Previous studies on divorce have found consistent empirical evidence that a negative association exists between marital disruption and children's wellbeing, life chances and intergenerational relations. However, there are relatively few studies on the long-term effect of divorce on parent–child contacts. Furthermore, contemporary research is confronted with two declining effect hypotheses. Firstly, it has been argued that the negative effects of divorce may possibly decrease as children's age at time of divorce increases; secondly, it has been suggested that the consequences of divorce might be less severe the more frequent marriage dissolution is in a society. By using data from Sweden, Denmark, Belgium and France from the Survey of Health, Ageing and Retirement in Europe, we analyze the existing relation between parents’ divorce and the intensity of parent–child contacts. Furthermore, we provide some empirical evidence on the two declining effect hypotheses. Our findings show that divorce has long-term negative effects on the intensity of intergenerational relations. The results do not provide any evidence in favor of either the individual or the collective declining effect hypothesis.

Family relations play an important role in determining individuals’ welfare and in the process of individuals’ social inclusion; this is even more so for the most vulnerable groups of the population, such as the economically poor, low educated, physically impaired, or elderly people. In particular, regarding the latter group, previous research has shown that the social networks of elderly individuals tend to be extremely weak and age-segregated, and that parent–child contacts represent a key factor in preventing social isolation. In this context, some recent socio-demographic changes in European societies – namely: population ageing, increasing childlessness rate, divorce rate and geographical mobility – raise some concerns about the social inclusion of the future elderly population.

Here we are particularly interested in studying how the divorce of the parents affects parent–child relations. Previous studies on divorce have found consistent empirical evidence that marital break-up has a negative effect on parent–child contacts in the years immediately subsequent to parental divorce. However, there is relatively little research carried out on the long-term effects of marriage dissolution. Furthermore, contemporary research on the topic is confronted with two declining effect hypotheses. Firstly, it has been suggested that the negative consequences of divorce on parent–child contacts decline as the age of the child at the time of divorce increases. Secondly, some authors advanced the hypothesis that the consequences of marital instability in a specific society tend to decrease when divorces become more frequent.

The present paper, by concentrating on the elderly population, aims at analyzing the long-term effects of parental divorce on parent–child contacts in four European societies, characterized by very different divorce histories and rates. Moreover, we will provide some evidence on the two above mentioned declining effect hypotheses. In particular, we will try to see if (i) the negative effects of divorce vary according to the age of child at the time of parents’ divorce (individual declining effect hypothesis); and if (ii) the more common divorce is in a society, the lesser its consequences (collective declining effect hypothesis).

The rates of separation and divorce have markedly increased in Western countries since the mid-1960s, albeit at a different pace and with varying intensity in different countries. An increasing number of children, therefore, have experienced their parents’ divorce while growing up. This latter phenomenon has attracted scholars’ attention. In particular, a significant number of studies have focused on the effects of parental divorce on children's wellbeing and life chances, and on intergenerational relations (for a review of these studies see for example Amato and Keith 1991a,b; Amato 2001; Sigle-Rushton and McLanahan 2004).

The academic discussion about the intergenerational effects of divorce has often been very controversial and the ‘conventional wisdom’ on the subject has changed several times (McLanahan and Sandefur 1994). Nowadays, after almost three decades of research, it is well established that there is a negative association between parental divorce and children's wellbeing, life chances and intergenerational relations. This general finding also applies to the specific outcome considered in the present article, i.e., parent–child contacts. There are several social mechanisms that can explain this negative relation.

The first mechanism is based on the selection hypothesis which argues that parental conflict prior to divorce, and not marital break-up per se, is what explains why children from intact families have more contacts with their parents than children from divorced families. In other words, those parents who have frequent conflict are those who are more at risk of separating and, at the same time, they are also those who have less contact with their children. Secondly, the negative effect of parental divorce on parent–child contacts might be produced by a change in social norms about family relations that are followed by family members. Thus, for example, in a study carried out in the USA, Cooney (1994) found that children's feelings about a given parent were strongly correlated with contacts with that parent in divorced, but not in intact, families. Next, Cooney advances the hypothesis that parental divorce is associated with weakened family obligations, and that family relationships may become more voluntary after divorce. A recent study based on data from Italy adds some complexity to this argument: Albertini and Saraceno (2008) found that parents’ divorce negatively affected their contacts with children, but it had no significant effect on the likelihood of receiving support from children. Thus, it would seem that parental divorce has a negative effect on social norms about family contacts but not on those about family support exchange – which are mainly driven by the logic of need and less connected to the quality of the relation. A third class of explanations has to do with the living arrangements and geographical mobility after divorce. In the years immediately subsequent to parental separation children often experience a decrease, or even a loss, of contact with the non-custodial parent (usually the father). In the case of fathers, this separation further strengths the asymmetry between mother–child and father–child intimacy already fostered by the gender division of labor in the family, and the primacy of the mother as caregiver and emotional centre of the family (Tomassini et al. 2004; Kalmijn 2007). However, contacts between custodial parents and children might also be reduced. In fact, custodial parents (usually mothers) often have to work long hours after the divorce and suffer from task overload (Amato 1993). This diminution of contacts between parents and children at a younger age also affects the frequency of contacts between them when the latter becomes an adult.

Recently two important questions about the intergenerational effects of parental divorce have arisen in the scientific literature. One is about the causality of this association: whether the effects of divorce on children's wellbeing and intergenerational relations are due to this event, per se, or to some other family/parent characteristics which are closely associated both with the risk of divorce and children's wellbeing (Painter and Levine 2000; Ní Bhrocháin 2001; Ginther and Pollack 2004). The second question has to do with the extent of the negative effects of divorce. It has been argued that these effects may possibly decrease as children's age at time of separation increases and, also, if marriage dissolutions are more frequent in a society. As mentioned above, in the present article we would like to provide some evidence on the second of these questions.

3.1. Individual declining effect hypothesis

In the literature on the intergenerational effects of divorce there have been a number of studies which have suggested the existence of some variation of these effects with regard to the age of child at the time in which the parents separated. However, the notion that the negative effects decline as children's age increases remains a contentious issue in the literature (Woodward et al. 2000). One group of studies suggests that marital dissolution may be more harmful for younger children than for older ones (Emery 1988; Allison and Furstenberg 1989). Others researchers have instead suggested that adolescent children are those who are particularly vulnerable to marriage break-ups (Wallerstein et al. 1988; Fergusson et al. 1994; Chase-Lansdale et al. 1995). Finally, a third group of studies found that the effects of divorce do not vary by children's age at the time of the event (Furstenberg and Teitler 1994).

One major problem in dealing with this topic is that it is very likely that the relation between the negative effects of divorce and children's age at the event differs markedly depending on the specific type of outcome we are taking into consideration (Hetherington et al. 1989). To the best of our knowledge, among the studies considering the effect of divorce on parent–child contacts, there is only one that, by using longitudinal data, examines how this effect varies according to the age of children at time of divorce, i.e., Booth and Amato (1994). In particular, the authors find some evidence to show that the younger the child is when divorce occurs the more disruptive the divorce is to father–child relations, whereas an insignificant effect is found for mother–child relations.

3.2. Collective declining effect hypothesis

The second declining effect hypothesis dealt with here is that the negative consequences of parental divorce decline when divorces become more frequent in a society. There are several theoretical arguments on which this hypothesis is based, and we briefly review here the most important. Firstly, it has been argued that the diffusion of marital break-up decreases the social stigma associated to it and, thus, decreases its negative effects (Wolfinger 1999; Sigle-Rushton et al. 2005). Secondly, selection processes into divorce have been called into question. When divorces are rare, only extremely dissatisfied and conflicting couples separate, whereas when divorce becomes a more widespread phenomenon moderately dissatisfied and conflicting couples will also separate (de Graaf and Kalmijn 2006). Thirdly, a declining effect of divorce might also be due to the increasing role of social policies and mediation programs. As a matter of fact, it is generally assumed that societies with higher divorce rates are more likely to develop mediation programs and generous family policies towards single parent families (Engelhardt et al. 2002). Fourthly, as suggested by Dronkers and Härkönen (2008), it may be that individuals’ own parents are not the only people who teach the children about marital and family behavior. According to this view, there is an aggregate effect of parental divorce that affects not only children from divorced families, but also children from intact families, and therefore this reduces the differences between them.

A number of studies have tried testing the collective declining hypothesis using different methodological approaches, and they have reached quite different conclusions. A first strategy is that of comparing studies from different decades while divorce rates have been on the increase. Adopting this approach Amato and Keith (1991a) found that the extent of divorce effects decreased in the US between the 1950s and 1980s. However, in an update of this meta-analysis Amato (2001) shows that the trend has reversed in studies carried out in the following decade. In contrast, by analyzing European longitudinal research, Wagner and Weiß (2006) found a negative association between the frequency of divorce and its intergenerational transmission. A second strategy adopted to test the collective declining hypothesis is that of comparing the effect of parental divorce across generations who experienced different divorce rates during their lives. Thus, Ely et al. (2000) comparing three British cohorts found that divorce effects had not decreased. Similar conclusions are reached by Sigle-Rushton et al. (2005) for the UK, and Biblarz and Raftery (1999) for the US. On the other hand, despite adopting the same approach, the results of Wolfinger (1999) point in the opposite direction. Eventually, a third strategy which has been adopted is that of comparing countries or geographical areas characterized by very different divorce histories and rates. By using this approach Ely et al. (1999) found no support for the declining effect hypothesis, and a similar conclusion is reached by Daatland (2007) and Tommasini et al. (2004). However, also in this line, we have studies which, despite adopting the same analytic strategy, find some support for a declining effect (e.g., Engelhardt et al. 2002; Kalmijn 2008). We can easily argue, therefore, that the empirical evidence on the collective declining hypothesis is mixed and further analysis is needed.

In the next part of this article we provide some additional evidence on the topic by adopting the latter of these approaches. We compare the long-term effects of divorce on parent–child contacts in Sweden, Denmark, Belgium and France. The justification for selecting specifically these countries is that they are characterized by very different divorce rates and histories, levels of social protection provided to divorced parents, and social acceptance of divorce. In that respect they are perfect candidates for applying the third of the above mentioned strategies. A further justification for choosing these countries is that in the data base we use there is a sufficient number of divorced parents as to reliably carry out our analyses, whereas, despite being quite interesting cases, we have had to exclude Mediterranean countries from our study; in fact they did not have sufficient cases of divorced individuals to perform reliable analyzes.

4.1. Trends in crude divorce rates

Since the 1960s, Nordic countries have had one of the highest divorce rates among Western societies.1 Thus, while in 1960 the crude divorce rate in Denmark and Sweden was equal respectively to 1.5 and 1.2, in France and Belgium much lower levels were registered, i.e., 0.7 and 0.5. Additionally, despite the huge increases registered during the following decades, in the 1980s differences remained substantial: 2.7 and 2.4 in the two Nordic countries, against 1.5 in France and Belgium. After this period, however, the former countries experienced a plateau in their divorce trends – this is also due to the significant increase in the number of couples that prefer cohabitation rather than marriage. On the other hand, crude divorce rates have been markedly on the increase both in France and Belgium. In 2004 marriage dissolutions in these two countries were as frequent as they were in Sweden and Denmark. It is worth noting, however, that in the former societies cohabitations are still much less frequent (Nazio 2008; Popenoe 2008).

4.2. Divorce legislation

All the four selected countries introduced divorce laws comparatively early and started liberalizing them – i.e., making it easier to divorce – from the 1960s onwards (Glendon 1996). Nevertheless, differences between their legislations on marriage dissolution were, and still are, quite significant.

The Nordic family law is based on the principles of the Lutheran reform which has traditionally been more permissive on family issues than the Catholic Church. In Denmark divorce was introduced in 1582 while the Swedish Lutheran State Church only introduced it in 1734 (Hussain and Kangas 2009). In the ancient régime the majority of family laws fell within the jurisdiction of the Catholic Church, and it was not until the 1792 revolution that the opportunity to divorce was introduced in France. Divorce was also allowed under the Napoleonic Code (1804) – although there were significant differences with respect to the previous norms; the law was abolished in 1816 and then reintroduced in 1884 (Loi Naquet), and it remained unchanged until the 1974 reform (Dutoit et al. 2000). Belgian divorce law originated from the French legal system, and the norms were those of the 1804 Napoleonic Code. The Belgian legislation of divorce remained unchanged until the 1974 reform (Senaeve 2001). Since the 1960s all of these countries have adopted much more liberal divorce laws; however marked differences remain.

In 1973 Sweden substantially modified the legislation on divorce, the main changes being: (i) all fault grounds were eliminated; (ii) unilateral divorce was made a matter of unqualified legal right; (iii) no reasons for divorce need to be given; (iv) all provisions of prior law requiring efforts to be made to bring the spouses to a mediator were eliminated; and (v) divorce was made possible without any waiting period unless one spouse opposes or has custody of children under 16, in which cases a 6-month period of consideration must be observed (Glendon 1996).

In 1969 the Danish Contraction and Dissolution of Marriage Act of 1922 was amended (Lund-Andersen and Krabbe 2003). Divorce was liberalized: the bases for separation were less restrictive; it was no longer necessary to give a specific reason for not being able to continue in the marriage; and unilateral separation was possible, when the court considers the marital relationship to be irretrievable. It is worth noting, however, that the liberalization in Denmark was less far reaching than the one implemented in Sweden. In fact, a period of separation before divorce is still required in certain circumstances, and both fault and non fault divorces are still possible (Verschraegen 2004). The Danish Marriage Act was modified in 1989, further extending the liberalization of the divorce process.

In France, until the reform of 1974, divorce was based on fault. Therefore, spouses who wished for a divorce had to allege faulty conduct of one partner, even though they both agreed upon termination of their marriage. The 1974 reform had the purpose of liberalizing divorce but it remained linked to the notion of fault (Glendon 1996; Verschraegen 2004). Three new grounds of divorce were introduced. First, divorce was made possible by mutual consent, with two possible variations: divorce on joint application by the spouses and divorce applied for by one and accepted by the other. The second ground for divorce was irretrievable breakdown of marriage. In this case, a spouse can present a unilateral petition of divorce without allegation of fault. However, to access this possibility 6 years of de facto separation was required and, moreover, a petition for divorce on this ground may be dismissed if it is established that the divorce would entail material or moral consequences of exceptional hardship for the unwilling spouse or for their children (Glendon 1987). The third ground for divorce was based on fault. This type of divorce could be only granted if one of the spouses has seriously or repeatedly violated the marital duties or s/he has been sentenced to a criminal punishment. In 2005 a new reform of divorce was approved and the notion of fault has been suppressed.

In Belgium, until 1974, there was the possibility of divorcing on the ground of fault or by mutual consent, although the latter possibility was strictly regulated. After the reform divorce could be obtained after 10 years of separation, even against the will of the ‘non-faulty spouse’. The spouse of a mentally ill person could also divorce, after 10 years of separation. The following reform (introduced in 1982) reduced both terms to 5 years (Pintens and Torfs 2003). From the nineties successive reforms have radically changed the divorce procedure, although the grounds for divorce remained unchanged. On the other hand, the divorce procedure by consent has been fundamentally simplified and the law introduced in 2000 has reduced the duration of separation to 2 years.

The database we use in our empirical analyzes is the Survey of Health, Ageing and Retirement in Europe (SHARE).2 SHARE is a longitudinal, multidisciplinary and cross-national survey representing the European population aged 50 and older. Additionally, the partners of selected individuals, independently of their age, were also interviewed. The first wave of SHARE took place in 2004 and 2005 with twelve participating countries.

SHARE contains detailed information on the social, economic and health situation of elderly Europeans, including information on the last 12 months of contact with each child – in particular, personal, phone and mail contacts are taken into consideration. Furthermore, one of the major advantages of SHARE is that information is provided on respondents’ children. Some general information – such as age, gender, or residential proximity to parents – is available for each child. Additional information – such as frequency of contact with parents, employment and marital status, number and age of own children – is available for the four children who live closest to the parental home. We have created parent–child dyads for each of the four children for which additional information is available (conditioned by whether the child is a natural child and not a step, fostered or adopted one). By choosing dyads as units of analysis we are able to consider the specific characteristics of each child as well as the parent–child relationship at the same time.3

It is worth noting that, despite the fact that we sometimes use the expression ‘parent–child relation’, what we are focusing on in the present paper is a variable providing information on parent–child contacts – and not, for example, on resource exchanges. The variable distinguishes the following categories, according to living arrangements and frequency of contact: (i) parent and child living in the same household; and among non co-residing parent–child dyads: (ii) daily; (iii) several times a week; (iv) one to four times a month; (v) less than once a month; and (vi) never.4 Our main independent variable is the parent's marital status. Unfortunately SHARE only provides information on the present marital status of interviewed individuals; additional information is available on whether they have a non registered partner at the time of interview. Following this, we were able to identify the following categories: (i) married/registered partnership; (ii) divorced/separated but has a partner; (iii) widowed but has a partner; (iv) divorced/separated; (v) never married; and (vi) widowed. In our analyses we excluded all of those dyads in which the parent is divorced or widowed but actually has a co-residing partner.5

In the next section, after presenting a few descriptive analyses, we use multinomial logistic regression to analyze the long-term effect of divorce on parent–child contact. In particular we implemented two main regression models. The first one (model 1) includes the following controlling variables: (i) age, gender, education, self perceived health status, gross equivalent income quintile, and country of residence of the parent; (ii) if the parent has other children beside the one considered in the dyad; and (iii) age and gender of child. In fact, previous literature on parent–child relations has showed that these factors are connected with frequency of contact. Thus, for example, mothers tend to have more frequent contact with children than fathers, and the same applies to daughters vs. sons; also, it has been found that education and income are positively correlated with frequency of contact. The second model (model 2) adds to the previous one the following variables: education, presence and age of children, marital status, and labor force status of the child. These variables have also been found to be correlated with parent–child relations. The reason for implementing two models is to better disentangle the possible mediating effects that the latter variables might have. Indeed, as shown in previous literature, child's education, labor market position and family life is strongly influenced by parents’ marital status. Therefore, the introduction of these controlling variables can make the negative effect of parents’ divorce on parent–child contacts partially ‘hidden’. Thus, it is useful to see to what extent our results are confirmed by implementing both the first and the complete regression model.

As for our analytic strategy, we proceeded in the following way: firstly, we implemented the above mentioned models with the main independent variable, only distinguishing between the different marital statuses of the parents; thus, analyzing the association between parent's divorce and frequency of contact with the child. Secondly, we wanted to see to what extent the divorce effect varies according to the age of child at time of divorce, and so we implemented the same models but, among those who are divorced, we distinguished different groups according to the age of child at time of divorce. Finally we approached the task of analyzing the extent to which there is a significant variation in the effect of divorce according to the social context in which it takes place. In particular, to test the collective declining hypothesis, we implemented two different analytic strategies. Firstly, we implemented models that also include an interaction term between parent's marital status and country dummies; secondly, we implemented the above mentioned regression models independently on each country data, and then compared the coefficients we obtained.6

Our final sample of parent–child dyads in the four selected countries is made up of 21,654 cases: 5,756 from Sweden, 2,981 from Denmark, 6,979 from Belgium and 5,938 from France.7 The average age of parents is equal to 64.6 years, whereas that of children is 36.1. Parent–child dyads in which the parent is divorced represent about 6.6 percent of our sample. Recently divorced parents – i.e., those whose divorce took place less than 6 years before the interview – are slightly less than 20 percent, whereas those who divorced more than 20 years ago are about 32 percent. The average time that has passed since the parent's divorce took place is 17 years. In the largest majority of cases, therefore, we are observing long-term effects of divorce. In regard to the age of child at divorce one fourth of divorced parent–child dyads include a child whose age at divorce was lower than 11 years; in more than 40 percent of the cases the child was between 11 and 20 years old, and older than 20 in the remaining cases. As might be expected the largest proportion of divorced parent–child dyads is to be found in the two Nordic countries (see Table 1).

TABLE 1. 
Distribution of parent–child dyads according to parent's actual marital status by country
SwedenDenmarkBelgiumFrance
Married 75 74 80 78 
Separated/Divorced 11 10 
Widowed 15 16 15 16 
 100 100 100 100 
SwedenDenmarkBelgiumFrance
Married 75 74 80 78 
Separated/Divorced 11 10 
Widowed 15 16 15 16 
 100 100 100 100 

Source: SHARE; own calculation.

Descriptive statistics reveal that norms about family contacts vary considerably between the two Nordic countries and France and Belgium (see Table 2). In the two latter countries the proportion of children who still co-reside or have daily contact with their parents tends to be much higher than in Sweden and Denmark. However, irrespectively of the specific cultural traditions of individual countries, in all of the four societies the frequency of contact of divorced parent–child dyads is markedly lower than that of married parent–child dyads. Thus, for example, the proportion of dyads in which the intergenerational relation is absent or almost broken (i.e., with no or very rare contact) is much higher among divorced parent–child dyads than among married ones.

TABLE 2. 
Distribution of parent–child dyads according to frequency of contact by parent's actual marital status and country
SwedenDenmarkBelgiumFrance
MarriedDivorcedAllMarriedDivorcedAllMarriedDivorcedAllMarriedDivorcedAll
Same household 12 10 11 15 12 14 19 12 17 
Daily 18 14 19 18 15 18 24 16 23 16 11 17 
Several times a week 39 28 37 36 28 34 30 27 30 27 26 27 
One to four times a month 28 44 32 33 43 35 27 33 27 32 40 33 
Less than once a month 
Never 
 100 100 100 100 100 100 100 100 100 100 100 100 
SwedenDenmarkBelgiumFrance
MarriedDivorcedAllMarriedDivorcedAllMarriedDivorcedAllMarriedDivorcedAll
Same household 12 10 11 15 12 14 19 12 17 
Daily 18 14 19 18 15 18 24 16 23 16 11 17 
Several times a week 39 28 37 36 28 34 30 27 30 27 26 27 
One to four times a month 28 44 32 33 43 35 27 33 27 32 40 33 
Less than once a month 
Never 
 100 100 100 100 100 100 100 100 100 100 100 100 

Source: SHARE; own calculation.

Of course this latter result might be driven by a number of different factors. Also, that it is simply a compositional effect cannot be rejected at this point; for instance, we could hypothesize that divorced parents are less educated and thus, since less educated people tend to have weaker intergenerational relations, it is educational level and not marital status that explains the results shown in Table 2. Therefore, in order to shed some light on the relation between parent marital status and contacts with children, we need to apply multivariate analysis techniques. Furthermore, by implementing different regression models we will also test the two declining effects hypotheses discussed above.

First of all, we want to analyze the relation between parent's marital status and parent–child contacts once we have controlled for other relevant parent's and child's characteristics. The models shown in the upper panel of Table 3 clearly show that the divorced parent–child dyads are much less likely to have frequent contact (i.e., more than once a week) and more likely to have rare or no contact than married parent–child ones. Therefore, adding to the existing literature – which mainly concentrated on the short-term effects of marital instability – our results show that divorce has long-term effects. Secondly, we want to provide some evidence on the variation of the divorce effect according to the age of child at time of divorce. Thus, we examine the relation between parent–child contacts and parent's marital status also distinguishing, among the divorced, between divorced parent–child dyads for which the age of child at divorce was different. The results of this analysis are reported in the lower panel of Table 3. In general these results do not provide evidence in favor of the hypothesis that the negative effect of divorce decreases as the age of the child increases, nor do they support the alternative hypothesis that divorce effects are stronger for teenagers. Indeed, despite the fact that in some cases differences between the coefficients exist, their confidence intervals largely overlap. Furthermore, even if we only consider the size of the coefficients, no meaningful gradient emerges with respect to the above mentioned hypotheses. Our results, therefore, partly contradict those of Booth and Amato (1994). As a matter of fact, we found no evidence that divorce is more disruptive of parent–child relations when it occurs when children are young.

TABLE 3. 
Effect of parent's marital status (upper panel) and age of child at divorce (lower panel) on parent-child contact, Relative Risk Ratios from multinomial logistic regressions
Model 1Model 2
Same householdDailySeveral times a weekLess than once a monthNeverSame householdDailySeveral times a weekLess than once a monthNever
Married Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. 
Widowed 4.07 [2.74–6.06] 1.09 [0.86–1.40] 1.10 [0.89–1.36] 1.10 [0.71–1.72] 2.42 [1.38–4.23] 3.18 [2.15–4.71] 1.01 [0.79–1.29] 1.06 [0.86–1.31] 1.07 [0.68–1.68] 2.24 [1.30–3.88] 
Divorced 0.26 [0.17–0.40] 0.45 [0.33–0.60] 0.65 [0.52–0.81] 1.43 [0.94–2.16] 3.02 [1.58–5.76] 0.17 [0.10–0.27] 0.41 [0.30–0.55] 0.63 [0.50–0.78] 1.40 [0.92–2.14] 2.75 [1.50–5.05] 
           
Pseudo r2(N) 0.15 (21654)     0.20 (21654)     
           
Divorced- child 0–10 0.19 [0.09–0.41] 0.49 [0.28–0.84] 0.63 [0.41–0.96] 1.68 [0.81–3.52] 3.19 [1.28–7.95] 0.12 [0.05–0.28] 0.42 [0.24–0.71] 0.60 [0.39–0.91] 1.70 [0.82–3.51] 2.99 [1.19–7.51] 
Divorced- child 11–20 0.29 [0.17–0.49] 0.38 [0.24–0.61] 0.66 [0.48–0.91] 1.36 [0.71–2.61] 2.64 [1.09–6.41] 0.18 [0.10–0.33] 0.36 [0.22–0.57] 0.64 [0.46–0.88] 1.32 [0.68–2.57] 2.18 [0.87–5.49] 
Divorced- child >20 0.26 [0.09–0.78] 0.45 [0.29–0.71] 0.63 [0.45–0.88] 1.63 [0.89–2.97] 2.80 [1.27–6.18] 0.23 [0.06–0.79] 0.43 [0.27–0.68] 0.61 [0.44–0.86] 1.61 [0.87–2.96] 2.70 [1.22–5.97] 
           
Pseudo r2(N) 0.15 (21654)     0.20 (21654)     
Model 1Model 2
Same householdDailySeveral times a weekLess than once a monthNeverSame householdDailySeveral times a weekLess than once a monthNever
Married Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. 
Widowed 4.07 [2.74–6.06] 1.09 [0.86–1.40] 1.10 [0.89–1.36] 1.10 [0.71–1.72] 2.42 [1.38–4.23] 3.18 [2.15–4.71] 1.01 [0.79–1.29] 1.06 [0.86–1.31] 1.07 [0.68–1.68] 2.24 [1.30–3.88] 
Divorced 0.26 [0.17–0.40] 0.45 [0.33–0.60] 0.65 [0.52–0.81] 1.43 [0.94–2.16] 3.02 [1.58–5.76] 0.17 [0.10–0.27] 0.41 [0.30–0.55] 0.63 [0.50–0.78] 1.40 [0.92–2.14] 2.75 [1.50–5.05] 
           
Pseudo r2(N) 0.15 (21654)     0.20 (21654)     
           
Divorced- child 0–10 0.19 [0.09–0.41] 0.49 [0.28–0.84] 0.63 [0.41–0.96] 1.68 [0.81–3.52] 3.19 [1.28–7.95] 0.12 [0.05–0.28] 0.42 [0.24–0.71] 0.60 [0.39–0.91] 1.70 [0.82–3.51] 2.99 [1.19–7.51] 
Divorced- child 11–20 0.29 [0.17–0.49] 0.38 [0.24–0.61] 0.66 [0.48–0.91] 1.36 [0.71–2.61] 2.64 [1.09–6.41] 0.18 [0.10–0.33] 0.36 [0.22–0.57] 0.64 [0.46–0.88] 1.32 [0.68–2.57] 2.18 [0.87–5.49] 
Divorced- child >20 0.26 [0.09–0.78] 0.45 [0.29–0.71] 0.63 [0.45–0.88] 1.63 [0.89–2.97] 2.80 [1.27–6.18] 0.23 [0.06–0.79] 0.43 [0.27–0.68] 0.61 [0.44–0.86] 1.61 [0.87–2.96] 2.70 [1.22–5.97] 
           
Pseudo r2(N) 0.15 (21654)     0.20 (21654)     

Notes: The reference outcome is ‘one to four times a month’.

95% C.I. reported in squared parenthesis, coefficients for controlling variables are omitted

Source: SHARE; own calculation.

Finally, the last goal of our regression analyses is that of testing the between countries differences in the negative effect of divorce. As argued above, this comparison provides an indirect test for the hypothesis that the effect of divorce diminishes the more that it becomes a frequent phenomenon in a specific society. If this was the case we should find that the negative consequences that marital instability has on parent–child relations are smaller in countries with high divorce rates – i.e., Sweden and Denmark – than in countries with low divorce rates – i.e., France and Belgium. In a first step of these analyses, we simply added to the two above implemented regression models an interaction between country of residence and parent's marital status. The results of these models show that the negative effect of divorce on parent–child contacts is not significantly different in the four countries considered – the only exception being that the negative effect on the probability of having several contacts in a week is slightly smaller in France than in the other three countries. These results are also confirmed once we run our regression models separately on the data for the four countries (Table 4). In fact, despite (sometimes relevant) existing differences in the divorce effect coefficients, the confidence intervals for these effects largely overlap. The only exception being for the fact that, as in previous analyses, the divorce effect on the outcome ‘several times a week’ is significantly smaller in France than in Sweden, and that on the outcome ‘never’ is significantly higher in Denmark than in Sweden and Belgium – but this only holds for the first regression model. Furthermore, generally speaking, if we take a look at the size of the coefficients the only pattern we are able to detect is some tendency for the negative effect of divorce to be smaller in France than in the Nordic countries. In sum from our regression analyses we conclude that no support is found for either the individual or the collective declining hypothesis.

TABLE 4. 
Effect of parent's divorce status (vs. married) on parent–child contacts, Relative Risk Ratios from multinomial logistic regressions
SwedenDenmarkBelgiumFrance
Model 1Model 2Model 1Model 2Model 1Model 2Model 1Model 2
Same household 0.17 [0.10–0.28] 0.14 [0.07–0.25] 0.22 [0.10–0.51] 0.22 [0.08–0.56] 0.25 [0.16–0.40] 0.13 [0.07–0.25] 0.28 [0.18–0.43] 0.16 [0.09–0.31] 
Daily 0.36 [0.27–0.50] 0.34 [0.23–0.50] 0.51 [0.34–0.78] 0.46 [0.27–0.79] 0.41 [0.29–0.57] 0.38 [0.25–0.58] 0.46 [0.32–0.66] 0.41 [0.26–0.66] 
Several times a week 0.40 [0.32–0.52] 0.39 [0.29–0.52] 0.63 [0.46–0.86] 0.60 [0.40–0.92] 0.57 [0.43–0.76] 0.58 [0.41–0.81] 0.75 [0.58–0.98] 0.72 [0.52–0.99] 
One to four times a month Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. 
         
Less than once a month 1.89 [1.15–3.10] 1.84 [1.01–3.35] 3.02 [1.43–6.38] 3.88 [1.38–10.9] 3.05 [1.84–5.03] 3.16 [1.75–5.70] 1.29 [0.82–2.02] 1.25 [0.72–2.16] 
Never 1.42 [0.62–3.29] 1.67 [0.44–6.25] 8.65 [3.76–19.9] 8.84 [2.87–27.23] 1.85 [1.05–3.25] 1.68 [0.76–3.71] 3.48 [1.82–6.62] 3.08 [1.31–7.23] 
         
Pseudo r2(N) 0.18 (5756) 0.20 (5756) 0.16 (2981) 0.18 (2981) 0.12 (6979) 0.19 (6979) 0.16 (5938) 0.21 (5938) 
SwedenDenmarkBelgiumFrance
Model 1Model 2Model 1Model 2Model 1Model 2Model 1Model 2
Same household 0.17 [0.10–0.28] 0.14 [0.07–0.25] 0.22 [0.10–0.51] 0.22 [0.08–0.56] 0.25 [0.16–0.40] 0.13 [0.07–0.25] 0.28 [0.18–0.43] 0.16 [0.09–0.31] 
Daily 0.36 [0.27–0.50] 0.34 [0.23–0.50] 0.51 [0.34–0.78] 0.46 [0.27–0.79] 0.41 [0.29–0.57] 0.38 [0.25–0.58] 0.46 [0.32–0.66] 0.41 [0.26–0.66] 
Several times a week 0.40 [0.32–0.52] 0.39 [0.29–0.52] 0.63 [0.46–0.86] 0.60 [0.40–0.92] 0.57 [0.43–0.76] 0.58 [0.41–0.81] 0.75 [0.58–0.98] 0.72 [0.52–0.99] 
One to four times a month Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. 
         
Less than once a month 1.89 [1.15–3.10] 1.84 [1.01–3.35] 3.02 [1.43–6.38] 3.88 [1.38–10.9] 3.05 [1.84–5.03] 3.16 [1.75–5.70] 1.29 [0.82–2.02] 1.25 [0.72–2.16] 
Never 1.42 [0.62–3.29] 1.67 [0.44–6.25] 8.65 [3.76–19.9] 8.84 [2.87–27.23] 1.85 [1.05–3.25] 1.68 [0.76–3.71] 3.48 [1.82–6.62] 3.08 [1.31–7.23] 
         
Pseudo r2(N) 0.18 (5756) 0.20 (5756) 0.16 (2981) 0.18 (2981) 0.12 (6979) 0.19 (6979) 0.16 (5938) 0.21 (5938) 

95% c.i. reported in squared parenthesis, coefficients for controlling variables are omitted.

Source: Weighted results.

After several decades of research, sociologists have provided abundant empirical evidence of the existing negative association between parents’ divorce and children's wellbeing, life chances and intergenerational relations. However, there are relatively few studies on the long-term effects of divorce. Thus, it remains an open question if divorce has long lasting consequences on parent–child relations, or just short-term ones. Furthermore, at present the scientific community is confronted by two relevant questions about this relation: (i) is there a causality in this association?; and (ii) how do intergenerational effects of divorce vary according to the age of child at divorce and in different societal contexts? In the present paper, by concentrating on parent–child contacts, we aim, firstly, at providing some additional evidence on the negative association between divorce and intergenerational relations. Secondly, we also analyze to what extent the strength of the negative relation between parental divorce and parent–child contacts varies according to the age of child at time of divorce, and to the country of residence.

The results from multinomial logistic regression models provide evidence of an existing negative association between parental divorce and intergenerational relations still many years after the event. In particular, it is shown that divorced parent–child dyads are less likely to have daily or weekly contact than married parent–child dyads. Furthermore, the proportion of dyads for which the relation is absent or almost broken is significantly higher among the former group.

Further analyses are implemented to test for the two declining effect hypotheses. As for the age of child at time of divorce we found no evidence supporting the idea that younger children suffer the most severe consequences of marriage dissolution. Neither did we find evidence that, as suggested by other scholars, adolescent children are more vulnerable than others to this family event. Next, by comparing the relation between parental divorce and parent–child contacts in four countries with very different divorce histories and rates, we tested the hypothesis that the increasing number of divorces in a society might contribute to a decrease in its negative effects. Similarly to what has been found in recent analyses (Tomassini et al. 2004; Daatland 2007) we found no support for the collective declining effect hypothesis.

It can be concluded, therefore, that while our analyses document a negative relation between parental divorce and the intensity of intergenerational relations many years after the event, they do not provide any evidence in favor of a declining effect. However, it is worth noting that the limitations derived from the type of data available impose some limitation on the interpretation and generalization of our results. First of all, it should be said that we are not addressing here the fundamental problem of causality which is behind any study on the consequences of divorce (Ní Bhrolcháin 2001). Secondly, even assuming a causal nexus, we cannot disentangle the effects of pre-divorce family life, divorce proceedings themselves, and events that follow divorce. What we are comparing here, in fact, are simply parent–child contacts between currently divorced and currently married parents.

Despite these qualifications, there are several policy implications of our analyses. Firstly, and most importantly, if divorced parents experience weaker relations with their children, even many years after the event, then we should be more concerned with the social inclusion of the future elderly population. As a matter of fact, as shown by previous research, the social networks of elderly people tend to be quite weak and strongly age-segregated. Parent–child contacts often represent a key factor in preventing the social isolation of the elderly. In a context of an ageing population and increasing childlessness and divorce rates, the weakening of intergenerational relations associated with parental divorce might represent an additional factor that negatively affects levels of social inclusion in European societies. Secondly, our data do not support the idea that we should expect stronger negative effects of divorce for younger children, nor for adolescents. According to these findings, therefore, there is no specific reason for supporting programs to address a specific group of the divorced children population. Finally, the substantial similarity of the negative effects of divorce across societies characterized by very different histories, divorce rates and social acceptance of divorce confirms that these consequences will not decrease as divorce becomes more common.

1.

The source of the following data on crude divorce rates is Eurostat (n.d.).

2.

This paper uses data from release 2.0.1 of SHARE 2004; see www.share-project.org for a full list of funding institutions, and for an introduction to the SHARE data set and its methodological aspects.

3.

Standard errors of all statistics reported below have been corrected for the clustering of dyads within the same household.

4.

Despite the fact that the variable is clearly ordinal, in our multivariate analysis we preferred to treat it as multinomial. As a matter of fact, not only do we want to maintain the full complexity of this variable, but also a Wald test by Brandt preformed on an ordinal regression model shows that the largest part of our variables violates the parallel regression assumption.

5.

Despite the comparison of the divorce effects for re-partnered and non re-partnered parents is of great interest, we were forced to delete re-partnered divorced parents from our sample due the very low number of respondents who share this characteristic.

6.

It is worth noting that due to lack of space we only report coefficients for the variable marital status, all controlling variables coefficients are omitted. Moreover, models that consider interactions terms between parent's marital status and country are not shown. The full results of the analyses can be obtained from the corresponding author.

7.

The corresponding number of respondents is 9,659: 2,584 from Sweden, 1,351 from Denmark, 3,177 from Belgium, 2,547 from France.

Albertini
,
M.
, and
Saraceno
,
C.
,
2008
. “‘Intergenerational contact and support: The long term effects of marital instability in Italy’”. In:
Saraceno
,
C.
, ed.
Families, Ageing and Social Policy: Generational Solidarity in European Welfare States.
Cheltenham
:
Edward Elgar
;
2008
. pp.
194
216
.
Allison
,
P. D.
, and
Furstenberg
,
F. F.
,
1989
. ‘
How marital dissolution affects children: Variations by age and sex
’,
Developmental Psychology
25
(
1989
), pp.
355
70
.
Amato
,
P. R.
,
1993
. ‘
Children's adjustment to divorce: Theories, hypotheses, and empirical support
’,
Journal of Marriage and the Family
55
(
1993
), pp.
23
38
.
Amato
,
P. R.
,
2001
. ‘
Children of divorce in the 1990s: An update of the Amato and Keith (1991) meta-analysis
’,
Journal of Family Psychology
15
(
2001
), pp.
355
70
.
Amato
,
P. R.
, and
Keith
,
B.
,
1991a
. ‘
Parental divorce and adult well-being: A meta-analysis
’,
Journal of Marriage and the Family
53
(
1991a
), pp.
43
58
.
Amato
,
P. R.
, and
Keith
,
B.
,
1991b
. ‘
Parental divorce and well-being of children: A meta-analysis
’,
Psychological Bulletin
110
(
1991b
), pp.
26
46
.
Biblarz
,
T. J.
, and
Raftery
,
A. E.
,
1999
. ‘
Family structure, educational attainment, and socioeconomic success: Rethinking the “Pathology of Matriarchy”
’,
American Journal of Sociology
105
(
1999
), pp.
321
65
.
Booth
,
A.
, and
Amato
,
P. R.
,
1994
. ‘
Parental marital quality, parental divorce, and relations with parents
’,
Journal of Marriage and the Family
56
(
1994
), pp.
21
34
.
Cooney
,
T. M.
,
1994
. ‘
Young adults’ relations with parents: The influence of recent parental divorce
’,
Journal of Marriage and the Family
56
(
1994
), pp.
45
56
.
Chase-Lansdale
,
L. P.
,
Cherlin
,
A. J.
, and
Kiernan
,
K. E.
,
1995
. ‘
The long-term effects of parental divorce on the mental health of young adults: A developmental perspective
’,
Child Development
66
(
1995
), pp.
1614
34
.
Daatland
,
S. O.
,
2007
. ‘
Marital history and intergenerational solidarity: The impact of divorce and unmarried cohabitation
’,
Journal of Social Issues
63
(
2007
), pp.
809
25
.
Dronkers
,
J.
, and
Härkönen
,
J.
,
2008
. ‘
The intergenerational transmission of divorce in cross-national perspective: Results from the Fertility and Family Surveys
’,
Population Studies
62
(
2008
), pp.
273
88
.
Dutoit
,
B.
,
Arn
,
R.
,
Sfondylia
,
B.
, and
Taminelli
,
C.
,
2000
.
Le divorce en droit comparé. Volume 1: Europe.
Geneva
:
Librarie Droz
;
2000
.
Ely
,
M.
,
Richards
,
M. P. M.
,
Wadsworth
,
M. E. J.
, and
Elliot
,
J. B.
,
1999
. ‘
Secular changes in the association of parental divorce and children's educational attainment – evidences from three British cohorts
’,
Journal of Social Policy
28
(
1999
), pp.
437
55
.
Ely
,
M.
,
West
,
P.
,
Sweeting
,
H.
, and
Richards
,
M. P. M.
,
2000
. ‘
Teenage family life, life chances, lifestyles and health: A comparison of two contemporary cohorts
’,
International Journal of Law Policy and the Family
14
(
2000
), pp.
1
30
.
Emery
,
R. E.
,
1988
.
Marriage, Divorce and Children's Adjustment.
Newbury Park, CA
:
Sage
;
1988
.
Engelhardt
,
H.
,
Trappe
,
H.
, and
Dronkers
,
J.
,
2002
. ‘
Differences in family policy and the intergenerational transmission of divorce: A comparison between the former East and West Germany
’,
Demographic Research
6
(
2002
), pp.
296
324
.
Eurostat cronos database, Theme 3: Population and Sicial Conditions
, (n.d.), http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database.
Fergusson
,
D. M.
,
Lynskey
,
M. T.
, and
Horwood
,
J. L.
,
1994
. ‘
The effects of parental separation, the timing of separation and gender on children's performance on cognitive tests
’,
Journal of Child Psychology and Psychiatry
35
(
1994
), pp.
1077
92
.
Furstenberg
,
F. F.
, and
Teitler
,
J. O.
,
1994
. ‘
Reconsidering the effects of marital disruption: What happens to children of divorce in early adulthood?
’,
Journal of Family Issues
15
(
1994
), pp.
173
90
.
Ginther
,
D. K.
, and
Pollak
,
R. A.
,
2004
. ‘
Family structure and children's educational outcomes: Blended families, stylized facts, and descriptive regressions
’,
Demography
41
(
2004
), pp.
671
96
.
Glendon
,
M. A.
,
1987
.
Abortion and Divorce in Western Law: American Failures, European Challenges.
Cambridge, MA
:
Harvard University Press
;
1987
.
Glendon
,
M. A.
,
1996
.
The Transformation of Family Law: State, Law and Family in United States and Western Europe.
Chicago, IL
:
Chicago University Press
;
1996
.
de Graaf
,
P. M.
, and
Kalmijn
,
M.
,
2006
. ‘
Divorce motives in a period of rising divorce: Evidence from a Dutch life-history survey
’,
Journal of Family Issues
27
(
2006
), pp.
483
505
.
Hetherington
,
M. E.
,
Stanley-Hagan
,
M.
, and
Anderson
,
E.
,
1989
. ‘
Marital transitions. A child's perspective
’,
American Psychologist
44
(
1989
), pp.
303
12
.
Hussain
,
A. M.
, and
Kangas
,
O.
,
2009
. “‘New holes in the safety net? Economic and social consequences of divorce in Denmark’”. In:
Andreß
,
H.-J.
, and
Hummelsheim
,
D.
, eds.
When Marriage Ends. Economic and Social Consequences of Partnership Dissolution.
Cheltenham
:
Edward Elgar
;
2009
. pp.
143
79
.
Kalmijn
,
M.
,
2007
. ‘
Gender differences in the effects of divorce, widowhood and remarriage on intergenerational support: Does marriage protect fathers?
’,
Social Forces
85
(
2007
), pp.
1079
104
.
Kalmijn
,
M.
,
2008
. “‘The effects of separation and divorce on parent–child relationships in ten European countries’”. In:
Saraceno
,
C.
, ed.
Families, Ageing and Social Policy: Generational Solidarity in European Welfare States.
Cheltenham
:
Edward Elgar
;
2008
. pp.
170
93
.
Lund-Andersen
,
I.
, and
Krabbe
,
L.
,
2003
. “‘Brief history of main developments of Danish divorce law’”. In:
Boele-Woelki
,
K.
,
Braant
,
B.
, and
Summer
,
I.
, eds.
European Family Law in Action, Vol. 1: Grounds of Divorce.
Antwerp
:
Intersedia
;
2003
. pp.
26
7
.
McLanahan
,
S.
, and
Sandefur
,
G.
,
1994
.
Growing up with a Single Parent: What Hurts, What Helps?.
Cambridge, MA
:
Harvard University Press
;
1994
.
Nazio
,
T.
,
2008
.
Cohabitation, Family and Society: European Experiences.
Oxford
:
Routledge
;
2008
.
Ní Bhrocháin
,
M.
,
2001
. ‘
Divorce effects and causality in the social sciences
’,
European Sociological Review
17
(
2001
), pp.
33
57
.
Painter
,
G.
, and
Levine
,
D. I.
,
2000
. ‘
Family structure and youths’ outcomes: Which correlations are causal?
’,
The Journal of Human Resources
35
(
2000
), pp.
524
49
.
Pintens
,
W.
, and
Torfs
,
L.
,
2003
. “‘Brief history of main developments of Belgian divorce law’”. In:
Boele-Woelki
,
K.
,
Braant
,
B.
, and
Summer
,
I.
, eds.
European Family Law in Action, Vol. 1: Grounds of Divorce.
Antwerp
:
Intersedia
;
2003
. pp.
13
14
.
Popenoe
,
D.
,
2008
.
Cohabitation, Marriage and Child Wellbeing. A Cross-National Perspective.
Piscataway, NJ
:
Rutgers, The State University of New Jersey – The National Marriage Project
;
2008
.
Senaeve
,
P.
,
2001
. “‘Family law’”. In:
Bocken
,
H.
, and
de Bondt
,
W.
, eds.
Introduction to Belgian Law.
Rijnland
:
Kluwer Law International
;
2001
. pp.
128
222
.
Sigle-Rushton
,
W.
, and
McLanahan
,
S.
,
2004
. “‘Father absence and child well-being: A critical review’”. In:
Moynihan
,
D. P.
,
Smeeding
,
T. A.
, and
Rainwater
,
L.
, eds.
The Future of the Family.
New York
:
Russel Sage Foundation
;
2004
. pp.
116
59
.
Sigle-Rushton
,
W.
,
Hobcraft
,
J.
, and
Kiernan
,
K. E.
,
2005
. ‘
Parental divorce and subsequent disadvantage: A cross-cohort comparison
’,
Demography
42
(
2005
), pp.
427
46
.
Tomassini
,
C.
,
Kalogirou
,
S.
,
Grundy
,
E.
,
Fokkema
,
T.
,
Martikainen
,
P.
,
van Groenou
,
M. I. B.
, and
Karisto
,
A.
,
2004
. ‘
Contacts between elderly parents and their children in four European countries: Current patterns and future prospects
’,
European Journal of Ageing
1
(
2004
), pp.
54
63
.
Verschraegen
,
B.
,
2004
. “‘Divorce’”. In:
Glendon
,
M. A.
, ed.
International Encyclopedia of Comparative Law, Vol. IV, Persons and Family.
Tübigen
:
Martinus Nijhoff Publishers
;
2004
. pp.
1
91
.
Wagner
,
M.
, and
Weiz
,
B.
,
2006
. ‘
On the variation of divorce risks in Europe: Findings from a meta-analysis of European longitudinal studies
’,
European Sociological Review
22
(
2006
), pp.
483
500
.
Wallerstein
,
J. S.
,
Corbin
,
S. B.
, and
Lewis
,
J. M.
,
1988
. “‘Children of divorce: A ten year study’”. In:
Hetherington
,
M. E.
, and
Arasteh
,
J. D.
, eds.
Impact of Divorce, Single-Parenting and Stepparenting on Children.
Hillsdale, NJ
:
Erlbaum
;
1988
. pp.
198
214
.
Wolfinger
,
N. H.
,
1999
. ‘
Trends in intergenerational transmission of divorce
’,
Demography
36
(
1999
), pp.
415
20
.
Woodward
,
L.
,
Fergusson
,
D. M.
, and
Belsky
,
J.
,
2000
. ‘
Timing of parental separation and attachment to parents in adolescence: Results of a prospective study from birth to age 16
’,
Journal of Marriage and the Family
62
(
2000
), pp.
162
74
.

Marco Albertini is assistant professor of Sociology at the University of Bologna. His current research interests include the consequences of divorce, the intergenerational exchange of social and economic support, and the distribution of household income. He has published on these issues in the European Sociological Review, Journal of European Social Policy, European Societies and Ageing & Society.

Anna Garriga will receive a PhD in Political and Social Science at University Pormpeu Fabra, in Barcelona. She is currently a Research Associate at the Department of Social Policy and Social Work at the University of York. She studies the effect of parental divorce on children's well-being. She is particularly interested in the comparison of the effect of parental divorce in different countries and generations and in the possible explanations of this effect, at a micro- and macro-level. Her main research interests focus also on the effects of marital quality on quality of parenting and children's well-being. She is also working on predictors of marital quality and family dissolution, and on the policy issues related to family changes.

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the use is non-commercial and the original work is properly cited. For a full description of the license, please visit https://creativecommons.org/licenses/by-nc/4.0/legalcode.