1 Introduction: The Study of Informal Social Support

The social support an individual receives is a phenomenon of growing interest, since it influences the subject’s well being through various mechanisms. The term social support is complex, and we can identify three levels (Gottlieb 1981:32): (1) social support defined in terms of the person’s degree of integration, (2) social support as a product of the interaction of the individual in a social network with specific structural properties, and (3) social support understood as access to resources typically present in one’s most intimate relationships. There are three corresponding levels on which to analyze the relationship between the individual and the social environment: the macro-level (social integration), the meso-level (social network), and the micro-level (intimate relationships). This paper focuses on the third level, studying the closest relationships (family and friends) to which the individual can turn when seeking informal support (as opposed to the formal support provided by institutions).

Intimate relationships are the greatest source of social support (Cutrona 1996). The support gained from intimate relationships, rather than social relations in general, is associated with one’s state of physical and mental health (Lin et al. 1986; Rogers 1987). Beach et al. (1996) stress the importance of one’s partner as a source of social support: the spouse is often mentioned as the person one is most likely to ask for help in a moment of need (Dakof and Taylor 1990; Berg-Cross 1974). The partner plays a critical role in giving various kinds of support (Reiss 1990) and is the person who provides almost all of the kinds of help for married people (Beach et al. 1993), to such an extent that the presence of support from other sources cannot compensate for lack of support from one’s spouse (Brown and Harris 1978; Coyne and DeLongis 1986; Lieberman 1982). This phenomenon is due to the fact that caring, understood as the wish to help the other by giving aid and emotional support when needed, is the central component of the vision of love (Kelley 1983), a component superimposed on the definition of social support that conceives of it as responsivity to the other’s needs.

Informal sources of social support depend on multiple factors, from age (making it less likely that one will ask for help from one’s parents when they are very old and impossible when they are no longer living) to whether one is married (one of the most important sources for married people is their spouse) or time living in a place (the longer one has been living somewhere, the more likely is it that one will have created ties of friendship with its inhabitants). Given the importance of the spouse as a source of social support demonstrated in previous studies, this paper focuses on people in a stable couple (both married and living together) to identify the factors that influence whether one chooses one’s spouse as the primary source of support. This contribution is new, first because sociologists have usually studied the population in general and have concluded that having a partner is one of the most important variables for explaining social support. In this case, however, we have chosen to study people in couples because marriage requires particular attention in social research. It is probably the most common social context in which adults find a relationship that is vital to their psychological and social well being (Acitelli 1996). Further, marital relationships provide a relatively homogeneous context for examining social support, since the exchanging help is normative within this relationship (Beach et al. 1996). In spite of this relative homogeneity, we must take into account that the role of the spouse is influenced by social context, since it is affected by norms, behavioral expectations, and macrosocial values, as well as characteristics of the personal network of both members of the couple. Second, this study contributes to determining the extent to which the role of the spouse is similar or different in Western countries by comparing the role of the spouse as a source of social support in thirteen countries.

Since the nineties, when comparative analyses of Western countries began to be performed, one of the most frequently used perspectives has been to classify countries according to the different welfare regimes proposed by Esping-Andersen (1990, 1999). Traditionally, the literature on these regimes has focused on the principal care-taker (state, market, or family) for the elderly and the ill, on the one hand, and preschool age children on the other. The third contribution of this research is that it analyzes situations of need that affect the population as a whole (and not specific sectors) and that, if not attended to, erode the individual’s well being: achieving aid when one is sick, receiving support when one is sad, and being lent money. As both Esping-Andersen and Palier (2010) indicate, throughout the life cycle, the role of the welfare state is important in the first and last years of a person’s life, whereas most social aid is provided by the family.

The classic division by Burgess et al. (1960) between institutional family and family based on companionship has given way to a pluralization of kinds of couples which, in turn, shows that the phenomenon of individualization has reached the area of the family. Beck-Gernsheim (2002) argues that a transition to a “post-familial family” is occurring (Rosenmayr 1992), characterized by increasingly elective relationships between members. Relationships between couples are probably the family relationships that have evolved most. If decades ago one’s relationship with one’s spouse had a markedly prescriptive character and was established for life, today the existence and duration of this relationship are much more strongly conditioned by whether both members of the couple are satisfied; in this sense, these relationships bear more resemblance to friendships. This process of individualization has led some authors to stress the erosion of social networks composed of family members and friends (Beck 1992) and even to affirm that the family is no longer the “natural” unit of social support due to the rise in divorce rates, which has made family relationships unstable (Van Dijk 1998). Along similar lines, Jacobson (1990) indicates that the expectations associated with the family become more confusing in the case of reconstituted step-family members, composed of one or two spouses who have been married previously. Against these pessimistic views of family ties, our research seeks to determine whether the partner has been weakened as a provider of social support.

The sources of social support for people in couples are not limited to the spouse, however significant a source he or she is. Two other kinds of relationships are also very important: family relationships, especially with the mother, and friendships. Although relationships between family members are increasingly elective, this voluntary quality occurs to a greater extent in the case of the spouse and his or her family, since relationships with one’s closest relatives (especially with parents) are not chosen and the norms for them, particularly those referring to the relationships between parents and children, are clearly established. As to relationships between siblings, relationships that are at least in their origin not chosen, it is striking that Cicirelli (1994), in an international comparison, stresses that these links tend to be discretionary in industrial societies and compulsory in non-industrial societies. In non-industrial societies, cooperation between siblings is fundamental to achieving economic and marital goals. As a general rule, one expects support from one’s family (at least most immediate family) in the face of any adversity, without asking anything in exchange and putting aside conflicts that might exist between their members. Altruism is the main motive mentioned in the literature for giving support, and it is generally assumed that it is inherent in family relationships and remains constant (Van Dijk 1998). In contrast, friendships are chosen based on common interests and shared activities and therefore tend to be based more strongly on reciprocity and exchange than on family ties (Allan 1986; Roberto 1989). We can thus expect friendships to continue as long as both parties feel that the benefits outweigh the costs. International studies of friendship show that the elements that define friendship in all countries are companionship and understanding, that is, expressive characteristics that focus on the quality of the relationship, not its instrumental character (De Federico 2011). There is empirical evidence that family and friends play complementary roles in the support network: the family provides instrumental or material support, and friends emotional support (Gerstel and Gallagher 1994; Jerome 1990; Litwak 1985; Rook 1989).

2 Objectives and Hypotheses

The first goal of this paper is to confirm what importance the partner, family, and close friends have in providing informal social support to people in couples. Scholars who follow the theory of individualization believe that traditional ties with family and friends are eroded in developed societies (Beck 1992). Social network analysts suggest, however, that what we are seeing is a transformation in the kind of personal networks, which are more dispersed and less dense, but that social ties continue to be crucial, shifting from group-based networks to individualized networks (Wellman 2001). This study attempts to confirm the importance of the personal network in providing well being.

The second goal proposed is to discover what factors (both individual and macrosocial) influence choice of the person to whom one turns in a situation of need. We believe that the sociodemographic variables and variables related to the characteristics of personal networks used traditionally in the literature are relevant for studying this phenomenon. However, researchers in this field stress that, in addition to individual characteristics and characteristics of the properties of the relationships that facilitate or prevent giving and receiving support, one must take into account sociocultural factors that promote or discourage people when lending support (House 1981). We must thus consider macrosocial factors, such as national contexts, since they influence the way of understanding the different relationships (parental, couples, friendship) and thus structure the “specific sociability” (De Federico 2011) of each country (Höllinger and Haller 1990).

The now classic distinction by Esping-Andersen (1990), subsequently revised (1999), between different welfare regimes (liberal, social democratic, and conservative) can shed light on the countries in which it is more likely for people to turn to family and friends as sources of informal social support, as opposed to other, more formal sources, whether provided by the state or the market. The concept of defamiliarization (1999), which Esping-Andersen adds to the concept of decommodification proposed in his first work (1990), is especially relevant here. Decommodification refers to the degree to which welfare states affect the monetary nexus when they guarantee certain rights independently of participation in the market. This conception assumes (as feminist studies have criticized) that individuals are mercantilized, that is, that they are wage earners on the market, which does not take into account the contribution of the family, and especially that of women, in providing well being (Orloff 1993; O’Connor 1993; Daly 1994). Responding to this criticism, Esping-Andersen incorporates the concept of defamiliarization in his review of welfare regimes, which refers to policies that reduce the dependence of the individual on the family. A defamiliarizing state decreases the responsibilities related to well being and attending to the family unit, whereas a familialist welfare state assigns the family the maximum obligations, obligations that fall mainly on the woman, hindering her participation in the labor market.

According to the two objectives mentioned above, we attempt to corroborate three hypotheses:

  1. 1.

    Informal social support continues to be the main source of help in cases of need among people in couples, since most turn first to their spouse and second to close relatives (parents and, to a lesser extent, children and siblings).

  2. 2.

    There are social characteristics and characteristics of the composition of personal networks that influence choice of the person from whom one requests aid. People with a higher degree of individualism are more likely to turn to their spouse and friends, whereas those with a more collectivist or group character turn first to their parents, children and siblings.

  3. 3.

    Countries’ degree of defamiliarization influences the decision of whom one turns to when asking for help. Liberal and social democratic countries are more defamiliarized and thus turn more to elective relationships (partner and friends) and less to family than conservative and Mediterranean countries.

3 Methodology: Sources of Data and Variables Used

The database used is the module on social support, “Social Relations and Support Systems (Social Network II)” of the International Social Survey Programme (ISSP), performed in 2001. Specifically, we have chosen national samples (of individuals over 18 years of age) of people with a stable partner (married or living together), from Austria, Canada, Denmark, Finland, France, East and West Germany, Great Britain, Italy, New Zealand, Norway, Spain, and the United States. The total sample of people in couples is 9,996, with samples ranging from 564 cases from the United States to 1,134 from Norway. The country with the lowest percentage of people in couples relative to the total national sample is the United States, with 49 %, and the highest is Italy, with 86 %. This percentage ranges in most of the countries from 60 to 70 %.

3.1 Dependent Variables

We use three questions to study social support for the individual. The very complexity of the concept of social support we have just mentioned also explains the diverse measurements used to capture it (Barrera 1981:70). Some scholars analyze who provides support, whereas others focus on the subjective evaluation that the individual makes of support. Still others study the activities involved in providing this aid. Perceived support is a better predictor of the state of health than more objective indicators of a person’s social support network (Sarason et al. 1990). As Hobfoll (1988) indicates, a person’s evaluation of the social support he or she perceives contains much objective information that is not measured by a limited number of objective indicators of support received. Due to the data available, this paper will study the perception of whether support is available. From the different kinds of support about which participants were asked in the survey, we have chosen common situations that can happen to any person. The first two items refer to instrumental support, in a case of domestic help (care when sick) and in the case of economic need (loan), and the third to emotional support. The precise phrasing of these questions in the questionnaire is as follows:

Now we would like to ask you how you would get help in situations that anyone could find herself or himself in.

First, suppose you had the’ flu and had to stay in bed for a few days and needed help around the house, with shopping and so on. Who would you turn to first for help?

Now, suppose you needed to borrow a large sum of money. Who would you turn to first for help?

Now suppose you felt just a bit down or depressed, and you wanted to talk about it. Who would you turn to first for help?

To create the dependent variables, we inspected the table of frequencies of these three questions for the whole sample of people in couples and for each of the countries (Table 1). For all of the respondents, we observe that the partner is the person most mentioned in all cases as the main informal support: over eight of every ten respondents would turn first to their partner if they were sick, over six of every ten would talk to their partner if they were depressed, and around a fourth would ask their partner for money.Footnote 1 It is not surprising that the partner would be seen less as a source of economic support. On the one hand, in most cases, spouses think of themselves as forming an economic unit. If one of the members needs money, the other probably does too and thus would not be the person to whom to turn. Second, in some countries, the rate of employment for women is much lower than for men, decreasing the possibility of men turning to their partners for economic support. After one’s partner, the second-most frequently mentioned person to whom one would turn first when needing help was the mother in the case of flu (5 %). For needing money, the mother and father show similar figures (18 % if we add both) and, for someone to talk to when one is sad, respondents turned to a close friend (14 %).

Table 1 Distribution of frequencies by country of first person whom respondents in couples would ask for supporta
Full size table

The pattern found for all of the survey respondents concerning whom one asks first for help was confirmed in the countries when they were analyzed individually.Footnote 2 In all countries, the respondents turn first to their partner if they have the flu or are sad, although in the case of Italy, the partner is only slightly higher as a choice than a close friend if one is depressed (36.6 % as opposed to 33 %). In Italy, the father scores higher than the partner as the first person one would ask to borrow money, but this would be the mother in East Germany. If we take the sum of the two parents, however, several countries show that respondents would ask parents first, before asking their partner: the United States, East Germany, and France.

National differences are considerable, with the greatest divergence between the countries in asking for aid from one’s partner when one is sick or depressed, where Italy and Norway represent the minimums and maximums (50.7 and 96.1 % for having the flu and 36.6 and 76.9 % for being sad). We can classify countries according to whether they are above or below the average in turning to spouses in search of the three kinds of support. The countries above average are Finland, Great Britain, New Zealand, and Norway, and below are Spain and Italy. The other countries show some forms of support as above average and some as below.

As to the second most-frequently mentioned person to whom people would turn if they needed money, respondents mention one of their parents, with quite similar figures for both father and mother, except in the cases of Great Britain and East Germany (where the mother scores considerably higher than the father) and Italy (where the opposite is the case). Being sick is the situation that shows the greatest variation among the countries. In some countries, the mother is the second most-mentioned person to whom people would turn (especially in Mediterranean countries, but also in the United States), while in others respondents would turn to a similar extent to their mother and their children (Great Britain and France). In a third group, children score higher than their mother as providers of help in case of illness. For help when sick, in four countries, no respondents mention the father as the first source of support (New Zealand, Austria, East Germany, and Finland). We thus confirm the first hypothesis through descriptive analysis: in cases of need one turns principally to informal social support, especially that of one’s partner and, to a lesser extent, one’s parents or children.Footnote 3 The countries that are farther from the average in turning to members of the previous generations to ask for economic aid or help in illness, are Canada, New Zealand, Austria, Denmark, Norway, and Finland (below average) and the United States, East Germany, Spain, and Italy (above average).

To conclude, the data in Table 1 show that citizens in liberal and social democratic countries turn to their partner more often, and in Mediterranean countries less often. Just the opposite occurs in the case of asking for aid from parents. This pattern makes sense, since the partner is chosen and the relationship is a freer one, characteristic of countries with higher degrees of individualization and higher rates of employment for women, which enables men to ask for economic help from their spouses. In contrast, Italy and Spain, characterized by familism based on family and parental solidarity (Saraceno 1994; Alberdi 1999; Naldini 2006), confirm this trait, as individuals turn more to parents than in other countries when seeking aid. Finally, respondents turn more to friends in France, West Germany, Italy, and the United States and less in Austria, Canada, Denmark, Finland, East Germany, Great Britain, Norway, and Spain. Spain stands out as the only country in which one turns more to one’s children than to close friends (9 % as opposed to 7.8 %). We do not find a clear pattern here, because countries belonging to the different regimes score both above and below the average. In this case, Italy and Spain belong to different groups, since the Italians turn to friends to a greater extent than do Spaniards (33 % as opposed to 7.8 %). Although this initial descriptive analysis generates interesting results, it is limited, since factors such as the different composition of personal networks or sociodemographic variables may influence the person one chooses. Therefore, the next section will control for these variables when the analysis advances from the descriptive to the explanatory phase.

Based on our examination of Table 1, we decided to use as dependent variables the three categories of people mentioned most often as providers of social support for the sample as a whole.Footnote 4 We therefore created three categorical dependent variables and applied a multinomial logit regression to each to explain the determinants of choosing one’s partner, one’s mother, or a close friend in the case of feeling depressed; one’s partner, mother, or offspring in case of the flu; and one’s partner, parents, or siblings in the case of needing money. For this last variable, we decided to combine the answers in which the respondent declared he/she would turn to the father or mother first to ask for money, since the figures in almost all of the countries were very similar for each of the parents and we assume that the income for older couples can in many cases not be clearly differentiated, such that the respondent does not make a sharp distinction between the two progenitors in requesting this aid, in contrast to the other two kinds of support.

3.2 Independent Variables

The independent variables chosen to perform the analysis are divided into three kinds: sociodemographic; network-related, that is, related to the kind of personal network of the person surveyed; and contextual, or referring to the respondent’s country of residence.

3.2.1 Sociodemographic Variables

We have included six variables: (1) sex (giving men the value 1 and women the value 0); (2) age (classified as 18–24 years, 25–34, 35–44, 45–54, 55–64 and 65 or over, taking as reference value people 35–44 years old)Footnote 5; (3) education (classified as: primary education/secondary education/university study, taking as a reference secondary educationFootnote 6); (4) time of residence (with three categories: the respondent has lived in the place for less than five years, has lived in the place for more than five years, or was born in the place, taking birthplace as the reference categoryFootnote 7); (5) religion, recodifying the original variable into the categories nonobservant (does not practice a religion), infrequent observance (attends several times a year), medium observance (attends several times a month) and high observance (at least once a week), taking as the reference category several times a year; (6), finally, we have habitat, with two categories, rural and urban, taking urban as the reference.Footnote 8

3.2.2 Variables Related to the Personal Network

This second group includes four variables, three related to family and one to friends: the frequency with which one sees or visits the sibling with whom one has the most contact,Footnote 9 the frequency with which, one sees or visits one’s father and mother, and size of network of friends. As Kalmijn (2006) indicates, geographical proximity and face-to-face contact are an important condition for the development of what has been termed “family solidarity” (Silverstein and Bengtson 1997), defined as the degree to which the members of the family care mutually for each other. It is argued that support is more likely when the geographical and social distances are smaller. Face-to-face contact is also a good indirect measure of intergenerational support because it includes many forms of instrumental support that are too idiosyncratic to measure (Kalmijn 2006). Distance can limit aid that can be given in the home (care for the sick, preparation of meals, shopping) but affects emotional and economic support less (Litwak and Kulis 1987; Hogan et al. 1993).

The categories of the three variables, treated as continuous, for family visits are: live in the same household/visit daily/visit several times a week/once a week/several times a month/several times a year/fewer than several times a year. In the case of the father and mother, we also include “never” and “not living.” We give “never” the value 0 and exclude the response “not living.” When respondents state that they do not visit a sibling or do not have one, we give the response the value 0. We constructed the last variable, size of network of friends, by adding the responses to the three questions: number of close friends the respondent has in the workplace, close friends living nearby, and other close friends outside the workplace, neighborhood, and relatives.Footnote 10

3.2.3 Contextual Variable

This variable includes the individual’s country of residence. To present these countries, we have classified them according to the welfare regime to which they belong, following the classification by Esping-Andersen (1990, 1999), except that we have kept the countries in southern Europe as a group separate from the conservative continental European regime. Here we follow arguments by many scholars who have found significant differences between the two groups of countries in diverse indicators related to social well-being (Ferrera 1996; Leibfried 1992; Antonnen and Sipilä 1996; Bonoli 1997; Requena 2010, 2013). We did not want to group the countries according to welfare regimes because we wished to confirm whether there were significant differences within each of these regimes or whether all of their members follow similar patterns.

According to this classification, Great Britain, the United States, Canada, and New Zealand belong to the liberal regime; the conservative regime is composed of Austria, France, and the two Germanies; the social democratic regime includes Norway, Denmark, and Finland; finally, Italy and Spain form the Mediterranean welfare regime. East Germany is a special case, as there is no consensus on where to situate it: Scheepers et al. (2002) decide to include it with the social democratic regime, since their research analyzes individuals over the age of 60 who were socialized during the Communist period, whereas Kääraiäinen and Lehtonen (2006) prefer to include it in a separate model composed of various countries in Eastern Europe. Since this paper has not analyzed other Eastern European countries, we preferred to include East Germany in one of the four existing models and decided to place it in the conservative model for two reasons: first, the state policies that regulate the family and the labor market are homogeneous throughout Germany, which had been unified for over a decade at the time of this study (2001); and, second, as will be shown in the following section, the frequencies of the dependent variables were very similar in both Germanies, although there are some differences that recommended keeping them separate for independent study.

4 Analysis and Results

As mentioned above, the likelihood of turning to one person or another from one’s personal network is influenced by factors in the composition of the network. The three kinds of support studied in this research are economic, domestic, and emotional. The three dependent variables refer to turning to one’s partner, mother, or offspring (for domestic help); partner, parents, or siblings (for economic help); and partner, mother, or close friend (for emotional help). We must therefore ask whether the differences found between the countries in Table 1 are due to the differing availability of these people (father, mother, and close friend) among the respondents who are members of a couple, or to other kinds of sociodemographic variables. Table 2 provides an initial approach to this question, showing the individuals to whom people in couples turn when both parents, or the mother and at least one close friend are alive (n = 5,551). The study of the sociodemographic variables will be performed in a later analysis.

Table 2 Distribution of frequencies by country of whom one would ask for help first among interviewees with a partner, both parents or the mother living, and close friendsa
Full size table

The results in the table show, as we expected, that asking for help is conditioned by the composition of the network. In choosing a sample in which, in addition to the partner, at least the mother and one close friend are available, the percentages of the person from whom one would ask for help first vary, especially in the cases of the partner and mother, although the spouse continues to be the main provider of support. Comparing the minimum and maximum in Tables 1 and 2 for each kind of aid, we see that the minimum values for the partner drop, especially in the case of domestic help, whereas the maximum values remain at similar levels. In contrast, the minimum and maximum figures for both parents are higher in Table 2. The mother has special importance in giving economic aid (mentioned on twice the number of occasions in the case of Canada in the smaller sample), followed by help in case of sickness (especially in Italy and Spain), and to a lesser extent comfort when one is depressed. The father is also mentioned more often as provider of money (especially in Austria, where the figure doubles) but does not increase in significance for the other two kinds of aid, such that his role is limited to instrumental, economic support. Finally, although there are no large differences in the other kinds of social support, close friends also have higher values, both minimums and maximums, in the case of emotional help, which increases in importance to a greater extent in France.

The analysis by country shows some interesting variations that affect primarily the mother as provider of social support. For economic aid, the number of countries increases in which one turns to the parents first, before one’s partner. To the United States, East Germany, and France, we add three new countries: East Germany, Denmark, and Spain. Further, in Italy, both the mother and the father are mentioned more often than one’s partner as the first person to whom one would turn for economic support. The classification of countries according to whether they score above or below the average in seeking support from their partners does not vary much, except that Spain ceases to be below the average in all kinds of help and Italy is the only country below the average in all three cases. Italy is also the only country in which close friends score higher than the partner as the main providers of emotional support (37.7 % as opposed to 34.3 %).

As to the second most-mentioned person to whom people turn, the mother becomes very important as the person from whom one requests money and is mentioned more than the father in Canada, Austria, France, Finland, and Spain, as well as the cases of Great Britain and East Germany mentioned in the analysis of Table 1. On the other side are Italy (mentioned above) and East Germany, where individuals turn more often to the father than to the mother. The case of being sick shows the greatest variations with respect to Table 1, as the mother acquires more importance in all cases. She is the second most-mentioned person, not only in the Mediterranean countries and the United States but also in Austria, the two Germanies, France, and Great Britain. Canada joins the countries with similar figures for both family members (Great Britain and France). The case of Italy stands out again, as it is the country where the mother comes closest to the partner as the main source of domestic help, although behind the spouse (38.1 % as opposed to 43.7 %). Finally, in all of the countries, friends are mentioned more often as a source of emotional support than in Table 1 although the increases are not as significant as in the case of the father and the mother. The only statistic that stands out, as mentioned above, is that of Italy, where friends replace the partner as the main providers of support.

Although we restricted the sample to make its network composition more homogeneous (taking all respondents surveyed for whom a partner, parents, or at least the mother and a close friend were available), we continue to find significant differences between the countries. This leads us to perform a multinomial logit regression analysis for each of the dependent variables, including each country as a dummy variable.Footnote 11 In this way, we aim, while controlling for individual variables, to investigate the countries in which individuals turn to a greater or lesser extent to their partner, parents, children, siblings and friends in seeking help, taking Spain as a reference. We preferred to include the countries independently instead of in groups by welfare regimes in order to confirm whether there are relevant differences between countries that in theory belong to the same welfare regime.

Tables 3, 4, and 5 show the results of these regressions. In many cases, the same variables are significant in predicting to whom one would turn, but the coefficients points in the opposite direction depending on the person in question. To achieve greater clarity in the explanation, we will analyze each kind of aid separately, studying first the sociodemographic variables, second those related to the personal network, and finally, the national contexts.

Table 3 Marginal effects of the probability of asking for domestic help among interviewees with a partner, both parents, or the mother living, and with close friends: multinomial logit model
Full size table
Table 4 Marginal effects of the probability of asking for economic help among interviewees with a partner, both parents, or the mother living, and with close friends: multinomial logit model
Full size table
Table 5 Marginal effects of the probability of asking for emotional help among interviewees with a partner, both parents, or the mother living, and with close friends: multinomial logit model
Full size table

4.1 Domestic Help

The results show (Table 3) that men turn more often to their partner when they are sick and women to their mothers and children. People aged 45–64 turn more to their partners and offspring and less to their mothers than the reference category (35–44 years old), while people from 25 to 34 years of age turn less to their offspring. This pattern may be due, on the one hand, to the fact that the parents of the reference category are still young and in good health, in contrast to the older people surveyed, and, on the other hand, to the relatively young age of the children of people 35–44 years old. People with university degrees more often ask for help from their mothers first when they are sick. They request this help less often from their children and do not differ significantly from the other respondents in the case of the partner, while people with primary education turn more to their offspring. Another significant variable is the time people have been living in a place: people who have been living in a place for over 5 years are more likely to turn to their children. The last relevant sociodemographic variable is religious practice: nonobservers turn more often to their mothers when they are sick and less to their partners. Since religion is controlled by age, this pattern cannot be explained by arguing that the younger respondents observe their religion less. We thus need further research to clarify the reasons for this phenomenon. As to the variables related to personal support networks, the data show that respondents who visit their parents more often turn least to their partners to ask for care and more to the mother.

Finally, we analyze national contexts. Taking Spain as reference country, the first figure to point out is that all countries except Italy are significantly different from Spain in the case of turning to the partner when one is sick. In all countries, individuals turn more to their partner to request domestic help. As to requesting help from the mother, 11 of the 12 countries differ from Spain, and all in the direction of requesting less help. The exception is Italy, where individuals ask for more help from the mother. For example, New Zealand’s inhabitants have 14 % less probability of requesting help from their mothers than Spaniards, and 17 % more of turning to their partners, while Norway and Denmark have 13 % less and 16 % more, respectively. The only country that does not differ from Spain is West Germany. Finally, all countries except East Germany and Italy differ from Spain in turning less to their children to request domestic aid. We thus confirm the strong family solidarity between parents and children in the Mediterranean model, which remains constant even when we control for geographical proximity (traditionally in Southern Europe, people live closer to their parents) through variables such as time of residence or frequency of visits to family members.

4.2 Economic Help

As to economic aid (Table 4), women turn more often to their partners in search of economic support (a pattern probably related to their lower presence and greater precariousness in the labor market), and men more often request to borrow money from parents and siblings. Age is a significant variable, repeating the pattern that the younger respondents (18–34 years old) ask for more money from their parents and less from their partners than do the oldest respondents (45 and over), who show the opposite behavior, due possibly to the economic situation of their parents, most of whom are no longer working and are thus less secure economically than the parents of young respondents. Education is not a significant variable, except for individuals with university degrees, who turn less to their siblings when they need money.

The data show, again, the importance of face-to-face contact as a significant condition for family solidarity, even in the case of economic support, which need not require physical proximity: the more often people visit their parents (especially the mother), the more often they turn to their parents and less to their partners to request money. Contact with siblings (at least the one with whom one has the closest relationship—the question asked in the survey) is relevant in requesting money: those who visit more their siblings ask them more frequently for money and ask their parents less. The second pattern may be due to two reasons: that one receives economic support from one’s siblings, or that having siblings means that parents have fewer resources available to lend, since they have to distribute their resources among all of their offspring. Last, having more friends (elective relationships) increases the probability of asking one’s partner (also an elective relationship) and not one’s siblings (not elective) for help.

As to national contexts, economic aid shows fewer divergences among countries than does domestic help when people turn to their partners, as only seven countries diverge significantly from Spain. This shows that economic solidarity between spouses is extensive throughout Western countries.Footnote 12 In Austria, Norway, Finland, and Great Britain, people turn more to their partners than in Spain. The maximum difference occurs between Spain and Finland: The Finnish have 18 % more probability of requesting help from their partners than Spaniards. At the other extreme, the United States, East Germany, and Italy request such aid less. On the other hand, economic support from parents is significantly lower in Finland and Austria and higher in the United States, France, Italy, and East Germany. Lastly, few countries differ from Spain in asking for economic help from siblings, but individuals in all of these countries request such aid less frequently (Austria, France, Norway, and Denmark), and only one asks for more help than Spain—Italy.

4.3 Emotional Support

In this case (Table 5), men turn more to their spouses and women to their close friends and mothers. These results are consistent with the literature on social support. For example, Miller and Ingham (1976) and McFarlane et al. (1983) find that women have more confidants, whereas Lowenthal and Haven (1968) confirm that husbands cite their wives much more often as confidants than wives cite their husbands. Young people turn less to their partner and more to their mothers to talk when they are depressed, perhaps because their love relationships are not as stable at these ages and they place less trust in their partner. Further, the relationship itself may be causing the person to be depressed. On the other hand, after age 55, people turn more to their partners and less to their friends to talk if they are depressed. For example, people over 65 have 35 % less probability of talking with a friend than people 35–44 years old. This result agrees with the literature on friendship relations, which finds that after age 45 people turn less to friends (De Federico 2011). The data from the study by Scheepers et al. (2002) on social contacts with family and friends in different European societies point in the same direction, showing that, among individuals over age 60, members of a couple have more family ties and fewer ties with friends.

People with university degrees talk less to their mothers when they are depressed. In addition, the time one has been living in a place is significant. The people who turn more to their partners and less to their friends when they are sad are those who have been living less than five years in the place. It is likely that these people’s support networks in a close radius are smaller because family members are not nearby and because the individuals have not yet created close friendship ties. Religion is the last sociodemographic variable that shows relevance: those who observe their religion often turn more to their partner, and nonobservers turn less to their partners and more to their friends. Taking into account that this variable is controlled by age and education, it seems that believing in a religious doctrine (primarily Christian in the countries analyzed) encourages one to find consolation and support in one’s spouse.

As to personal network, the frequency with which one visits one’s mother, a factor that influences asking one’s partner for economic and domestic support, also plays a role in the case of emotional support. The more contact one has with one’s mother, the less likely one is to seek aid from one’s partner and the more likely one is to turn to the mother and to close friends. Finally, the size of the network of friends does not influence the decision to turn to a friend when one is depressed. Therefore, the data do not show that greater availability (a greater number of friends) necessarily means that one seeks help from them first. Rather, number of friends shows a positive association with turning more to one’s partner and less to one’s mother when one is sad, confirming the second hypothesis: the support requested in elective relationships (partner and friend) shows similar patterns for the variables that promote or inhibit their existence.

National contexts show greater homogeneity here, but four countries differ sharply from Spain. In no country do individuals turn more to the partner than in Spain, and four countries turn less (United States, Canada, France, and Italy). Italy shows the most differences from Spain: Italians have a 26 % lower probability of turning to their partners than do Spaniards. On the other hand, the same four countries (in addition to West Germany) have more probability of asking for emotional help from friends, and Italy is again the country whose result is farthest from that of Spain. It is worth pointing out that no country requests less emotional support from friends than Spain, confirming previous studies that indicate that having many friends does not mean that one turns to them in seeking this or any other kind of support, but rather that friendships have a more expressive and fun-oriented role that in other places, a characteristic Spain shares with Brazil and that differs from Italy, where individuals do turn to friends to request emotional support (De Federico 2011). Lastly, five countries seek less emotional support from their mothers: the United States and Canada differ from Spain, as in the case of the partner and close friend, but so do Norway, Denmark, and Finland.

After analyzing the results of the multinomial logit regressions, we can conclude that all of the variables included in the model are significant, except size of habitat. Although it is common in the literature to find studies in which people who live in rural towns, whose environment is more traditional, have stronger ties with family and thus find more support in it, this study has not found a significant relationship between type of habitat and the person to whom one turns in search of help. It is possible that having included other variables in the regression, such as contact with family members, has diluted this association between rural environment and family support, which may be due principally to the availability of family members due to geographical proximity. We must also take into account that the degree of urbanization has different effects on personal networks depending on the country (De Federico 2011), and this may cause its significance to decrease.

On the other hand, the variables related to the characteristics of the network have become relevant, even when these are largely homogeneous personal networks (since all survey respondents have a spouse, close friends, and the mother or two parents living). We have shown the importance of face-to-fact contact with family members in asking for help, a result that agrees with previous studies (Kalmijn 2006).

Finally, including countries in the model was a good choice, since this variable turns out to be very significant in most cases. That is, if we take into account the usual sociodemographic variables in social research, the composition of the support networks and their properties, there continue to be significant differences between countries according to their degree of defamiliarization. The minimal support asked from partners in Italy in all three types of help analyzed here, and in domestic help for the case of Spain as compared to the other countries, may seem surprising at first glance, since both countries have traditionally been classified as familialist, but we must complement these data with the data on support from parents and other relatives. First, in no other country does one turn more to the mother if one needs care or feels depressed than in Italy and Spain, confirming that both states have a strong family component, not of the elective family (the partner) but of the blood family (parents). Second, Italy and Spain stand out among Western countries because individuals ask for more economic help from siblings and for more domestic help from offspring. Nevertheless, Italy differs from Spain in the weaker role of the spouse as provider of social support: the partner is asked less in the case of economic or emotional help.Footnote 13 Additionally, the role of close friends is more important in Italy than in Spain, as mentioned above. The only country that differs from Spain in the three variables related to support from one’s spouse is the United States, where the partner is asked less in the case of emotional or economic help but provides more support in the case of domestic help. Finally, Finland is the only country where people ask for less help from mother/parents in all three types of support. This phenomenon confirms the distance between social-democratic countries, which are more defamiliarized, and Mediterranean ones.

We should mention that the two Germanies coincide in not differing from Spain in the case of requesting emotional help from the partner and mother, but only West Germany resembles Spain in requesting aid from the mother when one is sick or in the three categories analyzed in economic help. An interesting difference between the two Germanies is found in the role of friends. Only West Germans turn more to friends than do Spaniards, whereas East Germans show similar patterns to those of Spaniards. Previous studies show that the countries that have had a totalitarian political regime usually have smaller friendship networks and ask for less help from their friends (De Federico 2011; Völker 1995). Since the ISSP asks for close friends and such relationships are usually established over an extended period of time, it may be that the political past of East Germany influences the differences observed with West Germany.

5 Conclusions and Discussion

This study makes three new contributions to the study of social support. First, it focuses on people in couples, since the literature on this topic reveals that the partner is the main source of social support. Second, it analyzes whether there are international differences in the role of the partner (as well as other relatives) as provider of support. Finally, it applies the classification of welfare regimes to situations of everyday need to confirm whether the kind of aid sought differs based on the degree of defamiliarization in the country in which the respondent lives.

The current study analyzes the importance of informal social support to people in couples and the factors that influence the group of highly developed Western countries. We have confirmed (Hypothesis 1) that the main source of social support (economic, emotional, and domestic) is the family—first, the partner and, second, the closest relatives (parents, children, and siblings). This main source is the same in all of the countries analyzed, independently of whether the country’s welfare state is highly developed as provider of goods and services (social democratic) or its market provides these services to a greater or lesser extent (liberal). In all of these countries, the personal network plays a fundamental role as provider of social support. The data do not, therefore, confirm the erosion of social ties augured by some researchers (Beck 1992).

We have found that emotional and domestic support are provided fundamentally by women to their husbands, whereas women turn to friends and mothers. This result confirms previous research, developed mainly from the area of social psychology, which finds that women have more close friends than men and stress intimacy and speaking with these friends, whereas men prefer doing activities with friends, such as sports (Parker and de Vries 1993). Based on a review of diverse studies on social support, Cutrona (1996) concludes that the more consistent difference between men and women has been found when they face a problem. Women tend to seek more social support in people other than their partners, leading to what Belle (1982) describes as a support gap in the relationship.

One contribution of this paper is to confirm that variables have different influence on social support depending on whether one turns to an elective relationship (partner or close friend) or to a “compulsory” relation of blood ties. We assumed that people with a higher degree of individualism—measured through their higher education level, fewer visits to their families, and having a larger network of friends—would turn much more to their partner to request economic, as well as emotional support (Hypothesis 2). Nevertheless, results are mixed. On the one hand, education is not relevant in predicting the probability of asking the partner for help, and people with more education turn to their mothers less for emotional help but more when they are sick. On the other hand, as expected, individuals who maintain more contact with their parents (especially the mother) tend to turn to their parents to a greater extent to receive any type of help, and it is more probable that one will ask for economic help from siblings when one has more contact with them. Many studies find an association between frequency of face-to-face contact and request for support in domestic tasks but indicate that contact has less influence on economic support (Martín-Lagos 2011; Litwak and Kulis 1987; Hogan et al. 1993). The data obtained show, however, that greater contact with parents (and siblings) increases the likelihood of asking them for money. It would have been desirable to incorporate the postmaterialism index in this analysis to confirm the second hypothesis, but unfortunately the ISSP does not include this item. We advise confirming this hypothesis in future research. Lastly, young people turn more to their parents than to their friends or partners, perhaps because the relationship with the partner is not yet stable and the degree of intimacy and trust is lower than with their parents and close friends.

Another contribution of this study has been to confirm that informal social support varies according to national contexts. We performed a multinomial logit regression analysis, which shows that the countries differ significantly as to whom one asks for help, controlling for sociodemographic variables and those of the personal network. Taking Spain as reference, we see that, in all of the other countries except Italy, people turn more to their partner in case of illness, and in no other country does one turn more to the mother if one needs care or if feels depressed than in Italy and Spain, confirming that both states have a strong family component, not the elective family (the partner) but the blood family (parents). Italy and Spain also differ from the other countries analyzed in that individuals request more economic help from siblings and more domestic help from offspring. In contrast, we find fewer international differences in emotional help given by the spouse. Only four countries diverge from Spain, all turning less to the spouse (United States, Canada, France, and Italy). On the other hand, Italy and Spain are characterized by turning much more than the others to parents, children and siblings, although they differ in the emotional support found in friends, which is greater in Italy than in Spain.

Hypothesis 3 assumes that the citizens of more defamiliarized countries turn more to the partner and friends and less to parents. Following Esping-Andersen’s classification, at the defamiliarized extreme, we find social democrats, followed by the liberals. The conservative countries are more familialist than these two groups, and at the extreme of familialism are the Mediterranean countries. The results of the regressions confirm the hypothesis, although with some qualifications: at one extreme, the most familialist countries (Spain and Italy) turn more to close relatives (mother, siblings, and offspring), in accordance with our expectations. The conservative countries differ significantly from the Mediterranean countries in domestic help from partners (in all cases), from the mother (except in West Germany), and from offspring (except East Germany). Some of these countries do not differ from the Mediterranean countries, however, in the economic support granted by parents, spouse, and siblings. The data thus confirm that, although the conservative countries are those most similar to the Mediterranean ones, we are justified in considering Italy and Spain as a model different from that of continental Europe.

Social democratic countries, which according to the classification of regimes would be at the opposite extreme from Mediterranean countries, always differ from Mediterranean in the case of domestic support from the partner, the mother, and offspring. In matters of emotional support, individuals in social-democratic countries ask for less help from their mothers. In the case of economic support, Norway and Finland ask for more help from their partners, but only Finland requests less support from parents. Liberal countries show patterns very similar to those of social democratic countries, with the United States as the country most distant from Spain.

We have seen that the Mediterranean countries differ from the rest in turning to parents, offspring, and siblings much more when seeking support. This pattern remains even when controlling for the degree to which we maintain face-to-face contact with family members. Thus, it is not a matter—or at least not only a matter—of people in the Southern countries having parents who are more available because they live closer and see each other more often, but of other factors that influence greater familism in these countries. One possible explanation would be that values of family solidarity are more deeply rooted in these countries than in the others. As Esping-Andersen (1999) indicates, these countries have a legal prescription that parents (or children) are responsible for their children (or their parents) in case of need. However, this legal norm is also present in Austria and Germany, which have shown significant differences from the Mediterranean countries regarding their familism. Future research should explore this issue further to confirm what characterizes this Mediterranean familism.

The analysis confirms the appropriateness of taking into account national contexts when studying individuals’ social support networks. Without denying the importance of the individual factors traditionally included in this kind of analysis, there is no doubt that a better understanding of this phenomenon should take into account macrostructural variables that can influence them, such as cultural traditions or welfare regimes. The analysis also shows that Spaniards and Italians have on occasion different sociability patterns: Italians turn much less to their partners and much more to their friends when they are sad, but there are no statistically significant differences between these countries in support sought from the partner when one is sick or the mother when one is sad. Precisely this result shows the need to perform future research, analyzing the countries separately and not only grouped by welfare regimes, since we often find “anomalies” that deviate from the theoretical models and significant differences between countries that belong to the same regime.

This study has a series of limitations due to the source of the data used. First, it does not take into account the connectivity of the respondent’s network, a variable whose importance for studying marital roles has been demonstrated (Bott 1957), due to lack of data on this topic in the survey. Second, it does not consider whether the respondent and his/her spouse are employed and, if this is the case, the nature of their work. This variable is important to determine the degree of availability of a person to lend support (especially in case of sickness). One approximation of occupation, although an imperfect one, is education level, since higher levels of education are associated with greater qualification and dedication to work and there is a tendency to enter into a relationship with people of a similar educational level. Due to lack of data, the study also does not take into account the age of the parents or their state of health, but both factors are related to their availability to give help. Nevertheless, our inclusion of the respondents’ ages in the analysis helps, at least partially, to estimate their parents’ age. It would be interesting for future research to incorporate all of these factors to confirm whether they have an influence, and in what way, in providing social support.