Keywords

Relatedness with others is a basic human need (Deci and Ryan 2002); as a result, social relationships play an important role in determining individuals’ subjective well-being. Among the various relationships in one’s social network, family relationships and friendships are usually the two most important sets of ties. Family relationships and friendships are both considered to be influential on subjective well-being (Adams and Blieszner 1995; Cheng et al. 2009; Fiori et al. 2006). However, as Asian countries are known for their collectivistic cultures and the emphasis on maintaining close and harmonious family relationships (Fuligni et al. 1999; Markus and Kitayama 1991; Yang 2006), it is important to investigate whether the effect of friendship on subjective well-being differs in the Asian context. This chapter first introduces previous theories about the different roles that family ties and friendship ties play in people’s lives. Then, empirical studies on the contributions of family relationship and friendship to subjective well-being are reviewed. A comparison is made between findings in Western samples and those in Asian samples. Finally, the implications for friendship ties in the context of weakening family structures in Asia as well as future research directions are discussed.

Family Relationship Versus Friendship

In this section, we review theories about how social network contributes to subjective well-being, particularly concerning distinctions and similarities between the roles played by family relationships and friendships. In this chapter, family relationships broadly refer to relationships brought by marriage (e.g., marital relationship, relationship with in-laws) or genetic relatedness (e.g., parent-child relationship, sibling relationship) unless specified otherwise. To describe the function of different sources of social support from one’s social network, three theoretical models have been proposed, namely, the hierarchical-compensatory model (Cantor 1979), the task-specific model (Dono et al. 1979; Litwak and Szelenyi 1969), and the functional specificity model (Simons 1983–1984).

According to the hierarchical-compensatory model (Cantor 1979), people have a hierarchy of preferred support providers. Across all support domains (e.g., instrumental, emotional, informational), individuals would first seek support from the most preferred source. Only when the most preferred source is not available would the next preferred source be utilized to compensate for the vacuum left by the more preferred support provider. Nuclear family members, especially the spouse, are usually the most preferred source of social support, followed by relatives in the extensive family and friends. Empirical studies have provided support to the hierarchical-compensatory model (e.g., Cantor 1979; Connidis and Davies 1990; Penning 1990). With a sample of 1552 older adults in New York, Cantor (1979) found that most people turned to kin, rather than to non-kin, for help in ten daily life situations ranging from instrumental ones (e.g., help with doctor’s visits, financial assistance) to affective ones (e.g., confiding). Penning (1990) replicated and further illustrated the finding in a Canadian sample. Focusing on five areas of assistance such as grocery shopping, emotional support, and help with emergencies, most supportive functions were found to be provided by spouses. When a spouse was not available, children became the major provider of social support. When both spouse and children were unavailable, other relatives took the role of major support provider. In all situations, non-kin relationships (i.e., friends and neighbors) were a minor source of support compared with kin relations.

However, although kin cover more support functions than non-kin, depending on the nature of the assistance required, the proportion of people seeking support from non-kin varies. For example, in Cantor’s (1979) classic study, 27.3 % of the respondents approached friends for help when they felt lonely and wanted to talk, but only 6.4 % turned to friends when they did not have money for a medical bill. Thus, the task-specific model proposes that because the nature of different tasks in daily life varies, people’s preference for support providers is task specific (Dono et al. 1979; Litwak and Szelenyi 1969; Messeri et al. 1993). Daily tasks can be differentiated on multiple dimensions, such as required proximity, duration of commitment, or shared life experience (Messeri et al. 1993). Family members are the most suitable support providers for tasks that need long-term commitment, such as chronic illness care, but friends may be the more preferred support providers for tasks that require shared life experiences or shared interests, such as discussing career-related concerns or playing tennis together.

Following a similar rationale, Simons (1983– 1984) proposed the functional specificity model of social relationships. Instead of being task specific, different relationship types are considered to serve distinctive functions. Security, intimacy and self-esteem are identified as the three basic desires to be fulfilled in social relationships. Different relationship types serve distinctive functions by satisfying different desires (Simons 1983–1984). In particular, whereas family relationships are more effective in satisfying the desire for security, friendships are more useful in satisfying the desire for self-esteem. Moreover, when a certain relationship is not available, its function can be substituted by other relationships. For unmarried individuals, frequent contact with friends could provide people with a sense of security, serving the same function as spouse and children do for married individuals (Carbery and Buhrmester 1998; Simons 1983 –1984).

Indeed, the three theoretical models of social relationships are not exclusive to each other (Messeri et al. 1993; Penning 1990). Family roles and relationships are ascribed and family members are obligated to support each other when in need. So, people feel more freely to seek help from family members across a range of circumstances and family members indeed provide most of the social support to people. In contrast, friendships are formed voluntarily by mutual agreement. No definite responsibilities are attached to the relationship. Support is usually provided based on the reciprocity principle. Hence, compared with family members, friends are less frequently sought for support in general. However, friends are usually at the same life stages and face similar challenges, and friendship is typically based on shared interests and shared experience; thus friendship can be particularly efficient when dealing with certain tasks (e.g., sharing about family problems) or fulfilling certain psychological needs (e.g., promoting self-esteem). Family and friends both have unique and indispensible roles in people’s lives.

Other theories have examined the role of social network from a life-span developmental perspective. The social convoy model argues that social networks are not static but are dynamic resources that change and provide support in response to changing circumstances over the life course. Relationship closeness is graphically represented by three concentric circles (i.e., the inner, middle, and outer circle) around the individual, with the inner circle being the closest (Kahn and Antonucci 1980). Across different stages of the individual’s life development and family life cycle, the position of each social partner dynamically moves in and out of the convoy circles while new members may be added. For example, old friendship may be rekindled to substitute for lost family ties as in widowhood. Usually, the nuclear family members, which change from parents and siblings in childhood and adolescence to spouse and children during adulthood, stay in the inner circle of the convoy. Friends enter and sometimes reach the inner circle of the social convoy during adolescence and young adulthood. But the importance of friendships drops later, especially after individuals start their own family (Levitt et al. 1993), so friends tend to stay in the middle or outer circle of the social convoy during later life stages. The general changing pattern of the social convoy across life span was found to be similar in the United States and Japan (Antonucci et al. 2004).

Similarly, socioemotional selectivity theory (Carstensen 2006; Carstensen et al. 1999) argues that as people perceive their future time as increasingly limited from early adulthood to later adulthood, their priority shifts from future-oriented goals to emotionally meaningful goals. Hence one’s social network is also managed to match with the prioritized goals at different stages of adulthood. To achieve emotionally meaningful goals in later adulthood, interaction frequency with peripheral social partners (i.e., those in the middle or outer circle of the social convoy, such as friends or acquaintances) declines, while that with close social partners (i.e., those in the inner circle of the social convoy, such as spouse or siblings) increases (Carstensen 1992; Lang 2001). Moreover, Heller et al. (1991) found that once the long-term close social ties were lost in later adulthood, they could not be replaced by establishing new friendship ties. Hence, both the social convoy model and the socioemotional selectivity theory suggest that the closeness of friendships peaks during adolescence and early adulthood, whereas the closeness of family relationships remains prominent across the life span.

Family, Friend, and Subjective Well-being in Western Culture

Many studies using Western samples have found that both family relationships and friendships are influential on subjective as well as physical well-being (Adams and Blieszner 1995; Antonucci et al. 2001; Dean et al. 1990; Dupertuis et al. 2001; Fiori et al. 2006; Larson et al. 1986; Reinhardt 1996). However, given the different nature of the two relationship types, there are several notable distinctions in their effects, which are discussed below.

First, more family interactions are not always related to better subjective well-being (Lee and Ellithorpe 1982; Lee and Ishii-Kuntz 1987; McCulloch 1990; Rook 2001, 2003). Family relationships are attached with both deep affections and strong obligations, so interaction with family members is a mixture of sweetness and bitterness. In a study of retired people in the U.S., Larson et al. (1986) found that married people spent nearly half of their time with their spouse and/or children, while the majority of the remaining time was spent alone. Not surprisingly, time spent with family would be much less for people who are working. According to a statistical report from Canada (Turcotte 2007), workers spent 206 min with their family on average on a typical working day in 2005. The time was less than it used to be in 1986 (250 min), but it was still far more than the time spent with friends, which was only 19 min per working day in 2005. As family members, especially the ones in the same household who spend so much time together, they go through nearly every life events jointly, just as the wedding vow says, “for better for worse, for richer for poorer, in sickness and in health”. Thus, family relationships are not all about fun. They are intertwined with repetitive routines (e.g., housework) and inevitable frustrating details of the reality (e.g., financial hardship). Family relationships can also be undermined by factors outside the household (Conger et al. 1990; Karney and Bradbury 1995). For example, when an individual has a stressful day at work, the negative impact of the stress can spill over to his/her family relationship. Negative social interactions at work or heavy workload could both lead to problematic marital interactions on the same day (Story and Repetti 2006). The effects of the mixed positive and negative interactions with family members may cancel each other out (Davey and Eggebeen 1998; Rook 2001, 2003), resulting in nonsignificant association observed between family interaction and subjective well-being (Lee and Ellithorpe 1982; Lee and Ishii-Kuntz 1987; McCulloch 1990).

Second, compared to family relationships, friendships can be more beneficial to subjective well-being (Antonucci 1990; Demir and Özdemir 2010; Larson et al. 1986; Nussbaum et al. 2000). Friendships are mainly based on mutual liking and shared interests. People spend much less time with friends than with family, and the nature of the interaction is less characterized by concomitant positive and negative elements in friendships than in family relationships (Larson et al. 1986; Turcotte 2007). Most of the interactions with friends happen during leisure activities, with the major purpose to relax and to enjoy. People can also choose whether to participate an event with friends following their own will. It is not surprising to find that people would experience more positive and more aroused affect when interacting with friends (Larson et al. 1986). Also, as friends mostly share the positive and recreational aspects of their lives, they have fewer conflicts of interest and can be better providers of emotional support to each other (Antonucci 1990; Antonucci and Akiyama 1995; Nussbaum et al. 2000). Another notable feature of friendship is that it is a voluntary relationship. When the friendship with someone stops being enjoyable and fruitful, one can choose to disengage from it. Through such a screening process, sustained friendships are usually characterized by mutual support and quality exchanges. For these reasons, friendships are particularly beneficial to subjective well-being. Demir and Özdemir (2010) further identified that the provisions of companionship, instrumental assistance, intimacy, reliable alliance, self-validation, and emotional security were six characteristics of good friendship. They also found that friendships contributed to happiness by meeting individuals’ three basic needs, namely, autonomy, competence, and relatedness (Deci and Ryan 2000). With the company of friends, people can feel free to be who they are, feel “capable and effective,” and feel “loved and cared about” (Demir and Özdemir 2010, p. 248).

Third, during difficult times, such as illness or financial hardship, family relationships are a more reliable source of instrumental support (Adams and Blieszner 1995; Antonucci 1990) and are expected to contribute more significantly to both physical and mental health. One reason is that family members share more aspects of their lives, and have more knowledge of each other’s difficulties. In most occasions, individuals would feel freer and more comfortable exposing their vulnerability to family members at home, than to friends during social gatherings. Moreover, the bonds between family members are involuntary. Unlike friendships, mutual support between family members is not only determined by personal preference, but also obligations to each other. It ensures the stability of the support, which is usually required during difficult times.

Testing the effects of different support sources on subjective well-being during hardship, Friedman (1993) specifically examined older women suffering from heart disease and their social support system. She found that patients who received support from family members reported more positive affect and higher life satisfaction than those who received support only from nonfamily sources. The same pattern was true for both emotional and instrumental support. One possible mechanism for the particularly beneficial effect of family relationship on subjective well-being during hard times is that people are more at ease and less physiologically aroused when they are with family than with other people (Holt-Lunstad et al. 2003; Spitzer et al. 1992). Such relaxed and calm states can be especially valuable when dealing with difficult or prolonged problems. Thus, support from the family, rather than friends, is more beneficial to subjective well-being during challenging times in life.

Lastly, to be a rewarding relationship that contributes to subjective well-being, the rule of reciprocity is of far greater importance in friendships than in family relationships. Clark and Mills (1979, 1993) proposed that social relationships can be categorized into two groups, namely, exchange relationships and communal relationships. Exchange relationships are characterized by an emphasis on the equity rule, whereas communal relationships focus on meeting each other’s needs. To be specific, friendships are mostly exchange relationships and function according to the reciprocity norm. The support exchanges between friends are mutual and relatively comparable in value, although the “payback” may not be immediately binding and may not take the same form. For example, it is possible that in a friendship, one person provides more materialistic support (e.g., paying bills for shared dinners) while the other person provides more emotional support (e.g., being a good confidant). In contrast, family relationships are the typical example of communal relationships. Family members can provide support to each other solely because it is needed. From the evolutionary perspective, such altruistic behaviors between family members are adaptive because of the biological connections between these individuals (Hamilton 1964); to protect one’s family members ensures continuation of the clan through survival. As a consequence of such differences between friendship and family relationship, the level of reciprocity was found to be higher in friendships than in family relationships. More importantly, reciprocity in friendships was more significantly related to satisfaction than that in family relationships (Rook 1987).

To conclude, family relationship and friendship are both indispensible components of people’s social network that contribute significantly to subjective well-being, but their specific roles are different. Family relationships are essential to most individuals and provide them with most support. The connections between family members are so close that both positive and negative experiences are shared. The mixed interaction experience makes the connection between family relationship and subjective well-being not apparent in general. However, when it comes to really difficult times in life, family members’ unconditional support is particularly important to maintain individuals’ well-being. By comparison, friendship is relatively peripheral in the social network. The lives of friends are overlapped to a lesser extent in comparison with those between family members, and the interactions in most friendships follow the rule of reciprocity. A major characteristic of friendship is that the shared experience between friends is dominated by recreational and enjoyable activities, so friendship is a salient contributor to subjective well-being in everyday life. Another notable point is that when family relationship is not available, close friendships can function as fictive kin relationships and provide people with the needed support (Connidis and Davies 1990; Taylor et al. 2001). In other words, family relationship and friendship jointly form a dynamic support system that protects individuals’ subjective well-being.

However, all the studies reviewed so far are based on samples from individualistic cultures. The effects of family relationship and friendship on subjective well-being may be different in the collectivistic context. The next section specifically introduces recent findings about how family and friends contribute to subjective well-being in the Asian context.

Family, Friend, and Subjective Well-being Among Asians

Asian countries are dominated by the collectivistic culture and a major characteristic of collectivistic culture is the emphasis on in-group relationships (Markus and Kitayama 1991). Individuals exposed to collectivistic culture have an interdependent self-construal, which incorporate close relationships as part of their self-concept. Their subjective well-being is not only influenced by satisfaction with self-related attributes, but also affected by satisfaction with relationship-related attributes. Specifically, Kwan et al. (1997) identified relationship harmony as a more important determinant of life satisfaction in Hong Kong than in the United States. Tam et al. (2012) also found that when bicultural individuals were primed with collectivistic stimuli, their life satisfaction and happiness were more strongly associated with satisfaction with close relationships, compared with when they were primed with individualistic stimuli. In other words, people from individualistic cultures draw a clear distinction between self and others, whereas people from collectivistic cultures emphasize more on the distinction between in-group and out-group. People with collectivistic beliefs were found to allocate more resources to in-group members and evaluate them more leniently (Gómez et al. 2000; Hui et al. 1991). Likewise, trust in collectivistic societies is mainly built on group affiliation and long-term collaboration, so that generalized trust toward out-group people is negatively associated with the collectivism level of a society (Allik and Realo 2004; Gheorghiu et al. 2009; Yamagishi et al. 1998; Yamagishi and Yamagishi 1994). All these empirical findings suggest that collectivistic people are particularly dependent on in-group relationships, and the family is a prototype of such relationships.

The family is the most central and intimate group for most Asian people. Family relationships are considered to be superior to other types of relationships because “blood is thicker than water.” Individuals believe that the welfare of family members, harmony of family relationships, and prosperity of the whole family are their principal concerns. One could sacrifice one’s own interest for the sake of other family members. For example, adolescents with Filipino or Chinese background were found to endorse more family obligations, and provide more support and pay more respect to their family members compared with their European counterparts (Fuligni et al. 1999). Such cognitive, emotional, and behavioral tendency favoring family members, family relationships, and family prosperity, as well as the emphasis on mutual family obligations, is termed familism (Yang 2006). Thus, family relationships should play a more critical role in one’s social network and contribute more to subjective well-being in Asia than in the West.

Similar to findings in the Western culture, the family is a major and diffuse support provider for Asian people, while the role of friends is mainly to provide emotional support. Lee et al. (2005) surveyed people in Beijing and Hong Kong about whom they would first turn to in five different situations (i.e., need help with housework, need help during sickness, need to borrow a large amount of money, need important advice, and need a confidant when depressed). Across all situations, close family members were the most preferred source of support. The percentage of people who chose to seek support from close family members was highest when needing help during sickness (77.7 % in Hong Kong and 85.2 % in Beijing), and lowest when needing a confidant when depressed (43.3 % in Hong Kong and 45.4 % in Beijing). Meanwhile, friends were turned to mostly when emotional support was needed. When feeling depressed, 31.5 % of the Hong Kong sample and 26.3 % of the Beijing sample chose to talk with a close friend. Another similarity between the West and Asia is that family relationships are particularly beneficial to subjective well-being during difficult times in life. In a longitudinal study with older adults in Hong Kong, Chou and Chi (2003) found that for older adults who suffered from depression, the support they received from family members increased after 3 years whereas the support from friends decreased over the same period. Support from family members also resulted in the decrease of depression longitudinally.

However, unlike Western cultures, family relationships contribute more significantly to subjective well-being than do friendships even during ordinary days in the Asian context. Cheng et al. (2011) investigated how social exchanges with different types of kin and non-kin influence older Chinese adults’ subjective well-being, indicated by the level of life satisfaction, depression, positive affect, and negative affect. Based on social contact and exchange patterns, they examined two subgroups within kinship: (a) the vertically extended family, including parents, spouse, children, children-in-law, and grandchildren, and (b) the horizontally extended family, including siblings, cousins, distant in-laws, and so on. They found that positive exchanges (averaged by number of network members within certain relationship categories) with both vertical and horizontal family members significantly contributed to older Chinese adults’ subjective well-being, while negative exchanges with family members significantly impaired subjective well-being. However, although social exchanges with friends had a few significant correlations with the subjective well-being indices, the relationships disappeared after controlling for exchanges with family members. In other words, the quality of social exchanges with friends did not contribute to well-being beyond the effects of exchanges with family. Cheng et al. (2009, 2011) argued that due to the stronger reciprocity norm in collectivistic cultures, one may be discouraged from seeking help from friends because of the obligation to return favors. The harmony norm also discourage one from sharing personal problems in certain contexts as doing so may be seen as disrupting social harmony. Thus, the beneficial effects of social exchanges with friends may be limited by these factors in the Asian context.

In addition to investigating separate relationships, Cheng et al. (2009) also examined older Chinese people’s social network types and how social network structure affected subjective well-being. Using a sample of more than one thousand older Chinese adults in Hong Kong, five types of social network were identified: diverse, friend-focused, restricted, family-focused, and distant family. Similar to Western findings (Fiori et al. 2006; Litwin 2001; Litwin and Shiovitz-Ezra 2011), the diverse network (characterized by high levels of contact and exchanges with kin and nonkin) was associated with the best well-being indicators, including being high in morale and life satisfaction and low in depression. Different from Western findings, however, those having family-focused networks were not disadvantaged when compared to those with friend-focused networks in Chinese older adults. Their well-being indicators were similar, or even marginally better among those with family-focused networks (Cheng et al. 2009). Another study of Japanese older adults found that social network types, whether diverse, friend-focused, family-focused, or restricted, were not differentially associated with depression (Fiori et al. 2008). Although the findings for the Japanese were somewhat different from those for the Hong Kong Chinese, the lack of superiority of friend- over family-focused networks appears to be consistently found across Asian samples. Fiori and colleagues also found that U.S. older adults tended to rate their family relationships more negatively than friendships, whereas the same was not true for the Japanese, which might explain the lack of differences between friend- and family-focused networks in terms of associations with well-being indicators among Asian older adults. Another reason may be that many leisure and hence pleasurable activities are conducted with family members as well as friends in Asian societies.

Other studies using Asian samples also supported that family relationships are generally more influential on subjective well-being than friendships. Siu and Phillips (2002) studied how the quality and subjective importance placed on family relationships and friendships affect older women’s subjective well-being in Hong Kong. Quality of family relationships, but not friendships, was significantly related to more positive affect and less negative affect in the sample. The perceived importance of friendships was also significantly related to more positive affect. In another study involving a larger Hong Kong sample with both genders, Phillips et al. (2008) found again that the frequency and quality of family interactions were significantly related to older adults’ subjective well-being, but frequency and quality of friendships did not make a difference. Yeung and Fung (2007) also found that older Chinese adults’ life satisfaction was significantly associated with emotional and instrumental support from family member, but such associations were not found between life satisfaction and support from friends. Moreover, familism significantly moderated the relationship between family instrumental support and life satisfaction. Older adults who valued more about familism benefitted more strongly from family instrumental support.

Does this mean friendships are not important for Chinese older adults? Although social exchanges with friends were not associated with well-being, Cheng et al. (2011) found that the size of the non-kin network (i.e., the number of friends) was significantly associated with all well-being indicators, with a larger non-kin network size being associated with more positive affect, less negative affect and depression, and, to a lesser extent, higher life satisfaction. The size of the family network, whether vertically or horizontally extended family, were only associated with selected indicators and the effect sizes were much smaller, compared with those of the non-kin network. Social exchanges with kin and non-kin were controlled for in these analyses. Together, the findings suggest that the quality of the daily exchanges with family members is important for the subjective well-being of Chinese older adults. While friends may not be seen on a regular basis, a larger friendship network may be key to more diverse and frequent social activities, which are typically emotionally arousing (hence stronger associations with affective indicators of well-being). Also in support of the potential benefits of friendship was the study by Zhang et al. (2011). The authors found that for middle-aged and older Chinese adults with high interdependent self-construal, their loneliness benefitted longitudinally from increased number of peripheral social partners over a 2-year interval. Although peripheral social partners may not necessarily be friends, this effect may be attributed primarily to friendships as increases in family members are far less likely than increases in friends. Likewise, Phillips et al. (2008) found that the number of friends was positively associated with older Chinese adults’ subjective well-being, although the quality and frequency of interaction with friends did not matter.

The dominating effect of friendship quantity over quality in the Asian context is contrary to what is observed in Western samples. Lucas and colleagues (Lucas and Dyrenforth 2006; Lucas et al. 2008) suggested that the mere existence of friendships only had a small effect on subjective well-being based on a quantitative review of studies in the Western context. Demir et al. (2013) also argued that friendship quality was more important than friendship quantity in affecting happiness. Such cultural difference regarding friendship quality and quantity may result from different culture norms of seeking support from friends. Social support is considered as an important mediator between friendship and subjective well-being (Chan et al. 2006; Lu 1995, 1999). Meanwhile, compared with Westerners, Asians were found to be less willing to explicitly seek social support from social partners, as they were more concerned with the potential negative relationship implications caused by the support seeking behavior, such as “worrying others, disrupting the harmony of the group, losing face, and making the problem worse” (Kim et al. 2006, p. 1596). With direct support seeking behavior, individuals with high-quality friendships can receive necessary social support from their close friends, so friendship quality is more strongly associated with subjective well-being in Western societies. However, without the direct support seeking behavior, friendship quantity may be a better indicator of available social support for an individual than the quality of specific friendships, so friendship quantity is more influential to subjective well-being in the Asian context.

To conclude, as a result of the dominance of collectivism and familism, although friendship is still an important source of emotional support among Asians, the impact of friendship on subjective well-being is less remarkable than that in the individualistic context. In contrast, family relationships remain the most preferred source of support across various domains. Moreover, the significant influence of family relationships on subjective well-being is no longer confined to difficult times in life, but becomes a general day-to-day phenomenon. However, as many Asian countries are undergoing rapid economic growth and social changes in recent decades, the roles played by friendship and family relationship are also under dynamic changes. Recent developing trends in the Asian context and potential future research directions regarding family relationship, friendship, and subjective well-being are discussed in the following section.

Developing Trend and Future Research Directions

The major developing trend concerning Asian families is that the size and stability of families is decreasing in recent decades. According to a report from the United Nations (2011), the total fertility rate in Asia dropped from 5.82 during 1950–1955 to 2.28 during 2005–2010. The decreasing trend is particularly significant in China as the result of the one-child policy. The total fertility rate dropped from 6.11 during 1950–1955 to 1.64 during 2005–2010. As a result, the size of the family network shrinks significantly. Meanwhile, influenced by modern values about marriage, individuals are less likely to treat marriage as a lifetime commitment. Instead, marriage is considered as optional and can be ended when it fails to satisfy one’s needs. Thus, the age at first marriage is delayed and the divorce rate increases significantly. For example, from 1981 to 2006, people’s age at first marriage increased by about 4 years for both males and females in Hong Kong. During the same period, the number of divorce decrees granted per year increased from 2060 to 17,424, while the number of new marriages per year only slightly increased from 43,386 to 50,242 (Census and Statistics Department, Government of HKSAR 2007). As family relationships are becoming fewer and less stable in Asian societies, individuals may not be able to obtain adequate and consistent support from family members any more. According to the hierarchical-compensatory model (Cantor 1979), friendships, as the next preferable source of support in one’s social network, may become more important in determining subjective well-being in modern Asian societies. Future studies can follow up on these changing trends and empirically test whether friendships are becoming increasingly important for Asian people’s subjective well-being.

Moreover, the above changing trends in demographics are also observed in Western societies. Arnett (2000, 2007) proposed the concept of emerging adulthood to represent the period from late teens to late twenties during which individuals had achieved certain levels of independence but had not settled down yet. Other researchers suggest that this life period also exists in Asian countries (Nelson and Chen 2007; Rosenberger 2007). Emerging adults usually are not married but many of them are involved in romantic relationships. Romantic relationship was found to be more influential on emerging adults’ happiness than friendship in an American sample (Demir 2010). Future studies can investigate whether the same pattern is also true in Asian countries.

Another direction for future research is to investigate the impact of family relationship and friendship on subjective well-being across different stages of adulthood. Most previous studies about this topic in the Asian context focused on older adults. The reason may be that the benefits from social support are more salient during late adulthood. However, social exchanges with family and friends are an essential part for young and middle-aged adults as well. Individuals face different life tasks and possess different goals over the life course (Ferraro 2001). Family relationship and friendship can serve different functions for people at different life stages. According to the social convoy model (Kahn and Antonucci 1980) and the socioemotional selectivity theory (Carstensen 2006), the importance of friendships decreases from early adulthood to later adulthood. Walen and Lachman (2000) reported that friendship strain impaired young adults’ subjective well-being to a greater extent compared with that of older adults. Furthermore, in a meta-analysis based on Western studies, Pinquart and Sorensen (2000) found that for older adults, higher relationship quality with adult children was more strongly associated with older adults’ life satisfaction than the quality of friendships. However, similar studies are lacking in the Asian context. To understand how family relationship and friendship coordinately provide people with a functional social support network that meets the dynamically changing developmental needs, it is of great importance to examine the longitudinal changes of the impact of family relationship and friendship on subjective well-being in the Asian context as well.

Moreover, family relationships and friendships can interact to influence subjective well-being. For example, Okun and Keith (1998) tested positive and negative social exchanges with spouse, children, and other relatives/friends in a larger American sample. They found significant interaction effects between exchanges from the same source in explaining younger participants’ depression, but significant interaction effects between exchanges from different sources in explaining older participants’ depressive symptoms. Similarly, based on a national sample in the United States, Walen and Lachman (2000) identified several buffering effects between relationship support and relationship strains in determining subjective well-being, either within the same relationship category (e.g., family support buffering family strains) or cross relationships (e.g., friends support buffering partner strains). Some of the buffering effects were only evident in women. Demir (2010) also reported that that romantic relationship quality significantly buffered the negative effect of conflict with the best friend for emerging adults in the United States. However, no consensus regarding the interaction effects between family relationships and friendships have been achieved and few studies have looked at this issue in the Asian context. Future studies can further investigate how family relationships and friendships dynamically interact with each other to support individuals’ subjective well-being across different cultures.

Lastly, some studies have indicated that subjective well-being may be the cause, instead of the result, of supportive family relationships and friendships because happier people may draw more support from the people around them (Adams 1988; Lyubomirsky et al. 2005). Although there has already been some longitudinal studies supporting that changes in social relationships predict changes in subjective well-being (e.g., Zhang et al. 2011), more longitudinal studies are needed to clarify the causal relationship between social relationships and subjective well-being.

Conclusion

Although family relationship is the dominating source of support for most people, friendship is also an indispensible part of people’s social network with unique functions. Based on studies with Western samples, family relationship provides most of the support across varying domains but family interactions tend to contribute significantly to subjective well-being only when people are facing difficulties in life. Interaction with friend is more significantly associated with subjective well-being in general, and reciprocity is a critical factor that determines friendship quality. However, influenced by collectivism and familism, the impact of friendship is less salient in Asian countries. Family relationship is consistently found to be an equally, if not more, important determinant of subjective well-being when compared to friendship. Meanwhile, it is important to notice that Asian families are becoming smaller and less stable. The effect of friendship may become increasingly important as a consequence of these trends. Future studies can investigate the dynamic roles of family relationship and friendship on subjective well-being across different stages of adulthood.