Is the sense of obligation we feel towards our parents comparable to the one we feel towards our children? Most studies of normative solidarity measure only filial norms, that is, norms for children's obligations towards parents, whilst largely ignoring parental norms, that is, norms for parents’ obligations towards children. This article quantitatively investigates parental and filial norms in 20 countries within 5 European regions. The article examines the question of whether the family cultures of North-West Europe can be understood as cultures of descending familialism, that is, cultures in which parental obligations are emphasised over filial obligations, as opposed to cultures of extreme individualism. The article contributes to the literature on family norms theoretically by showing that family cultures should be differentiated not merely by their strength but also by their direction, and methodologically by highlighting the importance of developing precise measures of both parental and filial norms. For the Nordic countries in particular, the analysis shows that the family culture is pluralistic, with the question of intergenerational responsibilities being one likely to provoke discussion in these societies for years to come.

How much self-sacrifice do we owe to our parents and our children? All over Europe, ageing populations and a crisis-ridden economy challenge families and governments to meet the needs of care-dependent citizens, both young and old. In many countries, the established ways of caring for these groups are threatened, and reforms are necessary. The success or failure of such future reforms will greatly depend on the configuration of norms in the various countries, that is, to what degree people accept moral responsibility for family members, and whether they feel responsible for older and younger generations equally.

The concept of directionality in intergenerational exchange practices has been developed in several studies in Asia, Europe and the US (e.g. Kreager and Schröder-Butterfill 2008; Fingerman et al. 2012; Lin and Yi 2013), but few have explored the question of possible corresponding variations in the directionality of intergenerational norms. Likewise, while measuring the overall strength of family norms is standard in studies of intergenerational relations, (e.g. Cooney and Dykstra 2011), less attention has been paid to their directionality.

I argue that looking at the directionality of norms resolves the apparent contradiction between, on the one side, the supposedly weak family norms of the North and, on the other, those recent studies showing a complex picture of intergenerational exchange practices with both close-knit and more distant families existing in all European countries (Bordone 2009; Dykstra and Fokkema 2011) and some studies even indicating that relations in the ‘individualistic’ North are more often perceived as satisfactory than is the case in the ‘familistic’ South (Brandt and Deindl 2013).

One reason for the apparent discrepancy between norms and practices is that most studies of normative solidarity measure only filial norms, that is, norms for children's obligations towards parents, whilst largely ignoring parental norms, that is, norms for parents’ obligations towards children (e.g. Goerres and Tepe 2010).1 This one-sided attention to filial norms is problematic in two ways. Firstly, the exclusive focus on filial obligations implies that the parents are always the weaker party, that is, more or less passive objects for filial care or neglect. This premise is questionable, as European parents on average are net-givers to their adult children for most of their lives (Mudrazija 2016). Secondly, equating family norms with filial norms ignores a body of literature that suggest that the weighting of parental and filial norms may vary between cultures, that is, that cultures appearing ‘weak’ when measured by filial norms may appear ‘strong’ when measured by parental norms, or vice versa (e.g. Aboderin 2005).

This article quantitatively investigates parental and filial norms in 20 countries within five European regions, each comprising between three and five institutionally and culturally similar countries corresponding to recognised welfare regime typologies. The article examines the question of whether the family cultures of North-West Europe can be understood as cultures of descending familialism, that is, cultures in which parental obligations are emphasised over filial obligations, as opposed to cultures of individualism. The analysis uses data from the European Values Study (EVS) 2008 wave, which includes measures of parental and filial norms for these 20 countries.

The article contributes to the literature on family norms theoretically by showing that family cultures should be differentiated not merely by their strength but also by their direction, and methodologically by highlighting the importance of developing precise measures of both parental and filial norms. For the Nordic countries in particular, the analysis portrays a pluralistic family culture, where the question of intergenerational responsibilities is likely to be provoke debate for years to come.

The article is organised as follows: I review the literature on differences in European family cultures, followed by an overview of the scholarly interpretations of these differences. Then I present the data, methods and analytic strategy, after which I present the results of the analysis, the conclusion and discuss the academic and policy-oriented implications of the results.

In studies of European regional variations in normative solidarity, the Nordic countries rank consistently the lowest and the Mediterranean countries the highest (e.g. Lowenstein and Daatland 2006; Kalmijn and Saraceno 2008). The changes in family patterns known as the second demographic transition came early in the Nordic countries and gained widespread acceptance both formally and informally (Sobotka 2008; Lesthaeghe 2014). The formal changes include the recognition of unmarried cohabitation and the institutionalisation of same-sex unions (Syltevik 2010, 2014; Fernández and Lutter 2013). The informal changes include a general culture of sexual permissiveness (Therborn 2004; Liefbroer and Billari 2010) and acceptance of homosexuality, abortion and single parenthood (Halman et al. 2011; Minkov et al. 2013; Masterson and Hoobler 2015).

That individualism is strong within the Nordic societies is evident (Esping-Andersen 1990; Inglehart and Baker 2000; Therborn 2007; Halman et al. 2011). Empirical rankings of countries, from the most collectivist to the most individualist, generally place the Nordic countries at the extreme individualist end of the spectrum, closely followed by the Netherlands and the UK (Lin 2005; Inglehart 2008; Liefbroer and Billari 2010). Following a long tradition of legal individualism, the Nordic welfare regimes have been termed ‘de-familialising’ in that they regard the individual, not the family, as their basic unit (Esping-Andersen 1999; Craveiro et al. 2013). The Nordic and Mediterranean regions are the most extreme cases in Europe of, respectively, generational independence and co-residence (Albertini and Kohli 2012; Isengard 2013).

Scholars have raised the question of whether the strong individualism found in the Nordic countries is detrimental to family solidarity. The ‘crowding out’-hypothesis suggests that strong family norms are superfluous in generous welfare regimes (Yamagishi and Yamagishi 1994; Ganjour and Widmer 2016), spurring concerns that decreasing economic dependencies between generations weaken family ties to the detriment of care-dependent groups (Wolfe 1989; Popenoe 2009; Billari and Liefbroer 2010).

Other studies, however, complicate the straightforward categorisation of Southern ‘strong’ versus Northern ‘weak’ family cultures (e.g. Jappens and Van Bavel 2012). As regards institutions, Leitner (2003) demonstrates that familialism and de-familialisation are not opposite poles as much as separate policy dimensions. She identifies three ideal-typical welfare regimes in Europe – Explicit, Implicit and Optional. The Explicit Familialism-regimes of Austria and Germany support family carers through paid leave and pension recognition of care periods, but do not provide public alternatives to family care. The Implicit Familialism-regimes of Italy and Spain neither support family carers nor provide public alternatives, leaving the family to care by default. The Optional Familialism-regimes of Denmark and Sweden offer de-familialising services such as public care facilities alongside familialistic policy measures such as paid leave for carers (Leitner 2003; Craveiro et al. 2013).

As regards family practices, 80% of parents in the North and 90% of parents in the South have weekly contact with an adult child (Ogg and Renaut 2006; Bordone 2009, 2012). In the North, intergenerational exchanges are more frequent, but less intense than in the South (Albertini et al. 2007; Brandt 2013). ‘Familistic’ family types make up more than half of late-life families in all 11 countries studied by Dykstra and Fokkema (2011), and their ‘autonomous’ type, in which relatives have little contact and weak norms of obligation, is more common in France and Switzerland than in the ‘individualistic’ countries of Scandinavia and The Netherlands (Dykstra and Fokkema 2011). The ‘familism’ of the Mediterranean region is tempered by a widespread desire for state support (Ganjour and Widmer 2016); and a tentative shift towards situational norms is occurring even in Spain, generally considered a stronghold of familism (Caïs and Folguera 2013). In sum, both close-knit and more distant families exist in all countries.

While studies of intergenerational exchange practices recognise as a matter of course that resources may flow both from parents to children and vice versa, most studies of family norms assume that filial norms are an adequate indicator of the strength of family norms in general. Bengtson's model of family solidarity defines normative solidarity as ‘expectations regarding filial obligations and parental obligations’ (Bengtson 2001: 8, my emphasis). Nevertheless, most studies either include only measures of filial norms, ignoring parental norms (e.g. Bengtson and Roberts 1991; Kalmijn and Saraceno 2008; Calzada and Brooks 2013) or they combine both types into a single measure of family norms without taking directionality into account (Alesina and Giuliano 2010; Ganjour and Widmer 2016).

The implicit assumption that parental norms are always strong ignores the hypothesis that normative solidarity may take different forms in different cultural settings. In other words, some cultures may perceive the parent–child relationship as asymmetrical and thus be ‘strong’ on parental norms and ‘weak’ on filial norms, or vice versa. Several qualitative studies indicate that this may be the case. Aboderin finds an illustrative case in Ghana, where people ‘repay’ obligations to parents by giving to their own children (Aboderin 2005), and Finch and Mason find that parent–child commitments are strongest down the generations (Finch and Mason 1993: 30–1). Dykstra and Fokkema's ‘descending’ family type is more common compared to the ‘ascending’ type in the North, whilst the opposite is true in the South (Fokkema et al. 2012). Moreover, in one of the few quantitative studies including measures of both sets of norms, Daatland et al. find that parental norms are relatively stronger in France, Germany and Norway, and filial norms relatively stronger in Bulgaria, Georgia and Russia (Daatland et al. 2011, 2012). Thus, to conceptually grasp the diversity of normative positions, scholars must expand the dichotomy of ‘individualistic’ and ‘familistic’ cultures to include categories of ‘ascending familialism’ (where filial obligations are stressed over parental obligations) and ‘descending familialism’ (where the reverse is true).

I hypothesise that the family cultures of North-West Europe can be more accurately understood as cultures of descending familialism than as cultures of extreme individualism. I explore this hypothesis by analysing parental and filial norms in 20 countries within 5 European regions.

I use data from the EVS 2008 wave.2 The EVS has been conducted since 1981, with subsequent data collection waves in 1990, 1999 and 2008. The number of countries covered has increased from 16 in 1981 to 47 in 2008. Data are collected with face-to-face interviews with standardised questionnaires in representative samples of the populations over 18 in the participating countries. Full documentation including country questionnaires, errata and patches is available from the data repository at ZACAT – GESIS Online Study Catalogue. I weighted data for gender and age using the standardised weights constructed by the EVS (EVS 2011a, b).3

3.1. The dependent variable: intergenerational obligations

Three dichotomous items in the EVS questionnaire measure intergenerational norms (Table 1).

Table 1.
Items.
ConceptWording
Children's emotional obligations towards parents absolute or situational Which of these two statements do you tend to agree with?
A: Regardless of what the qualities and faults of one's parents are, one must always love and respect them.
B: One does not have the duty to respect and love parents who have not earned it by their behaviour and attitudes. 
Children's practical obligations towards parents absolute or situational Which of the following statements best describes your views about responsibilities of adult children towards their parents when their parents are in need of long-term care?
A: Adult children have the duty to provide long-term care for their parents even at the expense of their own well-being.
B: Adult children have a life of their own and should not be asked to sacrifice their own well-being for the sake of their parents. 
Parents’ obligations towards children absolute or situational Which of the following statements best describes your views about parents’ responsibilities to their children?
A: Parents’ duty is to do their best for their children even at the expense of their own well-being.
B: Parents have a life of their own and should not be asked to sacrifice their own well-being for the sake of their children. 
ConceptWording
Children's emotional obligations towards parents absolute or situational Which of these two statements do you tend to agree with?
A: Regardless of what the qualities and faults of one's parents are, one must always love and respect them.
B: One does not have the duty to respect and love parents who have not earned it by their behaviour and attitudes. 
Children's practical obligations towards parents absolute or situational Which of the following statements best describes your views about responsibilities of adult children towards their parents when their parents are in need of long-term care?
A: Adult children have the duty to provide long-term care for their parents even at the expense of their own well-being.
B: Adult children have a life of their own and should not be asked to sacrifice their own well-being for the sake of their parents. 
Parents’ obligations towards children absolute or situational Which of the following statements best describes your views about parents’ responsibilities to their children?
A: Parents’ duty is to do their best for their children even at the expense of their own well-being.
B: Parents have a life of their own and should not be asked to sacrifice their own well-being for the sake of their children. 

For each item, answer ‘A’ represents an absolute position and answer ‘B’ represents a situational position. The items are directional in that two of them measure filial norms (practical and emotional obligations, respectively), whilst the third measures parental norms. Using these items I have constructed a dependent variable with four discrete values, defining a typology of four exhaustive and mutually exclusive normative positions (Lazarsfeld 1937; Kluge 2000; Szelewa and Polakowski 2008). For simplicity I have merged both types of filial obligations, leaving four ideal-typical normative positions (Table 2).

Table 2.
Normative positions.
Filial obligations absoluteFilial obligations situational
Parental obligations absolute 1. Symmetrical familialism
Parental obligations absolute
Some or all filial obligations absolute 
2. Descending familialism
Parental obligations absolute
Filial practical and emotional obligations situational 
Parental obligations situational 3. Ascending familialism
Parental obligations situational
Some or all filial obligations absolute 
4. Situationalism
Parental obligations situational
Filial practical and emotional obligations situational 
Filial obligations absoluteFilial obligations situational
Parental obligations absolute 1. Symmetrical familialism
Parental obligations absolute
Some or all filial obligations absolute 
2. Descending familialism
Parental obligations absolute
Filial practical and emotional obligations situational 
Parental obligations situational 3. Ascending familialism
Parental obligations situational
Some or all filial obligations absolute 
4. Situationalism
Parental obligations situational
Filial practical and emotional obligations situational 

The resulting typology distinguishes between symmetrical normative positions (symmetrical familialism and situationalism) and asymmetrical normative positions (ascending and descending familialism).

As Table 3 shows, symmetrical familialism is the most common position in the sample, accounting for almost two-thirds (64.7%) of the responses, with the remaining third divided between descending familialism (14.8%), ascending familialism (11.5%) and situationalism (9.1%).

Table 3.
Shares in each normative position.
Normative positionNPer cent
Symmetrical familialism 15,413 64.7 
Descending familialism 3514 14.8 
Ascending familialism 2729 11.5 
Situationalism 2161 9.1 
Total 23,817 100.0 
Normative positionNPer cent
Symmetrical familialism 15,413 64.7 
Descending familialism 3514 14.8 
Ascending familialism 2729 11.5 
Situationalism 2161 9.1 
Total 23,817 100.0 

3.2. Independent variable: region

I have constructed the regions according to Leitner's classification of optional, explicit and implicit familialism-regimes (Leitner 2003). Great Britain, the Netherlands and Switzerland are notoriously difficult to classify within welfare regime typologies. Therefore, following Manow (2004), I have placed these in their own region of mixed welfare regimes. I include four post-communist countries to compare the normative configuration in this region with the more well-known familialism-regimes in the West. Table 4 shows the resulting regions:

Table 4.
Regions.
TypeCountriesNPer cent
Nordic (optional familialism) Denmark, Finland, Norway and Sweden. 4918 16.4 
Central (explicit familialism) Austria, Belgium, France, Germany and Luxembourg. 8205 27.4 
Mediterranean (implicit familialism) Greece, Italy, Portugal and Spain. 6072 20.3 
Eastern (post-communist) Czech Republic, Hungary, Poland and the Slovak Republic. 6353 21.2 
Western (mixed) The Netherlands, Switzerland and Great Britain. 4387 14.7 
TypeCountriesNPer cent
Nordic (optional familialism) Denmark, Finland, Norway and Sweden. 4918 16.4 
Central (explicit familialism) Austria, Belgium, France, Germany and Luxembourg. 8205 27.4 
Mediterranean (implicit familialism) Greece, Italy, Portugal and Spain. 6072 20.3 
Eastern (post-communist) Czech Republic, Hungary, Poland and the Slovak Republic. 6353 21.2 
Western (mixed) The Netherlands, Switzerland and Great Britain. 4387 14.7 

The post-communist societies of East and Central Europe have been included only to a limited extent in research about family patterns in Europe. Before 1990, these countries were examples of relatively de-familialising institutional regimes, with, for example, public child care more universally available than was the case in Western Europe. However, since the transition from communism, national policies have diversified (Kocourková 2002; Robila 2004; Szelewa and Polakowski 2008). Recent studies indicate that family cultures in East Europe are more similar to those of the Mediterranean region than to those of North-West Europe (e.g. Calzada and Brooks 2013; Borsenberger et al. 2016).

Table 5 shows the distribution of the normative positions by region. Symmetrical familialism is the most common position in all regions. In all but the Nordic region an absolute majority of respondents support this position. The ranking of the regions follows the familiar pattern, with the Mediterranean region having the largest share of respondents supporting symmetrical familialism, and the Nordic region the smallest. The Mediterranean region exhibits high normative consensus, with the symmetrical familialism position accounting for 84% of the responses. The consensus is slightly less pronounced in the Central and Eastern regions, where approximately two-thirds of the responses support symmetrical familialism.

Table 5.
Normative positions by regions.
Normative position
Symmetrical familialismDescending familialismAscending familialismSituationalismTotal
RegionNPct.NPct.NPct.NPct.NPct.
Nordic 1475 40.9 1353 37.5 285 7.9 498 13.8 3611 100 
Western 2017 56.0 936 26.0 330 9.2 318 8.8 3601 100 
Central 4151 62.7 792 12.0 943 14.3 731 11.1 6617 100 
Eastern 3396 70.9 218 4.6 737 15.4 436 9.1 4787 100 
Mediterranean 4374 84.1 215 4.1 434 8.3 178 3.4 5201 100 
Total 15,413 64.7 3514 14.8 2729 11.5 2161 9.1 23,817 100 
Normative position
Symmetrical familialismDescending familialismAscending familialismSituationalismTotal
RegionNPct.NPct.NPct.NPct.NPct.
Nordic 1475 40.9 1353 37.5 285 7.9 498 13.8 3611 100 
Western 2017 56.0 936 26.0 330 9.2 318 8.8 3601 100 
Central 4151 62.7 792 12.0 943 14.3 731 11.1 6617 100 
Eastern 3396 70.9 218 4.6 737 15.4 436 9.1 4787 100 
Mediterranean 4374 84.1 215 4.1 434 8.3 178 3.4 5201 100 
Total 15,413 64.7 3514 14.8 2729 11.5 2161 9.1 23,817 100 

Notes: χ2: 3546.424; df: 12; p = .000. Bold type is used to emphasise values that are divergent from the marginal distribution.

In contrast, the Nordic and Western regions exhibit normative pluralism, with no more than about half of the respondents supporting any one position. However, those not supporting the majority position do not support a position of situationalism, either. In both the Nordic and Western regions the second-most common position is one of descending familialism, which accounts for 37.5% of all responses in the Nordic region and 26.0% in the Western region. Situationalism accounts for 13.8% in the Nordic region and 8.8% in the Western, and ascending familialism for 9.2% and 7.9% respectively.

In the Central, Eastern, and Mediterranean regions, ascending familialism is the second largest position. In the Mediterranean region, ascending familialism accounts for 8.3%, descending familialism for 4.1% and situationalism for 3.4% of the responses. The distributions in the Eastern and Central regions are very similar. In the Eastern region, 70% support symmetrical familialism, 15.4% ascending familialism, 9.1% situationalism and 4.6% descending familialism; in the Central region, 62.7% support symmetrical familialism, 14.3% ascending familialism, 12% descending familialism and 11.1% situationalism.4

4.1. Multivariate analysis

For the multivariate analysis I have constructed dummy variables for each normative position and performed a separate logistic regression for each outcome.5 Calculating robust standard errors compensates for possible heteroscedasticity (Mood 2010).

4.1.1. Control variables

As control variables, I include age, gender, education, parenthood, religious affiliation and religious practice, all of which are known to influence family norms. Table 6 shows the operationalisation and distribution of the control variables, as well as the expected direction of the influence of each.6 The first category of each variable is the reference category.7

Table 6.
Control variables.
VariableMeanSDMedianModeNHypotheses
Age 47.03 17.88 46.00 30 29,803 Older people more familistic (Liefbroer and Billari 2010; Daatland et al. 2011
Variable Categories N Per cent Hypotheses 
Gender 13,708 45.8 Women more situational (Liefbroer and Billari 2010; Daatland et al. 2011
16,215 54.2 
Education Lower 8970 30.3 The higher education, the more situational (Liefbroer and Billari 2010
Middle 13,732 46.3 
Higher 6945 23.4 
Parenthood No children 8653 29.6 Parents stronger on parental norms than non-parents (Aboderin 2005; Daatland et al. 2011
Has children 20,543 70.4 
Religious affiliation None 8846 30.0 Protestants, non-affiliates more situational (Inglehart 1990; Halman and Riis 2003
Catholic 11,900 40.3 
Protestant 5850 19.8 
Other 2920 9.8 
Religious participation Frequent (weekly or more often) 4807 16.2 Practitioners more absolute (Friedman et al. 2008; Gans et al. 2009; Liefbroer and Billari 2010; Sabatier et al. 2011; Aassve et al. 2013
Infrequent (once a month or less) 13,575 45.8 
Never/practically never 11,275 38.0 
VariableMeanSDMedianModeNHypotheses
Age 47.03 17.88 46.00 30 29,803 Older people more familistic (Liefbroer and Billari 2010; Daatland et al. 2011
Variable Categories N Per cent Hypotheses 
Gender 13,708 45.8 Women more situational (Liefbroer and Billari 2010; Daatland et al. 2011
16,215 54.2 
Education Lower 8970 30.3 The higher education, the more situational (Liefbroer and Billari 2010
Middle 13,732 46.3 
Higher 6945 23.4 
Parenthood No children 8653 29.6 Parents stronger on parental norms than non-parents (Aboderin 2005; Daatland et al. 2011
Has children 20,543 70.4 
Religious affiliation None 8846 30.0 Protestants, non-affiliates more situational (Inglehart 1990; Halman and Riis 2003
Catholic 11,900 40.3 
Protestant 5850 19.8 
Other 2920 9.8 
Religious participation Frequent (weekly or more often) 4807 16.2 Practitioners more absolute (Friedman et al. 2008; Gans et al. 2009; Liefbroer and Billari 2010; Sabatier et al. 2011; Aassve et al. 2013
Infrequent (once a month or less) 13,575 45.8 
Never/practically never 11,275 38.0 

Table 7 shows the results of the four regressions.

Table 7.
Regression parameters logistic.
Symmetrical familialismDescending familialismAscending familialismSituationalism
Region Nordic 
 Western .4292206 *** −.3562383 *** .1658585 −.3021457 ** 
 Central .6136357 *** −1.317219 *** .7702528 *** −.0803156 
 Eastern .9308287 *** −.2.2499681*** .9100746 *** .1383626 
 Mediterranean 1.46466 *** −1,968932 *** .1876905 −.974706 *** 
Rel. practice Frequent practitioner 
 Infrequent practitioner −.6233902 *** .5503627 *** .3857391 *** .7643944 *** 
 Non-practitioner −.8474942 *** .8984222 *** .2990215 ** .9275764 *** 
Education Lower (<complete secondary) 
 Middle (complete secondary) −.3716663 *** .316395 *** .0876337 .4436364 *** 
 Upper (some tertiary)3 −.4086572 *** .5017789 *** .0290434 .3637885 *** 
Parenthood No children 
 Has children .1293644 ** .3964698 *** −.3915782 *** −.3307561 *** 
Gender Male 
 Female −.1738512 *** .2252228 *** .0017091 .1302132 ** 
Rel denom. None 
 Catholic .3267097 *** −.32346 *** .0662618 −.3917068 *** 
 Protestant .0609079 .0171378 .0721109 .0637927 
 Other (Muslim, Orthodox) .3700246 *** −.7923088 *** .2057688 * −.3657233 ** 
Age  .0006331 .0006352 .001414 .0008868 
Constant  .6686497 −1.8419344 −2.6884746 −2.883199 
Symmetrical familialismDescending familialismAscending familialismSituationalism
Region Nordic 
 Western .4292206 *** −.3562383 *** .1658585 −.3021457 ** 
 Central .6136357 *** −1.317219 *** .7702528 *** −.0803156 
 Eastern .9308287 *** −.2.2499681*** .9100746 *** .1383626 
 Mediterranean 1.46466 *** −1,968932 *** .1876905 −.974706 *** 
Rel. practice Frequent practitioner 
 Infrequent practitioner −.6233902 *** .5503627 *** .3857391 *** .7643944 *** 
 Non-practitioner −.8474942 *** .8984222 *** .2990215 ** .9275764 *** 
Education Lower (<complete secondary) 
 Middle (complete secondary) −.3716663 *** .316395 *** .0876337 .4436364 *** 
 Upper (some tertiary)3 −.4086572 *** .5017789 *** .0290434 .3637885 *** 
Parenthood No children 
 Has children .1293644 ** .3964698 *** −.3915782 *** −.3307561 *** 
Gender Male 
 Female −.1738512 *** .2252228 *** .0017091 .1302132 ** 
Rel denom. None 
 Catholic .3267097 *** −.32346 *** .0662618 −.3917068 *** 
 Protestant .0609079 .0171378 .0721109 .0637927 
 Other (Muslim, Orthodox) .3700246 *** −.7923088 *** .2057688 * −.3657233 ** 
Age  .0006331 .0006352 .001414 .0008868 
Constant  .6686497 −1.8419344 −2.6884746 −2.883199 

Notes: Italic type indicates parameters that are not significantly different from 0. *p < .05; **p < .010; ***p < .000.

4.2. Regression results

4.2.1. Symmetrical familialism

The largest differences in the propensity to support the symmetrical familialism position are found between regions; since all logit differences are positive, living in any region but the Nordic increases the probability.

Of the control variables, religious participation is the most important, followed by education, religious affiliation, gender and parenthood. Age is not significant. Both infrequent religious practitioners and non-practitioners are significantly less likely than frequent practitioners to support symmetrical familialism. Compared to persons with the lowest level of education, those with the medium or highest level of education have less likelihood of supporting symmetrical familialism. Both Catholics and Others (predominantly members of Muslim or Orthodox communities) are more likely to support symmetrical familialism, while Protestants are not significantly different from those with no religious affiliation. Women are less likely than men to support symmetrical familialism, while parents are more likely than are non-parents to support symmetrical familialism.

4.2.2. Descending familialism

Again, region is the most important contributing variable. As expected, Nordics are significantly more likely than all others to support this position, whilst Easterners are the least likely. Mediterraneans and Centrals are close behind, whilst Westerners are somewhat closer to Nordics but still significantly less likely.

Of the control variables, religious practice is again the most important, while religious affiliation, education and parenthood are approximately equally important, and gender less important still. Age is not significant. Both infrequent practitioners and non-practitioners are more likely than frequent practitioners to support descending familialism. Catholics and Others are less likely to support descending familism than are Protestants and the unaffiliated. Those with more education and parents are more likely to support descending familialism than are non-parents and the less educated. Women are more likely than men to support descending familialism.

4.2.3. Ascending familialism

Centrals and Easterners are significantly more likely to support ascending familialism than are Nordics, Westerners and Mediterraneans. Parents are significantly less likely to support ascending familialism, compared to non-parents. Both infrequent practitioners and non-practitioners are more likely than frequent practitioners to support ascending familialism. Of religious affiliations, only the groups of Others are significantly more likely to support ascending familialism, but the effect is only just significant. Neither education, age nor gender are significant.

4.2.4. Situationalism

Westerners and Mediterraneans both have less likelihood of supporting situationalism than Nordics, with Mediterraneans the least likely. There is no difference between Centrals and Easterners and Nordics. Infrequent religious practitioners and non-practitioners are both more likely than frequent practitioners to support situationalism. Any degree of education increases the likelihood of supporting situationalism. Catholics and Others are less likely to support situationalism than are Protestants and the unaffiliated. Parents are significantly less likely to support situationalism than are non-parents, while women are slightly more likely to do so than are men. Age, again, is insignificant.

4.3. Intra-regional differences

The regression confirms the inter-regional differences and points to interesting intra-regional differences. After region, religion (religious practice and religious affiliation) is the most important variable. Frequent religious practitioners support symmetrical familialism much more wholeheartedly than do infrequent and non-practitioners, both of which groups are divided between all three alternative positions, with non-practitioners being slightly more strongly represented as supporters of descending familialism and situationalism. Catholics and Others consistently support more binding norms than do Protestants and the unaffiliated. Education decreases support for symmetrical familialism and increases support for descending familialism and situationalism. Having children means more support for descending familialism and less support in particular for ascending familialism and situationalism. Gender is of relatively little significance, with women being somewhat less likely to support symmetrical familialism and somewhat more likely to support descending familialism and situationalism than are men.

4.4. Predicted probabilities

Table 8 shows predicted probabilities for selected groups.8

Table 8.
Predicted probabilities selected groups logit.
Symmetrical (%)Descending (%)Ascending (%)Situationalism (%)
Practicing Catholic male without children, low education Mediterranean 91.91 1.53 7.58 1.47 
Eastern 86.95 1.16 14.46 3.32 
 Central 82.91 2.90 12.81 4.10 
 Western 80.13 7.23 7.43 2.83 
 Nordic 72.42 10.02 6.37 3.79 
Non-practicing Protestant mother, high education Mediterranean 67.74 13.75 7.69 6.53 
Eastern 55.18 10.75 14.64 13.88% 
Central 47.28 23.43 12.98 16.71 
 Western 42,72 44,44 7.54 12.04 
 Nordic 32.68 53.32 6.46 15.62 
Symmetrical (%)Descending (%)Ascending (%)Situationalism (%)
Practicing Catholic male without children, low education Mediterranean 91.91 1.53 7.58 1.47 
Eastern 86.95 1.16 14.46 3.32 
 Central 82.91 2.90 12.81 4.10 
 Western 80.13 7.23 7.43 2.83 
 Nordic 72.42 10.02 6.37 3.79 
Non-practicing Protestant mother, high education Mediterranean 67.74 13.75 7.69 6.53 
Eastern 55.18 10.75 14.64 13.88% 
Central 47.28 23.43 12.98 16.71 
 Western 42,72 44,44 7.54 12.04 
 Nordic 32.68 53.32 6.46 15.62 

Note: Bold type is used to highlight the most frequently occurring profiles for each group.

Practising Catholic fathers with relatively little education have a very high probability of supporting symmetrical familialism, whilst highly educated, non-practising Protestant mothers have the lowest. The predicted probabilities in Table 8 show the normative positions taken by typical members of these extreme opposites in the five regions. In the Mediterranean region, even highly educated, non-practising mothers have a 67.7% probability of supporting symmetrical familialism, whilst the less educated, practising fathers have 91.9% probability. Nordic, less-educated practitioners are almost as likely to support symmetrical familialism as are the highly educated non-practitioners in the Mediterranean region, whilst the highly educated Nordic non-practitioners are more likely to support descending familialism than symmetrical familialism. In the Eastern and to some degree in the Central region, ascending familialism is relatively stronger than elsewhere.

The inclusion of control variables leaves unchanged the initial regional differences, with the Nordic and Western regions exhibiting a marked normative pluralism and the Mediterranean, Eastern and Central regions much more normative consensus. However, for those in the pluralistic Nordic and Western regions who do not recognise symmetrical family obligations, the alternative is not situationalism but a position of descending familialism; one that views parental norms as absolute but filial norms as situational. In contrast, ascending familialism is particularly strong in the Central and Eastern regions.

Given these results, talking about one set of family norms in the North-West of Europe in general and in the Nordic countries in particular is neither meaningful nor useful, because these regions clearly exhibit normative pluralism; and even in this pluralistic normative climate, no support exists for the hypothesis that North-West Europe is heading towards the general dissolution of family norms: indeed, the most common normative position remains one of symmetrical familialism, a stance that holds the obligations of both parents and children to be absolute. In so far as symmetrical familialism has lost ground, descending familialism appears to be the alternative. Situationalism is the position of a small minority.

The importance of religious practice and religious affiliation underlines the salience of institutional, rather than individual, characteristics in the reinforcement of norms. Religious practice generally takes place in group settings, and religious groups, all other things being equal, emphasise normative obligations over individual preferences. The results for parenthood support the altruism hypothesis that parenthood makes people lower their expectations of children and increase their expectations of themselves as parents; the result can be interpreted either as a realistic adjustment of expectations or as an expression of true altruism, where the interest of others are put above one's own.

The marked pluralism found in the Nordic region is surprising, given the reputation of these small and relatively homogeneous countries for bring high on consensus. One possible interpretation might be that, whilst many aspects of the ‘new family’ of the second demographic transition are uncontroversial in the Nordic countries, the inhabitants of that region have not yet found their stance in the area of intergenerational obligations. The generations who pioneered the SDT are only now reaching the age of retirement and increasing dependence on their adult children. Following the development of the normative positions in future waves of data collection should prove enlightening.

5.1. Limitations of the study

One limitation – common to all large-scale, multiple-topic surveys – is that the number of items measuring intergenerational norms is necessarily limited. A comparison with the findings of Daatland et al. shows that the results are robust, even though the items used are slightly different: Daatland et al. find 57% of Norwegians acknowledging filial responsibility is 57%, while the share in the Nordic region in this study is 59%, indicating that another data set would probably not have yielded very different results (Daatland et al. 2011). Nevertheless, including more items in future surveys is desirable for more precise measurements of parental and filial obligations, as well as more precise measures for varying types of obligations (emotional, practical, etc.). Analysis of future waves of EVS data will show whether the trend towards more situational perceptions of family obligations is continuing, or whether the economic hardships of the intervening years have made families more inclined to take care of their own.

Intergenerational obligations engender normative turbulence in the usually consensual Nordic region, to a degree where notions of general Nordic family norms is no longer meaningful. Whilst a surprisingly large group in these countries still support the traditional normative ideal of mutual responsibilities, a large minority supports an ideal of descending obligations, in which adult children only help parents from choice and inclination, not from a sense of sacrifice, duty or obligations. As reforms forcing adult children to, say, take on heavy care burdens will lack legitimacy both amongst the adult children and their dependent parents, policy makers should review policies in areas such as housing, transportation and work–life balance to allow families to build those committed relationships that will enable the voluntary assumption of responsibilities. Especially in the coming decades, when large groups of migrants from strong reciprocity cultures will grow old in countries where descending familialism is strong, normative clashes are likely to occur between parents’ expectations and those of their adult children socialised in the norms of the new country.

The configuration of norms and institutions in the Nordic countries increases the salience of the moral identity of the ‘good parent’ whilst decreasing the salience of the corresponding identity of the ‘good child’. When people do not acknowledge absolute filial obligations, those frequent intergenerational interactions and exchanges that occur take on a new significance in that they affirm the emotional qualities of the relationship and confirm that the parents have done a good job of raising their children. In other words, when one's children do not have to visit one as a duty means that if they do it anyway, one has been a good parent: their act of visiting confirms one's moral identity. Investigating to what degree this logic is present in the minds of parents and children in the Nordic countries should prove fruitful indeed.

I thank Margaretha Järvinen, Peter Gundelach and Kristian B. Karlson as well as two anonymous referees for constructive comments and advice on earlier drafts of this article.

No potential conflict of interest was reported by the author.

Bella Marckmann is Assistant Professor in the Department of Sociology at the University of Copenhagen. Her research interests are in the areas of family sociology, intergenerational relations, everyday life, housing, social capital, survey research and mixed methods research. She is currently working on a project entitled ‘The Moral Economy of Families’ funded by the Danish Council for Independent Research (DFF).

1

Another reason, as Eekelaar (2012) points out, is that many writers fallaciously assume that the withdrawal of institutional dominance over the individual equals egoism and a weakened sense of moral obligation. This discussion, however, is beyond the scope of this article.

2

Unfortunately, no comparative survey has yet been conducted with both good measures of intergenerational norms and data from all or most European countries. The Generations and Gender Programme (GGP) includes a number of items measuring intergenerational norms, but this survey has not been conducted in as many countries as the EVS, and several of the regions compared in the analysis are represented by one country only.  Appendix 1 contains a detailed overview of data sets considered for the analysis.

3

The weights also compensate for disproportional sample sizes in regions of Germany and Belgium. I have not weighted by population size, that is, smaller countries contribute as much to the analysis as larger countries. As the countries in the analysis represent institutional and cultural entities, it would defeat the purpose of the analysis to weight by population, since this would mean that the larger countries would entirely drown out the smaller ones, thus obscuring the intra-regional differences and giving a too homogeneous picture of each region. This would obviously be especially problematic in regions comprising both large countries such as Germany with a population of more than 80 million and Luxembourg with one of less than 1 million.

4

The regions are not entirely distinct, as variation within each region makes for a certain overlap.  Appendix 2 shows the distribution of normative positions within each country. The most striking case is the Netherlands, which has a normative profile that more closely resembles those of the Nordic region than those of Switzerland and Great Britain.

5

Opinions differ as to the type of regression appropriate for binary dependent variables.  Appendix 4 shows the results obtained by linear regression for comparison. Linear regression has the advantage compared to logistic regression that the resulting regression coefficients are readily interpretable, being probability differences (Rothman 1986; Hellevik 2009).

6

For detailed discussion of the correlation between religious belief and family norms, see Inglehart 1990; Halman and Riis 2003; Gans et al. 2009; Sabatier et al. 2011; Aassve et al. 2013.

7

 Appendix 3 shows details of the categories of the education variable.

8

Regression parameters are included only in the calculation of predicted probabilities for those positions where they are significant.

Aassve
,
A.
,
Arpino
,
B.
and
Billari
,
F. C.
(
2013
) ‘
Age norms on leaving home: multilevel evidence from the European Social Survey
’,
Environment and Planning A
45
(
2
):
383
401
. doi:
Aboderin
,
I.
(
2005
) ‘
Conditionality’ and ‘Limits’ of Filial Obligation: Conceptual Levers for Developing a Better Understanding of the motivational Basis and Societal Shifts or Patterns in Old Age Family Support
, Oxford Institute of Ageing Working Papers, Editor: George W. Leeson, Oxford.
Albertini
,
M.
and
Kohli
,
M.
(
2012
) ‘
The generational contract in the family: an analysis of transfer regimes in Europe
’,
European Sociological Review
29
(
4
):
828
40
. doi:
Albertini
,
M.
,
Kohli
,
M.
and
Vogel
,
C.
(
2007
) ‘
Intergenerational transfers of time and money in European families: common patterns – different regimes?
’,
Journal of European Social Policy
17
(
4
):
319
34
. doi:
Alesina
,
A.
and
Giuliano
,
P.
(
2010
) ‘
The power of the family
’,
Journal of Economic Growth
15
(
2
):
93
125
. doi:
Bengtson
,
V.
(
2001
) ‘
Beyond the nuclear family: the increasing importance of multigenerational bonds. The burgess award lecture
*’,
Journal of Marriage and Family
63
(
1
):
1
16
. doi:
Bengtson
,
V. L.
and
Roberts
,
R. E. L.
(
1991
) ‘
Intergenerational solidarity in aging families: an example of formal theory construction
’,
Journal of Marriage & the Family
53
:
856
70
. doi:
Billari
,
F. C.
and
Liefbroer
,
A. C.
(
2010
) ‘
Towards a new pattern of transition to adulthood?
’,
Advances in Life Course Research
15
(
2–3
):
59
75
. doi:
Bordone
,
V.
(
2009
) ‘
Contact and proximity of older people to their adult children: a comparison between Italy and Sweden
’,
Population, Space and Place
15
(
4
):
359
80
. doi:
Bordone
,
V.
(
2012
) ‘Social norms and intergenerational relationships’, in
G.
De Santis
(ed),
The family, the market or the state?
,
Dordrecht
:
Springer Netherlands
: pp.
159
78
.
Borsenberger
,
M.
,
Fleury
,
C.
and
Dickes
,
P.
(
2016
) ‘
Welfare regimes and social cohesion regimes: do they express the same values?
’,
European Societies
18
(
3
):
221
244
. doi:
Brandt
,
M.
(
2013
) ‘
Intergenerational help and public assistance in Europe
’,
European Societies
15
(
1
):
26
56
. doi:
Brandt
,
M.
and
Deindl
,
C.
(
2013
) ‘
Intergenerational transfers to adult children in Europe: do social policies matter?
’,
Journal of Marriage and Family
75
(
1
):
235
51
. doi:
Caïs
,
J.
and
Folguera
,
L.
(
2013
) ‘
Redefining the dynamics of intergenerational family solidarity in Spain
’,
European Societies
15
(
4
):
557
76
. doi:
Calzada
,
I.
and
Brooks
,
C.
(
2013
) ‘
The myth of Mediterranean familism
’,
European Societies
15
(
4
):
514
34
. doi:
Cooney
,
T. M.
and
Dykstra
,
P. A.
(
2011
) ‘
Family obligations and support behaviour: a United States-Netherlands comparison
’,
Ageing and Society
31
:
1026
50
. doi:
Craveiro
,
D.
,
Matos
,
A. D.
,
Silva
,
S. G.
,
Martinez-Pecino
,
R.
and
Schouten
,
M. J.
(
2013
) ‘31 Intergenerational support: the role of gender and social networks’, in
B-S.
Axel
,
B.
Martina
,
L.
Howard
and
W.
Guglielmo
(eds),
Active Ageing and Solidarity between Generations in Europe: First Results from SHARE after the Economic Crisis
,
Berlin
:
De Gruyter
, pp.
359
68
.
Daatland
,
S. O.
,
Herlofson
,
K.
and
Lima
,
I. A.
(
2011
) ‘
Balancing generations: on the strength and character of family norms in the West and East of Europe
’,
Ageing and Society
31
:
1159
79
. doi:
Daatland
,
S. O.
,
Veenstra
,
M.
and
Herlofson
,
K.
(
2012
) ‘
Age and intergenerational attitudes in the family and the welfare state
’,
Advances in Life Course Research
, Elsevier Ltd
17
(
3
):
133
44
. doi:
Dykstra
,
P. A.
and
Fokkema
,
T.
(
2011
) ‘
Relationships between parents and their adult children: a West European typology of late-life families
’,
Ageing and Society
31
(
4
):
545
69
. doi:
Eekelaar
,
J.
(
2012
) ‘
Self-restraint: Social norms, individualism and the family
’,
Theoretical Inquiries in Law
13
(
1
):
75
95
. doi:
Esping-Andersen
,
G.
(
1990
)
The Three Worlds of Welfare Capitalism
,
Cambridge
:
Polity Press
.
Esping-Andersen
,
G.
(
1999
)
The Social Foundation of Post-Industrial Economies
,
Oxford
:
Oxford University Press
.
EVS
(
2011a
)
European Values Study 2008: Integrated Dataset (EVS 2008). ZA4800 Data File Version 3.0.0
,
Cologne
:
GESIS Data Archive
.
EVS
(
2011b
)
EVS 1981 - 2008 Variable Report Longitudinal Data File.
Cologne
:
GESIS Data Archive
.
Fernández
,
J. J.
and
Lutter
,
M.
(
2013
) ‘
Supranational cultural norms, domestic value orientations and the diffusion of same-sex union rights in Europe, 1988–2009
’,
International Sociology
28
:
102
20
. doi:
Finch
,
J. V.
and
Mason
,
J.
(
1993
)
Negotiating Family Responsibilities
,
London
:
Routledge
.
Fingerman
,
K. L.
,
Pillemer
,
K. A.
,
Silverstein
,
M.
and
Suitor
,
J. J.
(
2012
) ‘
The baby boomers’ intergenerational relationships
’,
The Gerontologist
52
(
2
):
199
209
. doi:
Fokkema
,
T.
,
De Jong Gierveld
,
J.
and
Dykstra
,
P. A.
(
2012
) ‘
Cross-national differences in older adult loneliness
’,
The Journal of Psychology
146
(
1–2
):
201
28
. doi:
Friedman
,
D.
,
Hechter
,
M.
and
Kreager
,
D.
(
2008
) ‘
A theory of the value of grandchildren
’,
Rationality and Society
20
(
1
):
31
63
. doi:
Ganjour
,
O.
and
Widmer
,
É. D.
(
2016
) ‘
Patterns of family salience and welfare state regimes: sociability practices and support norms in a comparative perspective
’,
European Societies
6696
(
June
):
201
20
. doi:
Gans
,
D.
,
Silverstein
,
M.
and
Lowenstein
,
A.
(
2009
) ‘
Do religious children care more and provide more care for older parents? A study of filial norms and behaviors across five nations
’,
Journal of Comparative Family Studies
40
(
2
):
187
201
.
Goerres
,
A.
and
Tepe
,
M.
(
2010
) ‘
Age-based self-interest, intergenerational solidarity and the welfare state: a comparative analysis of older people's attitudes towards public childcare in 12 OECD countries
’,
European Journal of Political Research
49
(
6
):
818
51
. doi:
Halman
,
L.
and
Riis
,
O.
(eds.) (
2003
)
Religion in a Secularizing Society: The Europeans’ Religion at the end of the 20th Century
,
Leiden
:
Brill
.
Halman
,
L.
,
Sieben
,
I.
and
van Zundert
,
M.
(
2011
)
European Values Studies: Atlas of European Values, Volume 14: Trends and Traditions at the Turn of the Century
,
Leiden
:
Brill
.
Hellevik
,
O.
(
2009
) ‘
Linear versus logistic regression when the dependent variable is a dichotomy
’,
Quality and Quantity
43
:
59
74
. doi:
Inglehart
,
R.
(
1990
)
Culture Shift in Advanced Industrial Society
,
Princeton
:
Princeton University Press
.
Inglehart
,
R. F.
(
2008
) ‘
Changing Values among Western Publics from 1970 to 2006
’,
West European Politics
31
(
1-2
):
130
46
. doi:
Inglehart
,
R.
and
Baker
,
W. E.
(
2000
) ‘
Modernization, cultural change, and the persistence of traditional values
’,
American Sociological Review
65
(
1
):
19
51
. doi:
Isengard
,
B.
(
2013
) ‘
The Apple doesn't live far from the tree’: living distances between parents and their adult children in Europe
’,
Comparative Population Studies
38
(
2
):
237
62
.
Jappens
,
M.
and
Van Bavel
,
J.
(
2012
) ‘
Regional family norms and child care by grandparents in Europe
’,
Demographic Research
27
(
4
):
85
120
. doi:
Kalmijn
,
M.
and
Saraceno
,
C.
(
2008
) ‘
A comparative perspective on intergenerational support : responsiveness to parental needs in individual and familialistic countries
’,
European Societies
10
(
3
):
479
508
. doi:
Kluge
,
S.
(
2000
) ‘
Empirically grounded construction of types and typologies in qualitative social research
’,
Forum Qualitative Sozialforschung/Forum: Qualitative Social Research
1
(
1
), http://www.qualitative-research.net/index.php/fqs/article/view/1124/2497Volume
Kocourková
,
J.
(
2002
) ‘
Leave arrangements and childcare services in Central Europe: policies and practices before and after the transition
’,
Community, Work & Family
5
(
3
):
301
18
. doi:
Kreager
,
P.
and
Schröder-Butterfill
,
E.
(
2008
) ‘
Indonesia against the trend? Ageing and inter-generational wealth flows in two Indonesian communities
’,
Demographic Research
19
(
52
):
1781
810
. doi:
Lazarsfeld
,
P.
(
1937
) ‘
Some remarks on typological procedures in the social sciences
’,
Zeitschrift für Sozialforschung
6
:
119
39
. doi:
Leitner
,
S.
(
2003
) ‘
Varieties of familialism: the caring function of the family in comparative perspective
’,
European Societies
5
(
4
):
353
75
. doi:
Lesthaeghe
,
R.
(
2014
) ‘
The second demographic transition: a concise overview of its development: table 1
’,
Proceedings of the National Academy of Sciences of the United States of America
111
(
51
):
18112
115
.
Liefbroer
,
A. C.
and
Billari
,
F. C.
(
2010
) ‘
Bringing norms back in: a theoretical and empirical discussion of their importance for understanding demographic behaviour
’,
Population, Space and Place
16
(
4
):
287
305
. doi:
Lin
,
K.
(
2005
) ‘
Cultural traditions and the Scandinavian social policy model
’,
Social Policy & Administration
39
(
7
):
723
39
. doi:
Lin
,
J.-P.
and
Yi
,
C.-C.
(
2013
) ‘
A comparative analysis of intergenerational relations in East Asia
’,
International Sociology
28
(
3
):
297
315
. doi:
Lowenstein
,
A.
and
Daatland
,
S. O.
(
2006
) ‘
Filial norms and family support in a comparative cross-national context: evidence from the OASIS study
’,
Ageing and Society
26
:
203
23
. doi:
Manow
,
P.
(
2004
)
The Good, the Bad, and the Ugly: Esping-Andersen's Regime Typology and the Religious Roots of the Western Welfare State
, MPIfG working paper, No. 04/3.
Masterson
,
C. R.
and
Hoobler
,
J. M.
(
2015
) ‘
Care and career: a family identity-based typology of dual-earner couples
’,
Journal of Organizational Behavior
36
:
75
93
. doi:
Minkov
,
M.
,
Blagoev
,
V.
and
Hofstede
,
G. H.
(
2013
) ‘
The Boundaries of culture: Do questions about societal norms reveal cultural differences?
’,
Journal of Cross-Cultural Psychology
44
(
149
):
1094
106
. doi:
Mood
,
C.
(
2010
) ‘
Logistic regression: why we cannot do what we think we can do, and what we can do about it
’,
European Sociological Review
26
(
1
):
67
82
. doi:
Mudrazija
,
S.
(
2016
) ‘
Public transfers and the balance of intergenerational family support in Europe
’,
European Societies
6696
(
July
):
1
23
.
Ogg
,
J.
and
Renaut
,
S.
(
2006
) ‘
The support of parents in old age by those born during 1945–1954: a European perspective
’,
Ageing and Society
26
(
5
):
723
43
. doi:
Popenoe
,
D.
(
2009
) ‘
Cohabitation, marriage, and child wellbeing: a cross-national perspective
’,
Society
46
(
5
):
429
36
. doi:
Robila
,
M.
(
2004
) ‘
Families in Eastern Europe : context, trends and variations
’,
Contemporary Perspective in Family Research
5
(
4
):
1
14
.
Rothman
,
K. J.
(
1986
)
Modern Epidemiology
,
Boston, MA
:
Little, Brown and Company
.
Sabatier
,
C.
,
Mayer
,
B.
,
Friedlmeier
,
M.
,
Lubiewska
,
K.
and
Trommsdorff
,
G.
(
2011
) ‘
Religiosity, family orientation, and life satisfaction of adolescents in four countries
’,
Journal of Cross-Cultural Psychology
42
(
8
):
1375
93
. doi:
Sobotka
,
T.
(
2008
) ‘
Overview chapter 6: the diverse faces of the second demographic transition in Europe
’,
Demographic Research
19
:
171
224
. doi:
Syltevik
,
L. J.
(
2010
) ‘
Sense and sensibility: cohabitation in ‘cohabitation land
’,
The Sociological Review
58
(
3
):
444
62
. doi:
Syltevik
,
L. J.
(
2014
) ‘
Cohabitation from illegal to institutionalized practice: the case of Norway 1972–2010
’,
The History of the Family
20
(
4
):
515
29
.
Szelewa
,
D.
and
Polakowski
,
M. P.
(
2008
) ‘
Who cares? Changing patterns of childcare in Central and Eastern Europe
’,
Journal of European Social Policy
18
(
2
):
115
31
. doi:
Therborn
,
G.
(
2004
)
Between Sex and Power
,
London
:
Routledge
.
Therborn
,
G.
(
2007
) ‘
Is there a future for the family?
’,
Public Policy Research
14
(
1
):
41
6
. doi:
Wolfe
,
A.
(
1989
)
Whose Keeper? Social Science and Moral Obligation
,
Berkeley, CA
:
University of California Press
.
Yamagishi
,
T.
and
Yamagishi
,
M.
(
1994
) ‘
Trust and commitment in the United States and Japan
’,
Motivation and Emotion
18
(
2
):
129
66
. doi:
EVS (European Values Study)Oasis (Old age and autonomy: the role of service systemsGGP (Gender and Generations Programme)SHARE (Survey of Health, Ageing and Retirement in Europe)ESS (European Social Survey)
Year of most recent data collection wave 2008 2000–2003 Wave 1: 2004–2008
Wave 2: ?? 
Wave 1: 2004
Wave 5: 2013 
 
No. of countries included 20 Wave 1: 19 countries, wave 2: 10 countries (bold) 20 countries  
Names of countries included Nordic: DK, FI, NO, SE
Central: AU, BE, FR, GE, LUX
Med.: GR, IT, PO, SP
East: CZ, HU, PO, SL
West: NL, CZ, GB 
Nordic: NO
Central: GE
Med.: SP, IL
West: GB 
Nordic: NO, SE
Central: AU, BE, FR, GE
Med.: IT
East: CZ, HU, PL, BU, ES, LT, GO, RO, RF
West: NL
Other: AU, JP 
Nordic: DK, SE
Central: AU, BE, FR, GE, LUX
Med.: GR, IT, PO, SP, IL
East: CZ, PL, HU, SL, ES
West: NL, CH, IR 
 
Variables measuring intergenerational obligations/norms Three binary items: two filial, one parental Only filial Both waves:
Parental:
-Grandparents look after grandchildren if parents are unable to do so
-Parents should provide financial help adult children when in difficulty
-If adult children in need, parents adjust own lives to help them
Filial:
-Children should take responsibility for parental care if parents in need
-Children should adjust working lives to the needs of their parents
-Children should provide financial help if parents financial difficulty
-Children should live with parents when no longer look after themselves [When parents in need, daughters should take more caring responsibility] 
Actual help given/received, feeling of appreciation, no questions about norms None 
Population age groups in sample 18–80  18–80 50+  
Comments Good repr. of both N/W and S/E  Nordic, Mediterranean, Western under-represented, so not possible to make the comparisons I want between ascending/descending family cultures Nordic, Western under-represented  
EVS (European Values Study)Oasis (Old age and autonomy: the role of service systemsGGP (Gender and Generations Programme)SHARE (Survey of Health, Ageing and Retirement in Europe)ESS (European Social Survey)
Year of most recent data collection wave 2008 2000–2003 Wave 1: 2004–2008
Wave 2: ?? 
Wave 1: 2004
Wave 5: 2013 
 
No. of countries included 20 Wave 1: 19 countries, wave 2: 10 countries (bold) 20 countries  
Names of countries included Nordic: DK, FI, NO, SE
Central: AU, BE, FR, GE, LUX
Med.: GR, IT, PO, SP
East: CZ, HU, PO, SL
West: NL, CZ, GB 
Nordic: NO
Central: GE
Med.: SP, IL
West: GB 
Nordic: NO, SE
Central: AU, BE, FR, GE
Med.: IT
East: CZ, HU, PL, BU, ES, LT, GO, RO, RF
West: NL
Other: AU, JP 
Nordic: DK, SE
Central: AU, BE, FR, GE, LUX
Med.: GR, IT, PO, SP, IL
East: CZ, PL, HU, SL, ES
West: NL, CH, IR 
 
Variables measuring intergenerational obligations/norms Three binary items: two filial, one parental Only filial Both waves:
Parental:
-Grandparents look after grandchildren if parents are unable to do so
-Parents should provide financial help adult children when in difficulty
-If adult children in need, parents adjust own lives to help them
Filial:
-Children should take responsibility for parental care if parents in need
-Children should adjust working lives to the needs of their parents
-Children should provide financial help if parents financial difficulty
-Children should live with parents when no longer look after themselves [When parents in need, daughters should take more caring responsibility] 
Actual help given/received, feeling of appreciation, no questions about norms None 
Population age groups in sample 18–80  18–80 50+  
Comments Good repr. of both N/W and S/E  Nordic, Mediterranean, Western under-represented, so not possible to make the comparisons I want between ascending/descending family cultures Nordic, Western under-represented  

Appendix 2. Distribution of normative positions within each country

RegionCountrySymmetrical familialism (%)Descending familialism (%)Ascending familialism (%)Situationalism (%)Total (%)
Nordic 208 Denmark 36.90 31.30 12.30 19.50 100.00 
 246 Finland 42.40 29.60 9.90 18.10 100.00 
 578 Norway 48.60 40.00 4.10 7.40 100.00 
 752 Sweden 39.40 48.50 3.90 8.30 100.00 
Western 528 Netherlands 48.50 34.10 8.10 9.30 100.00 
 756 Switzerland 60.20 22.30 8.70 8.90 100.00 
 826 Great Britain 61.80 20.40 9.30 8.50 100.00 
Central 40 Austria 48.80 14.10 19.90 17.20 100.00 
 56 Belgium 62.90 12.60 15.00 9.50 100.00 
 250 France 73.10 9.10 13.80 4.00 100.00 
 276 Germany 53.30 11.80 12.70 22.30 100.00 
 442 Luxembourg 72.10 10.90 11.30 5.70 100.00 
Eastern 203 Czech Republic 59.30 4.50 19.70 16.50 100.00 
 348 Hungary 70.30 7.30 15.30 7.00 100.00 
 616 Poland 80.30 2.70 11.20 5.70 100.00 
 703 Slovak Republic 71.90 3.90 16.10 8.10 100.00 
Med 300 Greece 79.50 3.60 12.50 4.40 100.00 
 380 Italy 86.80 4.30 6.50 2.50 100.00 
 620 Portugal 85.40 3.40 8.20 3.10 100.00 
 724 Spain 82.10 6.40 7.80 3.70 100.00 
RegionCountrySymmetrical familialism (%)Descending familialism (%)Ascending familialism (%)Situationalism (%)Total (%)
Nordic 208 Denmark 36.90 31.30 12.30 19.50 100.00 
 246 Finland 42.40 29.60 9.90 18.10 100.00 
 578 Norway 48.60 40.00 4.10 7.40 100.00 
 752 Sweden 39.40 48.50 3.90 8.30 100.00 
Western 528 Netherlands 48.50 34.10 8.10 9.30 100.00 
 756 Switzerland 60.20 22.30 8.70 8.90 100.00 
 826 Great Britain 61.80 20.40 9.30 8.50 100.00 
Central 40 Austria 48.80 14.10 19.90 17.20 100.00 
 56 Belgium 62.90 12.60 15.00 9.50 100.00 
 250 France 73.10 9.10 13.80 4.00 100.00 
 276 Germany 53.30 11.80 12.70 22.30 100.00 
 442 Luxembourg 72.10 10.90 11.30 5.70 100.00 
Eastern 203 Czech Republic 59.30 4.50 19.70 16.50 100.00 
 348 Hungary 70.30 7.30 15.30 7.00 100.00 
 616 Poland 80.30 2.70 11.20 5.70 100.00 
 703 Slovak Republic 71.90 3.90 16.10 8.10 100.00 
Med 300 Greece 79.50 3.60 12.50 4.40 100.00 
 380 Italy 86.80 4.30 6.50 2.50 100.00 
 620 Portugal 85.40 3.40 8.20 3.10 100.00 
 724 Spain 82.10 6.40 7.80 3.70 100.00 

Appendix 3. The education variable

Highest educational level attained respondent (eight categories) (Q110) * education level (Q110) (recoded)

Education level (Q110) (recoded)Total
1 Lower2 Middle3 Upper
Highest educational level attained respondent (eight categories) (Q110) 1. Inadequately completed elementary education Count 2124 2124 
% within education level (Q110) (recoded) 10.6% 0.0% 0.0% 3.2% 
2. Completed (compulsory) elementary education Count 6594 6594 
% within education level (Q110) (recoded) 33.0% 0.0% 0.0% 9.8% 
3. Incomplete secondary school: technical/vocational type Count 11,251 11251 
% within education level (Q110) (recoded) 56.3% 0.0% 0.0% 16.7% 
4. Complete secondary school: technical/vocational type/secondary Count 6335 6335 
% within education level (Q110) (recoded) 0.0% 20.2% 0.0% 9.4% 
5. Incomplete secondary: university-preparatory type/secondary, Count 7849 7849 
% within education level (Q110) (recoded) 0.0% 25.0% 0.0% 11.7% 
6. Complete secondary: university-preparatory type/full secondary Count 17,175 17175 
% within education level (Q110) (recoded) 0.0% 54.8% 0.0% 25.6% 
7. Some university without degree/higher education – lower-level tertiary Count 9473 9473 
% within education level (Q110) (recoded) 0.0% 0.0% 59.8% 14.1% 
8. University with degree/higher education – upper-level tertiary Count 6370 6370 
% within education level (Q110) (recoded) 0.0% 0.0% 40.2% 9.5% 
Total Count 19,969 31,359 15,843 67,171 
% within education level (Q110) (recoded) 100.0% 100.0% 100.0% 100.0% 
Education level (Q110) (recoded)Total
1 Lower2 Middle3 Upper
Highest educational level attained respondent (eight categories) (Q110) 1. Inadequately completed elementary education Count 2124 2124 
% within education level (Q110) (recoded) 10.6% 0.0% 0.0% 3.2% 
2. Completed (compulsory) elementary education Count 6594 6594 
% within education level (Q110) (recoded) 33.0% 0.0% 0.0% 9.8% 
3. Incomplete secondary school: technical/vocational type Count 11,251 11251 
% within education level (Q110) (recoded) 56.3% 0.0% 0.0% 16.7% 
4. Complete secondary school: technical/vocational type/secondary Count 6335 6335 
% within education level (Q110) (recoded) 0.0% 20.2% 0.0% 9.4% 
5. Incomplete secondary: university-preparatory type/secondary, Count 7849 7849 
% within education level (Q110) (recoded) 0.0% 25.0% 0.0% 11.7% 
6. Complete secondary: university-preparatory type/full secondary Count 17,175 17175 
% within education level (Q110) (recoded) 0.0% 54.8% 0.0% 25.6% 
7. Some university without degree/higher education – lower-level tertiary Count 9473 9473 
% within education level (Q110) (recoded) 0.0% 0.0% 59.8% 14.1% 
8. University with degree/higher education – upper-level tertiary Count 6370 6370 
% within education level (Q110) (recoded) 0.0% 0.0% 40.2% 9.5% 
Total Count 19,969 31,359 15,843 67,171 
% within education level (Q110) (recoded) 100.0% 100.0% 100.0% 100.0% 

Appendix 4. Regression parameters and predicted probabilities from linear regression.

Regression parameters from linear regression

Symmetrical familialismDescending familialismAscending familialismSituationalism
Region Nordic 
 Western .1110915 *** −.0936618 *** .014282 −.0317117 *** 
 Central .1525415 *** −.232694 *** .0750192 *** −.0051333 
 Eastern .2228598 *** −.2969413 *** .0912946 *** .0172131 
 Mediterranean .30397 *** −.263971 *** .0175027 −.0575017 *** 
Rel. practice Frequent practitioner 
 Infrequent practitioner −.1024614 *** .0315539 *** .0363193 *** .0345883 *** 
 Non-practitioner −.1522853 *** .075288 *** .0270437 ** .0499536 *** 
Education Lower (<complete secondary) 
 Middle (complete secondary) −.0727095 *** .0333582 *** .0081245 .0312269 *** 
 Upper (some tertiary)3 −.0816375 *** .0613927 *** .0034575 .0237023 *** 
Parenthood No children 
 Has children .0256014 ** .0437301 *** −.0413003 *** −.0280312 *** 
Gender Male 
 Female −.0364543 *** .0253676 *** .0001789 .0112657 ** 
Rel denom. None 
 Catholic .074212 *** −.0299652 *** .0081974 −.0360494 *** 
 Protestant .0130202 .0016832 .0069509 .004386 
 Other (Muslim, Orthodox) .1841744 *** −.0686353 *** .0177506 −.0332897 *** 
Age  .00015 .0000508 .0001268 .000074 
Constant  .5995713 .2509651 .0633929 .0860707 
Symmetrical familialismDescending familialismAscending familialismSituationalism
Region Nordic 
 Western .1110915 *** −.0936618 *** .014282 −.0317117 *** 
 Central .1525415 *** −.232694 *** .0750192 *** −.0051333 
 Eastern .2228598 *** −.2969413 *** .0912946 *** .0172131 
 Mediterranean .30397 *** −.263971 *** .0175027 −.0575017 *** 
Rel. practice Frequent practitioner 
 Infrequent practitioner −.1024614 *** .0315539 *** .0363193 *** .0345883 *** 
 Non-practitioner −.1522853 *** .075288 *** .0270437 ** .0499536 *** 
Education Lower (<complete secondary) 
 Middle (complete secondary) −.0727095 *** .0333582 *** .0081245 .0312269 *** 
 Upper (some tertiary)3 −.0816375 *** .0613927 *** .0034575 .0237023 *** 
Parenthood No children 
 Has children .0256014 ** .0437301 *** −.0413003 *** −.0280312 *** 
Gender Male 
 Female −.0364543 *** .0253676 *** .0001789 .0112657 ** 
Rel denom. None 
 Catholic .074212 *** −.0299652 *** .0081974 −.0360494 *** 
 Protestant .0130202 .0016832 .0069509 .004386 
 Other (Muslim, Orthodox) .1841744 *** −.0686353 *** .0177506 −.0332897 *** 
Age  .00015 .0000508 .0001268 .000074 
Constant  .5995713 .2509651 .0633929 .0860707 

*p < .05.

**p < .010.

***p < .000.

Predicted probabilities for selected groups based on linear regression

Symmetrical (%)Descending (%)Ascending (%)Situationalism (%)
Practicing Catholic male without children, low education Mediterranean 97.07 <0 7.87 <0 
Eastern 88.96 <0 15.25 3.63 
 Central 81.93 <0 13.62 5.86 
 Western 77.78 12.49 7.54 2.18 
 Nordic 66.67 21.86 6.12 5.35 
Non-practicing Protestant mother, high education Mediterranean 63.87 19.21 7.59 9.33 
Eastern 55.76 15.91 14.97 13.36 
Central 48.73 22.33 13.34 15.60 
 Western 44.58 36.24 7.27 11.91 
 Nordic 33.47 45.60 5.84 15.08 
Symmetrical (%)Descending (%)Ascending (%)Situationalism (%)
Practicing Catholic male without children, low education Mediterranean 97.07 <0 7.87 <0 
Eastern 88.96 <0 15.25 3.63 
 Central 81.93 <0 13.62 5.86 
 Western 77.78 12.49 7.54 2.18 
 Nordic 66.67 21.86 6.12 5.35 
Non-practicing Protestant mother, high education Mediterranean 63.87 19.21 7.59 9.33 
Eastern 55.76 15.91 14.97 13.36 
Central 48.73 22.33 13.34 15.60 
 Western 44.58 36.24 7.27 11.91 
 Nordic 33.47 45.60 5.84 15.08 

Note: The results of the linear and logistic regression are, as expected, very similar as to which variables they point out as significant and the relative magnitude of their effects. The predicted probabilities do differ slightly in the predictions of the proportions supporting the descending familialism position in the Nordic and Western regions, where the linear regression tends to overestimate the probabilities for the practicing Catholic males with low education and underestimate those for the Protestant, non-practicing mothers with high education. This may be due to the very low number of observations in the group of practicing Catholic males in these regions.

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.