
2
13

(.
2

6
)

C
u

tp
o

in
t 

1
-5

.4
5

2
*

(.
0

0
)

-5
.5

2
1
*

(.
0

0
)

-5
.4

8
3
*

(.
0

0
)

-5
.5

4
6
*

(.
0

0
)

C
u

tp
o

in
t 

2
-3

.8
2

7
*

(.
0

0
)

-3
.8

9
3
*

(.
0

0
)

-3
.8

5
8
*

(.
0

0
)

-3
.9

2
1
*

(.
0

0
)

C
u

tp
o

in
t 

3
-1

.8
6

6
*

(.
0

2
)

-1
.9

41
*

(.
0

2
)

-1
.8

9
6
*

(.
0

2
)

-1
.9

6
0
*

(.
0

2
)

N
3

,7
8

8
3

,1
8

4
3

,7
8

8
3

,7
8

8
M

o
d

e
l C

h
i2

16
5

.1
*

17
0

.0
*

17
2

.5
*

16
4

.5
*

N
o

te
: 

p
-v

al
u

e
s 

in
 p

ar
e
n

th
e
se

s.
 C

h
ild

re
n

 w
h

o
 n

e
v
e
r 

se
e
 t

h
e
ir

 f
at

h
e
r 

ar
e
 e

x
cl

u
d

e
d

. 
M

o
d

e
l (

1
.)

 a
ls

o
 e

x
cl

u
d

e
s 

ch
ild

re
n

 w
h

o
 li

v
e
 w

it
h

 p
ri

-
m

ar
ily

 t
h

e
ir

 f
at

h
e
r 

o
r 

w
it

h
 n

e
it

h
e
r 

p
ar

e
n

t.
~

 p
 <

 0
.1

0
, 
* 

p
 <

 0
.0

5

S
o

u
rc

e
: 

O
w

n
 c

al
cu

la
ti

o
n

s 
b

as
e
d

 o
n

 C
IL

S
4
E

U
, 

2
0
1
0
/2

0
1
1



•    Matthijs Kalmijn268

Does it matter which parent is foreign-born? In Model 3, we make a distinction 
between three types of (second-generation) children: (a) both parents born abroad 
(n = 579), (b) only the father foreign-born (n = 347), and (c) only the mother foreign-
born (n = 175). Categories (b) and (c) are the children of an ethnically mixed mar-
riage. When only the mother is foreign-born, there is no difference with natives. 
When both parents are foreign-born, the contrast with natives is signifi cant but 
small. It is primarily when the father is foreign-born and the mother is native-born 
that we fi nd less frequent post-divorce contact compared to natives. In other words, 
children in mixed marriages where the father is foreign-born tend to lose contact 
with the father more frequently. For quality, we see the exact same pattern of re-
sults: children from mixed marriages have poorer ties to their divorced father but 
this is only true when the father is foreign and the mother is native. Remember, how-
ever, that fi rst-generation immigrants are also less likely to have frequent contact 
and evaluate the relationships with their divorced father more negatively. In other 
words, there is an effect of intermarriage but also an effect of immigrant status.

To what extent are there differences among the various immigrant groups? To 
examine this, we distinguish broad regions in Model 4. We fi rst conduct an extra 
test where the effects for all regions are the same. This hypothesis cannot be re-
jected (X2 = 5.5, p =.48). Hence, the difference between natives and immigrants 
does not depend on the subgroup we look at. The individual effects vary from -.36 
for Sub-Saharan African immigrants to -.64 for Latin American immigrants. These 
are not small differences but apparently not large enough to be signifi cant with the 
present sample size. For quality, we fi nd the same pattern but the group-specifi c ef-
fects are smaller and often not signifi cant.2 The test for equal effects across groups 
cannot be rejected (X2 = 2.1, p = .91). Hence, there is more evidence for a general 
immigrant background effect than for a group-specifi c effect.

Country differences in effects?

So far, we pooled the four countries and included a main effect. This assumes that 
the effects of the other variables are similar across settings. In the last part of the 
analyses, we examine this assumption by presenting models for each country sepa-
rately. We build one new model that summarises the main patterns. More specifi -
cally, we include all respondents regardless of living arrangement and look at the 
frequency of contact. This model takes into account that contact can be frequent 
because of a favourable living arrangement and therefore it combines the results for 
living arrangements and contact frequency. Table 7 presents the country-specifi c 
estimates. We abstain from testing the differences across countries and simply de-
scribe the patterns that we see, focusing on whether or not an effect is signifi cant. 
The effects of the family’s socioeconomic status exist in all countries but the effect 
of a mother’s employment is present only in England and Germany. Gender differ-

2 The general immigrant effect is signifi cant for quality but the region-specifi c effects “distribute” 
this effect among many dummy variables, which on their own are often not signifi cant.



Father-Child Relations after Divorce in Four European Countries    • 269

T
a

b
. 

7:
 

O
rd

e
re

d
 lo

g
it

 r
e

g
re

ss
io

n
 o

f 
fa

th
e

r-
ch

ild
 c

o
n

ta
c
t 

a
ft

e
r 

d
iv

o
rc

e
 f

o
r 

a
ll 

liv
in

g
 a

rr
an

g
e

m
e

n
ts

 b
y
 c

o
u

n
tr

y

E
n

g
la

n
d

G
e

rm
a
n

y
N

e
th

e
rl

a
n

d
s

S
w

e
d

e
n

b
p

b
p

b
p

b
p

M
o

th
e

r 
e

m
p

lo
y
e

d
.4

16
*

(.
0

0
)

.3
0

0
*

(.
0

4
)

.1
2

6
(.

4
5
)

.0
0
7

(.
9
7

)
F
am

ily
 s

o
ci

o
e

co
n

o
m

ic
 s

ta
tu

s
.1

4
2
*

(.
0

3
)

.2
4

0
*

(.
0

0
)

.2
41

*
(.

0
0

)
.2

0
7
*

(.
0

0
)

D
au

g
h

te
r 

v
e

rs
u

s 
so

n
-.

2
8

8
*

(.
0

3
)

-.
2

8
6
*

(.
01

)
-.

5
0

3
*

(.
0

0
)

-.
3

9
9
*

(.
0

0
)

A
g

e
 o

f 
ch

ild
-.

0
9

3
(.

4
6
)

-.
2

4
7
*

(.
0

0
)

-.
21

4
*

(.
0

2
)

-.
0

2
7

(.
9

2
)

N
u

m
b

e
r 

o
f 

si
b

lin
g

s
-.

01
8

(.
74

)
.0

4
0

(.
4

5
)

-.
0

0
3

(.
9
7

)
.0

15
(.

74
)

M
o

th
e

r 
n

e
w

 p
a
rt

n
e

r
-.1

7
0

(.
21

)
-.

5
3

0
*

(.
0

0
)

-.
31

4
*

(.
0

0
)

-.
3

2
4
*

(.
0

0
)

Im
m

ig
ra

n
t 

v
s 

n
a
ti

v
e

-.
7

9
7
*

(.
0

0
)

-.
51

6
*

(.
0

0
)

-.
6

3
9
*

(.
0

0
)

-.
51

9
*

(.
0

0
)

C
u

tp
o

in
t 

1
-2

.7
8

9
(.1

3
)

-5
.3

6
3
*

(.
0

0
)

-5
.3

9
5
*

(.
0

0
)

-3
.2

2
2

(.
3

8
)

C
u

tp
o

in
t 

2
-1

.7
3

6
(.

3
3

)
-4

.2
8

4
*

(.
0

0
)

-4
.5

8
3
*

(.
0

0
)

-2
.0

16
(.

5
8

)
C

u
tp

o
in

t 
3

-.
9

0
3

(.
6

2
)

-3
.1

8
2
*

(.
01

)
-3

.4
2

4
*

(.
01

)
-1

.0
0

3
(.

7
8

)
C

u
tp

o
in

t 
3

1.
21

4
(.

5
0

)
-1

.1
41

(.
3

3
)

-1
.2

9
4

(.
3

5
)

.9
10

(.
8

0
)

N
9

8
7

1,
0

4
2

9
2

5
1,

16
0

M
o

d
e

l C
h

i2
7
6

.4
*

9
6

.2
*

6
9

.9
*

61
.8

*

N
o

te
: 

p
-v

al
u

e
s 

in
 p

ar
e
n

th
e
se

s.
 A

ll 
ch

ild
re

n
 a

re
 in

cl
u

d
e
d

, 
re

g
ar

d
le

ss
 o

f 
liv

in
g

 a
rr

an
g

e
m

e
n

t.
 C

o
n

ta
ct

 f
re

q
u

e
n

cy
 w

as
 a

ls
o

 a
sk

e
d

 f
o

r 
ch

il-
d

re
n

 li
v
in

g
 w

it
h

 t
h

e
ir

 f
at

h
e
r.

* 
p

 <
 0

.0
5

S
o

u
rc

e
: 

O
w

n
 c

al
cu

la
ti

o
n

s 
b

as
e
d

 o
n

 C
IL

S
4
E

U
, 

2
0
1
0
/2

0
1
1



•    Matthijs Kalmijn270

ences are consistent across countries but stepfathers have a negative effect in all 
countries except England. Finally, in all four countries, immigrant children less often 
have contact with their father than native children.

5 Conclusion

This paper has described the father-child relationship after divorce in four European 
countries, focusing on children who are about 14 years of age. Although there has 
been much research on these relationships, most of the research is American and 
little is known about the situation in Europe. Moreover, there is considerable hetero-
geneity in post-divorce relationships and fi ndings about the causes of this heteroge-
neity are not consistent. Examples of such factors are the socioeconomic status of 
the family, the national context, and differences between natives and immigrants. 
Recent studies have made comparisons across countries, but these analyses focus 
on adult children who are no longer living at home. Studies of ethnic and racial 
differences in family functioning have been done but these have rarely looked at 
effects of divorce on father-child relationships.

Using nationally representative data on secondary school children in four coun-
tries, we fi nd that there are considerable differences among divorced families. Many 
children have good or very good ties to their father after divorce and a large group 
of children either lives with the father in a co-parenting arrangement or has at least 
weekly contact. However, there also is a non-trivial minority of children who do not 
see their father at all (about one in six). Moreover, of those who do have contact, a 
substantial minority does not have a good relationship with the father. We suspect 
that this will be perceived as a problem by many fathers and children. In this sense, 
poor father-child relationships are one of the more problematic consequences of 
the divorce revolution.

To understand heterogeneity, we looked at socioeconomic factors, demographic 
differences, and contextual variation. High-status families generally suffer less from 
the adverse consequences of divorce for father-child relationships. When measured 
by the parent’s educational level and their occupational status, socioeconomic sta-
tus is positively associated with co-parenting, with frequent contact, and with the 
quality of the relationship. The same applies to mother’s employment, which is also 
positively related to father-child ties, at least in two of the four countries. We inter-
pret these fi ndings in terms of the greater role that fathers play in children’s lives 
before divorce in such families, more awareness of the consequences of divorce 
for children, and better coping strategies to deal with the practical and emotional 
problems after divorce.

We also found considerable demographic differences. Previous research has not 
consistently found gender differences but our fi ndings are rather clear: sons more 
often live with the father, have more frequent contact if they do not live with him, 
and they also more often assess the relationship as positive. In addition, stepfathers 
“compete” with biological fathers in terms of contact frequency but the quality of 
the tie to the father is not negatively affected if the mother has a new partner. This 



Father-Child Relations after Divorce in Four European Countries    • 271

suggests that the substitution effect of the stepfather is probably more practical 
than emotional: there is less time to visit and less need for support if the stepfather 
already helps out with a problem. 

The context matters for understanding father-child relationships as well. Of the 
four countries, Sweden seems to have the most favourable post-divorce father-
child relationships and this is largely due to more frequent co-parenting. Germany 
seems to have the least favourable post-divorce relationships, at least in terms of 
having less frequent weekly contact. These results do not support an institutionali-
sation perspective since England and Sweden are both high-divorce countries but 
ties to fathers are much stronger in Sweden than in England. The special position of 
Sweden, and especially the fi nding on co-parenting, seems more consistent with a 
gender-role perspective. Obviously, larger samples of countries are needed to test 
these ideas conclusively. One previous study of father-child contact among adult 
children in ten European countries found a similar pattern: stronger negative ef-
fects of divorce on father-child contact in less gender egalitarian countries (Kalmijn 
2008). Our fi ndings are consistent with this earlier study.

Contextual differences were also analysed by looking at immigrant children. Im-
migrant children less often live  with the father after divorce and if they live with 
their mother, they less often have regular contact with the father. The weaker role of 
the immigrant father after divorce is observed for all investigated major immigrant 
groups. Several reasons were suggested to explain the differences. First, immigrant 
families tend to be more traditional in their gender roles so that fathers invest less in 
their children during marriage. Second, in some immigrant families, divorce is met 
with disapproval. This leads to a more selective group of parents who divorce (e.g. 
more confl ict-ridden families) and can also lead to more complications after divorce 
(e.g. more sanctions from the community). Finally, economic considerations can 
play a role. Immigrant men often have a weaker economic position in society (e.g. 
lower income) and this makes it more diffi cult for them to provide a comfortable 
(second) home for a child after divorce. Alimony payments can also be more uncer-
tain, and since a weak fi nancial tie is often associated with a reduction in contact, 
this can be part of the explanation as well (Seltzer et al. 1989).

What do the differences tell us about these explanations? First, we observe that 
second-generation immigrants are more likely to see their divorced father than fi rst-
generation immigrants. This is in line with a cultural perspective. Second-genera-
tion immigrants are more accustomed to Western European gender roles and may 
become more accepting of divorce (De Graaf et al. 2011). Secondly, we found more 
evidence for similarities than for differences among the major immigrant groups in 
the four countries. Immigrants from African and Caribbean origins have less fre-
quent contact with the non-resident father, just as immigrants from predominantly 
Muslim origin countries. In Muslim groups, divorce is disapproved of, but in African 
and Caribbean groups, divorce and non-marital childbearing are more accepted. 
That effects are nonetheless similar may be due to the matrifocal orientation of 
African and Caribbean groups (Sarkisian/Gerstel 2004).

The children of mixed marriages seem to occupy a special position. Children 
of mixed marriages in which the father is foreign-born and the mother is native 



•    Matthijs Kalmijn272

are more prone to lose contact with their father. Perhaps when mixed marriages 
break up, the immigrant father becomes disconnected from the mother’s network 
because without being tied to the mother through marriage it may be diffi cult to 
bridge ethnic group boundaries. Intermarriage is often believed to connect groups 
and the breakup of a mixed marriage may lead to the opposite effect with unintend-
ed negative effects for one parent. New studies need to look into this issue in more 
detail. The possibility that the father moves back to his origin country after divorce 
needs to be considered as well.

We conclude with some caveats about our data. Some information is missing 
as the CILS4EU was a general survey and not specifi cally targeted to family schol-
ars. This means that we have no information on whether the father repartnered (al-
though we do know if the mother repartnered), we do not know the year of divorce, 
and we do not know where the father is living (for example, if he moved back to the 
origin country). The design is cross-sectional, which means that we do not know if 
the relationships of children and parents were already poor before the breakup. We 
also do not know the temporal order of some of the variables (e.g. co-parenting and 
mother’s employment). It is clear that longitudinal data could yield more stringent 
tests of our hypotheses but it is rather diffi cult practically to collect longitudinal data 
with suffi cient numbers of parents who divorce in between waves. There are no 
data where the exact same research design was used in four European countries. 
In this sense, the data used here are unique and provide an important fi rst look at 
the problem.3 The main goal of the paper was to describe father-child relationships 
after divorce using nationally representative data on children. Moreover, we were 
able to show that there are important social differences in these relationships, re-
lated to the national context where families live, their socioeconomic status, and 
their immigrant status.

References

Adamsons, Kari; Johnson, Sara K. 2013: An updated and expanded meta-analysis of 
nonresident fathering and child well-being. In: Journal of Family Psychology 27,4: 
589-599 [doi: 10.1037/a0033786].

Albertini, Marco; Garriga, Anna 2011: The effect of divorce on parent-child contacts: 
Evidence on two declining effect hypotheses. In: European Societies 13,2: 257-278 
[doi: 10.1080/14616696.2010.483002].

Amato, Paul R. 2001: Children of divorce in the 1990s: An update of the amato and 
keith (1991) meta-analysis. In: Journal of Family Psychology 15,3: 355-370 [doi: 
10.1037/0893-3200.15.3.355].

Aquilino, William S. 2006: Noncustodial father-child relationship from adolescence into 
young adulthood. In: Journal of Marriage and Family 68,4: 929-946 [doi: 10.1111/j.1741-
3737.2006.00305.x].

3 The CILS4EU is an annual panel but there are few separations over the years.



Father-Child Relations after Divorce in Four European Countries    • 273

Arditti, Joyce A.; Keith, Timothy Z. 1993: Visitation frequency, child support payment, 
and the father-child relationship post-divorce. In: Journal of Marriage and the Family 
55: 699-712 [doi: 10.2307/353350].

Bernardi, Fabrizio; Radl, Jonas 2014: The long-term consequences of parental divorce 
for children’s educational attainment. In: Demographic Research 30: 1653-1680.

Bianchi, Suzanne M.; Casper, Lynne M. 2002: Continuity and change in the American 
family. Thousand Oaks: Sage.

Bianchi, Suzanne M. 2000: Maternal employment and time with children: Dramatic 
change or surprising continuity? In: Demography 37,4: 401-414 [doi: 10.2307/2648068].

Cheadle, Jacob E.; Amato, Paul R. 2011: A quantitative assessment of Lareau’s qualita-
tive conclusions about class, race, and parenting. In: Journal of Family Issues 32,5: 
679-706 [doi: 10.1177/0192513X10386305].

Cheadle, Jacob E.; Amato, Paul R.; King, Valarie 2010: Patterns of nonresident father 
contact. In: Demography 47,1: 205-225 [doi: 10.1353/dem.0.0084].

Cooksey, Elizabeth N.; Craig, Patricia H. 1998: Parenting from a distance: The effects of 
paternal characteristics on contact between nonresidential fathers and their children. 
In: Demography 35: 187-200 [doi: 10.2307/3004051].

Cooney, Teresa M. 1994: Young adults’ relations with parents: The infl uence of recent 
parental divorce. In: Journal of Marriage and Family 56: 45-56 [doi: 10.2307/352700]. 

Daatland, Svein O. 2007: Marital history and intergenerational solidarity: The impact of 
divorce and unmarried cohabitation. In: Journal of Social Issues 63,4: 809-825 [doi: 
10.1111/j.1540-4560.2007.00538.x].

De Graaf, Paul M. et al. 2011: Sociaal-culturele verschillen tussen Turken, Marokkanen 
en autochtonen: eerste resultaten van de Nederlandse LevensLoop Studie (NELLS). 
In: Bevolkingstrends 4: 61-70.

De Graaf, Paul M.; Fokkema, Tineke 2007: Contacts between divorced and non-divorced 
parents and their adult children in the Netherlands: An investment perspective. In: 
European Sociological Review 23,2: 263-277 [doi: 10.1093/esr/jcl032].

Edin, Kathryn; Kefalas, Maria 2005: Promises I can keep: Why poor women put mother-
hood before marriage. Berkeley: University of California Press.

Emmen, Rosanneke A. G. et al. 2013: Socioeconomic status and parenting in ethnic mi-
nority families: Testing a minority family stress model. In: Journal of Family Psychol-
ogy 27,6: 896-904 [doi: 10.1037/a0034693].

Ganzeboom, Harry B.G.; De Graaf, Paul M.; Treiman, Donald J. 1992: A standard inter-
national socio-economic index of occupational status. In: Social Science Research 
21,1: 1-56 [doi: 10.1016/0049-089X(92)90017-B].

Gauthier, Anne H.; De Gusti, Berenice 2012: The time allocation to children by parents 
in Europe. In: International Sociology 27,6: 827-845 [doi: 10.1177/0268580912443576].

Grundy, Emily; Read, Sanna 2012: Social contacts and receipt of help among older peo-
ple in england: Are there benefi ts of having more children? In: Journals of Gerontol-
ogy Series B-Psychological Sciences and Social Sciences 67,6: 742-754 [doi: 10.1093/
geronb/gbs082].

Heath, Anthony F.; Rothon, Catherine; Kilpi, Elina 2008: The second generation in West-
ern Europe: Education, unemployment, and occupational attainment. In: Annual Re-
view of Sociology 34: 211-235 [doi: 10.1146/annurev.soc.34.040507.134728].



•    Matthijs Kalmijn274

Hofferth, Sandra L.; Forry, Nicole D.; Peters, H. Elizabeth 2010: Child support, father–
child contact, and preteens’ involvement with nonresidential fathers: Racial/ethnic dif-
ferences. In: Journal of Family and Economic Issues 31,1: 14-32 [doi: 10.1007/s10834-
009-9172-9].

Hook, Jennifer L.; Wolfe, Christina M. 2012: New fathers? Residential fathers’ time 
with children in four countries. In: Journal of Family Issues 33,4: 415-450 [doi: 
10.1177/0192513X11425779].

Hynes, Kathryn; Dunifon, Rachel 2007: Children in no-parent households: The continu-
ity of arrangements and the composition of households. In: Children and Youth Ser-
vices Review 29,7: 912-932 [doi: 10.1016/j.childyouth.2007.02.001].

Juby, Heather et al. 2007: Nonresident fathers and children – Parents’ new un-
ions and frequency of contact. In: Journal of Family Issues 28,9: 1220-1245 [doi: 
10.1177/0192513X07302103].

Juby, Heather; Le Bourdais, Céline; Marcil-Gratton, Nicole 2005: Sharing roles, sharing 
custody? Couples’ characteristics and children’s living arrangements at separation. In: 
Journal of Marriage and Family 67,1: 157-172 [doi: 10.1111/j.0022-2445.2005.00012.x].

Kalmijn, Matthijs 2003: Country differences in sex-role attitudes: Cultural and economic 
explanations. In: Arts, Wil; Hagenaars, Jacques; Halman, Loek (Eds.): The cultural 
diversity of European unity: Findings, explanations and refl ections from the European 
Values Study. Leiden: Brill: 311-337.

Kalmijn, Matthijs 2007: Gender differences in the effects of divorce, widowhood and 
remarriage on intergenerational support: Does marriage protect fathers? In: Social 
Forces 85,3: 1079-1104 [doi: 10.1353/sof.2007.0043].

Kalmijn, Matthijs 2008: The effects of separation and divorce on parent-child relation-
ships in ten European countries. In: Saraceno, Chiara (Ed.): Families, Ageing And 
Social Policy: Intergenerational Solidarity in European Welfare States. Cheltenham: 
Edward Elgar: 170-193.

Kalmijn, Matthijs 2010: Racial differences in the effects of parental divorce and separa-
tion on children: Generalizing the evidence to a European case. In: Social Science 
Research 39,5: 845-856 [doi: 10.1016/j.ssresearch.2010.05.002].

Kalter, Frank et al. 2012: The Children of Immigrants Longitudinal Survey in Four Euro-
pean Countries (CILS4EU): Motivation, aims, and design. Unpublished manuscript. 
Mannheim.

King, Valarie 2009: Stepfamily formation: Implications for adolescent ties to mothers, 
nonresident fathers, and stepfathers. In: Journal of Marriage and Family 71,4: 954-968 
[doi: 10.1111/j.1741-3737.2009.00646.x].

King, Valarie; Harris, Kathleen Mullan; Heard, Holly E. 2004: Racial and ethnic diversity 
in nonresident father involvement. In: Journal of Marriage and Family 66: 1-21 [doi: 
10.1111/j.1741-3737.2004.00001.x].

Knudsen, Knud; Wærness, Kari 2008: National context and spouses’ housework in 34 
countries. In: European Sociological Review 24,1: 97-113 [doi: 10.1093/esr/jcm037].

Koo, Helen; Sunchindran, C. M.; Griffi th, Janet D. 1984: The effects of children on di-
vorce and re-marriage: A multivariate analysis of life table probabilities. In: Population 
Studies 38,3: 451-471 [doi: 10.1080/00324728.1984.10410303].

Kraaykamp, Gerbert 2002: Cumulative Advantages and Inequality in Lifestyle: A Dutch 
Description of Distinction in Taste. In: The Netherlands’ Journal of Social Sciences 38: 
121-43.



Father-Child Relations after Divorce in Four European Countries    • 275

Lichter, Daniel T. et al. 1992: Race and the retreat from marriage: A shortage of marriage-
able men? In: American Sociological Review 57,6: 781-799 [doi: 10.2307/2096123].

Lin, I-Fen 2008: Consequences of parental divorce for adult children’s support of their 
frail parents. In: Journal of Marriage and Family 70,1: 113-128 [doi: 10.1111/j.1741-
3737.2007.00465.x].

Mandemakers, Jornt J.; Kalmijn, Matthijs 2014: Do mother’s and father’s education con-
dition the impact of parental divorce on child well-being? In: Social Science Research 
44: 187-199 [doi: 10.1016/j.ssresearch.2013.12.003].

Manning, Wendy D.; Smock, Pamela J. 1999: New families and nonresident father-child 
visitation. In: Social Forces 78,1: 87-116 [doi: 10.1093/sf/78.1.87].

Manning, Wendy D.; Stewart, Susan D.; Smock, Pamela J. 2003: The complexity of fa-
thers’ parenting responsibilities and involvement with nonresident children. In: Jour-
nal of Family Issues 24,5: 645-667 [doi: 10.1177/0192513X03252573].

Morgan, S. Philip et al. 1993: Racial differences in household and family structure 
at the turn of the century. In: American Journal of Sociology 98,4: 798-828 [doi: 
10.1086/230090].

Mortelmans, Dimitri et al. 2011: Scheiding in vlaanderen. Leuven: Acco.

Norris, Pippa; Inglehart, Ronald 2002: Islamic culture and democracy: Testing 
the ‘clash of civilizations’ thesis. In: Comparative Sociology 1,3: 235-263 [doi: 
10.1163/156913302100418592].

Oppenheimer, Valerie Kincade 2003: Cohabiting and marriage during young men’s career-
development process. In: Demography 40,1: 127-149 [doi: 10.1353/dem.2003.0006].

Pels, Trees; Nijsten, Cecile 2003: Myths and realities of diversity in child rearing and 
parent-child relations: Non-indigeneous compared to indigeneous families in the 
Netherlands. In: Hagendoorn, Louk; Veenman, Justus; Vollebergh, Wilma A. M. (Eds.): 
Structural integration and cultural orientations in indigeneous and non-indigeneous 
Dutch citizens. Ashgate: Alderslot: 63-90.

Powell, Brian; Downey, Douglas B. 1997: Living in single-parent households: An inves-
tigation of the same-sex hypothesis. In: American Sociological Review 62,4: 521-539  
[doi: 10.2307/2657424].

Pryor, Jan; Rodgers, Bryan 2001: Children in changing families: Life after parental sepa-
ration. Malden: Oxford.

Ryan, Rebecca M.; Kalil, Ariel; Ziol-Guest, Kathleen M. 2008: Longitudinal patterns of 
nonresident fathers’ involvement: The role of resources and relations. In: Journal of 
Marriage and Family 70,4: 962-977 [doi: 10.1111/j.1741-3737.2008.00539.x].

Sarkisian, Natalia; Gerstel, Naomi 2004: Kin support among blacks and whites: Race 
and family organization. In: American Sociological Review 69,6: 812-837 [doi: 
10.1177/000312240406900604].

Sayer, Liana C.; Gauthier, Anne H.; Furstenberg, Frank F. 2004: Educational differences 
in parents’ time with children: Cross-national variations. In: Journal of Marriage and 
Family 66,5: 1152-1169 [doi: 10.1111/j.0022-2445.2004.00084.x].

Seltzer, Judith A.; Bianchi, Suzanne M. 2013: Demographic change and parent-child 
relationships in adulthood. In: Annual Review of Sociology 39: 275-290 [doi: 10.1146/
annurev-soc-071312-145602].

Seltzer, Judith A.; Schaeffer, Nora Cate; Charng, Hong-Wen 1989: Family ties after di-
vorce: The relationship between visiting and paying child-support. In: Journal of Mar-
riage and Family 51,4: 1013-1031 [doi: 10.2307/353213].



•    Matthijs Kalmijn276

Sobolewski, Juliana M.; Amato, Paul R. 2007: Parents’ discord and divorce, parent-child 
relationships and subjective well-being in early adulthood: Is feeling close to two par-
ents always better than feeling close to one? In: Social Forces 85,3: 1105-1124 [doi: 
10.1353/sof.2007.0056].

Spruijt, Ed; Duindam, Vincent 2009: Joint physical custody in the Netherlands and 
the well-being of children. In: Journal of Divorce and Remarriage 51,1: 65-82 [doi: 
10.1080/10502550903423362]. 

Stack, Carol B. 1974: All our kin: Strategies for survival in a black community. New York: 
Harper & Row.

Stephens, Linda S. 1996: Will Johnny see daddy this week? An empirical test of three 
theoretical perspectives of postdivorce contact. In: Journal of Family Issues 17,4: 466-
494 [doi: 10.1177/019251396017004003].

Sun, Yongmin 2003: The well-being of adolescents in households with no biologi-
cal parents. In: Journal of Marriage and Family 65,4: 894-909 [doi: 10.1111/j.1741-
3737.2003.00894.x].

Swiss, Liam; Le Bourdais, Céline 2009: Father-child contact after separation the in-
fl uence of living arrangements. In: Journal of Family Issues 30,5: 623-652 [doi: 
10.1177/0192513X08331023].

Van Damme, Maike 2010: Beyond marriage: Women’s economic independence and 
separation in comparative perspective. Tilburg: Ph.D. Thesis.

Van Tubergen, Frank 2004: The integration of immigrants in cross-national perspective: 
Origin, destination, and community effects. Amsterdam: Thesis.

Waldfogel, Jane; Han, Wen-Jui; Brooks-Gunn, Jeanne 2002: The effects of early mater-
nal employment on child cognitive development. In: Demography 39,2: 369-392 [doi: 
10.1353/dem.2002.0021].

Yaman, Ayşe et al. 2010: Parenting in an individualistic culture with a collectivistic cul-
tural background: The case of Turkish immigrant families with toddlers in the Nether-
lands. In: Journal of Child and Family Studies 19,5: 617-628 [doi: 10.1007/s10826-009-
9346-y].

Yeung, W. Jean et al. 2001: Children’s time with fathers in intact families. In: Journal of 
Marriage and the Family 63,1: 136-154 [doi: 10.1111/j.1741-3737.2001.00136.x].

Zander, Joep 2012: Parental alienation as an outcome of paternal discrimination. In: New 
Male Studies 1,2: 49-62.

Date of submission: 07.04.2014  Date of acceptance: 12.06.2015

Prof. Dr. Matthijs Kalmijn (). University of Amsterdam, Faculty of Social and 
Behavioural Sciences. Amsterdam, The Netherlands. E-mail: M.Kalmijn@uva.nl
URL: http://www.uva.nl/en/about-the-uva/organisation/staff-members/item/m.kalmijn.
html?f=Kalmijn



Published by / Herausgegeben von
Prof. Dr. Norbert F. Schneider

Federal Institute for Population Research 
D-65180 Wiesbaden / Germany

Managing Editor /
Verantwortlicher Redakteur
Frank Swiaczny

Assistant Managing Editor /
Stellvertretende Redakteurin
Katrin Schiefer

Copy Editor (German) /
Lektorat (deutsch)
Dr. Evelyn Grünheid

Layout / Satz
Beatriz Feiler-Fuchs

E-mail: cpos@bib.bund.de

Scientifi c Advisory Board / 
Wissenschaftlicher Beirat
Paul Gans (Mannheim) 
Johannes Huinink (Bremen) 
Michaela Kreyenfeld (Rostock)
Marc Luy (Wien) 
Clara H. Mulder (Groningen) 
Notburga Ott (Bochum) 
Peter Preisendörfer (Mainz)
Zsolt Spéder (Budapest)

Comparative Population Studies

www.comparativepopulationstudies.de

ISSN: 1869-8980 (Print) – 1869-8999 (Internet)

Board of Reviewers / Gutachterbeirat
Martin Abraham (Erlangen)
Laura Bernardi (Lausanne)
Hansjörg Bucher (Bonn)
Claudia Diehl (Konstanz)
Andreas Diekmann (Zürich)
Gabriele Doblhammer-Reiter (Rostock)
Jürgen Dorbritz (Wiesbaden)
Anette Eva Fasang (Berlin)
E.-Jürgen Flöthmann (Bielefeld)
Alexia Fürnkranz-Prskawetz (Wien)
Beat Fux (Salzburg)
Joshua Goldstein (Berkeley)
Karsten Hank (Köln)
Sonja Haug (Regensburg)
Hill Kulu (Liverpool)
Aart C. Liefbroer (Den Haag)
Kurt Lüscher (Konstanz)
Emma Lundholm (Umeå)
Nadja Milewski (Rostock)
Dimiter Philipov (Wien)
Roland Rau (Rostock)
Tomáš Sobotka (Wien)
Jeroen Spijker (Barcelona)
Olivier Thévenon (Paris)
Helga de Valk (Brussel)
Heike Trappe (Rostock)
Michael Wagner (Köln)

© Federal Institute for Population Research 2015 – All rights reserved