Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Universals and Cultural Differences in the Causes of Happiness

Over the past four decades, the study of subjective well-being has grown rapidly from a nascent field of inquiry to a major research endeavor. In its most succinct definition, subjective well-being (SWB) constitutes the level of well-being experienced by people according to their own subjective evaluations of their lives. These evaluations include cognitive judgments about life satisfaction, affective reactions to life events, interest and engagement, and satisfaction with specific domains, such as work, relationships, health, recreation, and meaning and purpose.

Interest in cross-cultural topics has flourished, primarily during the past 15 years. Several works fostered interest in the cultural–psychological perspective, namely, Veenhoven and Ehrhardt (1995), Diener et al. (1995a), Diener and Diener (1995), and Diener and Suh (2000). A growing body of research has shown that while many people around the globe are relatively happy, distinct differences in subjective well-being exist between nations which can be explained, to some extent, by the effects of culture. While most people within industrialized societies score above neutral in surveys of well-being (Diener and Diener 1996), international surveys reveal significant differences in subjective well-being levels (see Diener and Suh 2000 for a review) that can be explained, to some degree, by the level of economic development of the countries surveyed. For example, Diener et al. (1995a) found that wealthier nations frequently reported higher SWB and that it is not merely coincidental that some of the unhappiest nations are also the poorest (see also Stevenson and Wolfers 2008).

However, evidence from some cultures suggests that material wealth is not solely responsible for happiness across the globe. For example, Biswas-Diener et al. (2005) found that the Maasai, Amish, and Inughuit cultures reported levels of subjective well-being that were above neutral despite their relatively impoverished living conditions. This finding is consistent with data gathered from those living in slum housing, sex workers, and homeless pavement dwellers in Calcutta, who reported levels of life satisfaction that were only slightly below neutral (Biswas-Diener and Diener 2001). In addition, evidence suggests that the effects of poverty are moderated, to some degree, by culture. For example, while well-being in the homeless is predictably low across cultures, pavement dwellers in Calcutta show higher levels of life satisfaction than homeless groups in the United States (Biswas-Diener and Diener 2006). Interestingly, this occurs despite the fact that the homeless in the United States have better access to food, clean water, medical care, opportunities for employment, and adequate shelter. In addition, Biswas-Diener et al. (2012) found that low-income respondents in the United States reported higher negative affect and lower life satisfaction than low-income respondents in Denmark. Thus, the effects of income depend partly on the cultural context.

It is also important to note that the social benefits produced by a nation’s wealth may contribute to citizens’ experience of high subjective well-being. Diener et al. (1995a) found that human rights and social equality, factors which are highly correlated with subjective well-being, were highly correlated with national income. In addition, recent research (Tay and Diener 2011) has shown that psychosocial resources, such as feeling respected and being able to count on someone in an emergency, contribute significantly to well-being and, in some cases, are more important than material circumstances. In the past decade, researchers have increasingly focused on this distinction in the causes of well-being and have discovered the importance of both affluence and psychosocial resources in subjective well-being across nations. Diener et al. (2010) found that income best predicts life satisfaction, whereas psychosocial resources best predict positive emotions. We now know that the happiest nations are wealthy and are able to meet the psychosocial needs of their citizenry, whereas the unhappiest nations are less wealthy and are unable to meet these needs. Despite this general conclusion, intriguing cultural universals and differences in the causes and structure of subjective well-being in nations remain, which we will discuss in this chapter.

Components and Structure of Subjective Well-Being

The validity of these cultural comparisons depends on how subjective well-being is conceptualized within individual cultures. For example, do all societies understand the idea of life satisfaction? Are there a set of universal emotions, or are most emotions culture-specific? And if differences in the conception of subjective well-being do exist between cultures, how do we reliably and validly measure SWB in order to make cross-cultural comparisons?

As mentioned above, the concept of subjective well-being includes several key components. We will primarily discuss three: life satisfaction, frequent positive affect (or pleasant emotions), and infrequent negative affect (or unpleasant emotions). These components have been researched most thoroughly across nations and provide broad descriptions of subjective well-being around the globe.

A growing body of evidence reveals that cultures tend to have similar perceptions of which emotions are perceived as positive or negative (Diener et al. 2004; Watson et al. 1984; Shaver et al. 1992). A core cluster of positive emotions, such as pleasant and happy, as well as a core cluster of negative emotions, such as unpleasant, sad, and angry, tend to consistently reappear despite some variations in how noncore emotions, such as pride, are clustered (Diener et al. 2004). Tay et al. (2011) found that certain emotions are experienced as pleasant across cultural regions of the world, and the same is true of certain negative feelings. Evidence also suggests the similarity of structure of life satisfaction across cultures (Vittersø et al. 2002), providing support for the idea that SWB has a consistent general structure across the globe. Thus, there is reason to believe that these aggregates can be compared cross-culturally to yield valid results about the levels of subjective well-being in different societies.

Although high positive affect, low negative affect, and high life satisfaction are all components of high subjective well-being across nations, it is important to note that there are some differences between cultures in the relation of these components. Suh et al. (1998) found that life satisfaction and affect balance (the difference in frequency of positive and negative affect) correlated positively across 40 countries. However, Suh et al. also found that the correlations were stronger in individualistic societies and weaker in collectivistic societies. In the second part of their study, when life satisfaction was predicted from emotions and perceived norms, emotions were most predictive in individualistic cultures while perceived norms were most predictive in collectivistic cultures. Perhaps this difference exists because emotional feelings are considered very important in individualistic cultures, which emphasize people’s personal goals and feelings. In contrast, collectivistic cultures may produce a greater tendency to rely on normative standards as a guide for one’s actions and judgment because there is a greater emphasis on group harmony.

Norms for Subjective Well-Being

When discussing the well-being of nations, it may be surprising to learn that, on average, most societies are fairly happy. Diener and Diener (1995) found that the majority of respondents in 31 nations reported above-neutral levels of life satisfaction, and Diener and Diener (1996) found that 86% of a total 43 sampled nations reported above-neutral levels of subjective well-being. Even impoverished or nonindustrialized societies, such as the Maasai of Africa, the Inughuit of Greenland, and the Amish in the United States, report above-neutral levels of well-being (Biswas-Diener et al. 2005). This is not to say that suffering no longer exists in the world; importantly, many of the poorest nations of the world were often absent from the early data sets. People may be dispositionally prone to positive emotions due to their adaptive advantages (Fredrickson 1998).

Importantly, the societal norms for emotions and life satisfaction, which differ from culture to culture, affect the experience of well-being. For example, East Asian respondents often report lower SWB than Europeans or Americans (Diener and Diener 1995). While this difference could be linked to economic factors, it is interesting to note that Japan often reports lower SWB than Latin American countries, despite its far greater purchasing power (Diener and Diener 1995; Diener and Oishi 2000; Diener and Suh 1999). Some have posited that this phenomenon occurs because the Japanese and other Asians are more accepting of negative emotions than are other countries (Diener and Suh 1999). In addition, Asian-Americans and Chinese participants have been found to value low-arousal positive affect (calmness) more so than do European-Americans, whereas European-Americans value high-arousal positive affect (excitement) more than do Chinese participants (Tsai et al. 2006). Tsai et al. (2006) posited that Asian cultures show greater acceptance of negative emotions and value low-arousal emotions because these attributes foster greater attention to the social context. In this way, a culture’s perception of the desirability of certain emotions might influence people’s frequency of experiencing those emotions and their subsequent reporting of well-being. While many of these differences in societal norms are intriguing, it is important to note that valuing positive emotions less, for example, does not mean that East Asian cultures are unhappy. Rather, they only report less happiness relative to other cultures, such as those in Latin America. Actually, East Asians score consistently above the midpoint in SWB, and it is possible that, for a culture rooted in balance and moderation, these modest levels of well-being might be viewed as most desirable (Kitayama and Markus 2000; Schimmack et al. 2002a).

Do People Around the Globe Find SWB Desirable?

An important issue in assessing cross-cultural subjective well-being is the degree to which different societies value it. Evidence shows that the desire to be happy is virtually universal, with respondents in 47 nations ranking happiness highly, and as more important than wealth or physical attractiveness (Diener 2000; Diener and Oishi 2004). While on average, happiness is rated as most important, there is some variance between cultures in the degree of importance. For example, Latin Americans place more importance on happiness than do East Asians. However, despite variations in the degree to which SWB is valued, its universal desirability provides ample evidence that SWB should be studied cross-culturally.

Units for Studying Cultural Differences

In examining culture, one can study various sizes of units that capture the phenomena in different ways. Because shared practices, beliefs, and feelings can occur in various groupings of people, and because each of these will nevertheless contain some amount of heterogeneity, discrete cultural units that are completely homogeneous, but different from all other groups, do not exist. In other words, culture can be regarded as a group average that operates in a hierarchy, as below:

  1. 1.

    Cultural dimensions—for example, individualism and collectivism

  2. 2.

    Geopolitical/sociocultural regions—for example, sub-Saharan Africa and East Asia

  3. 3.

    Nations—for example, Denmark and Japan

  4. 4.

    Small distinct relatively homogeneous cultural groups—for example, the Maasai and the Amish

Studying each one of these levels of culture has its own advantages and limitations. For example, the dimensional approach has the advantage of being able to compare all people but has the potential problem of oversimplifying. For instance, the popular individualism–collectivism dimension can be applied to any world region, but at the same time, the collectivism of different regions often varies to some degree (Brewer and Chen 2007; Oyserman et al. 2002). The small-group approach, which has been most extensively used by anthropologists, has the advantage of focusing on relatively homogeneous groups with distinct characteristics, but at the same time, it leaves us with a very long list of different cultures and often very small sample sizes (see Henrich et al. 2005 for an exemplary work). The nation and geopolitical units define geographical regions with some cultural homogeneity owing to common factors, such as history, language, and politics, but at the same time, there can be distinct groups within these regions. Lucas and Diener (2008) discuss the issues involved in comparing the subjective well-being of nations.

All four approaches capture some of the cultural differences between groups. In each type of unit, there are shared cultural characteristics that, on average, apply to individuals within them and that, on average, are to some degree different from other groups. In each case, we can also examine the degree to which the cultural unit used is associated with differences in the factors that are associated with well-being. Thus, we analyze the universals and differences in the causes of well-being across these four types of cultural units. If the results tend to be consistent, we will have more confidence in our conclusions. We also examine mean-level differences between cultural groups and whether the structure of subjective well-being is consistent across culture. An important issue that we do not address is the degree of heterogeneity in cultural units.

Cultural Universals in the Causes of Well-Being

Cultural Dimensions

On a dimensional level, individualism has shown a strong correlation with well-being. Diener et al. (1995a) found that income, individualism, and human rights all showed strong relations to SWB in a survey of 55 nations. However, it has thus far been difficult to disentangle these variables. Because a nation’s economic development usually occurs in tandem with cultural changes, it can be challenging to ascribe heightened subjective well-being to a specific economic or social improvement. Despite greater social support in collectivistic societies, some have theorized that individualism may be a predictor of well-being because people have greater freedom in choosing the course of their lives in individualistic societies. People in these societies also may attribute their success to themselves rather than to the social network and, thus, feel greater personal satisfaction and fulfillment, at least if they are succeeding.

Sociocultural Regions

On a sociocultural region level, there is evidence not only that universal causes of well-being exist across cultures but that these universals exist in connections to specific types of SWB. Tay and Diener (2011) examined six types of needs (basic needs for food and shelter, safety and security, social support and love, respect and pride in activities, mastery, and self-direction and autonomy) in relation to three types of subjective well-being (life evaluations, positive feelings, and negative feelings) across eight sociocultural regions of the world in order to assess the association between needs and types of SWB. Life satisfaction was best predicted by the fulfillment of basic needs, whereas positive feelings were best predicted by the fulfillment of social needs. Importantly, the associations of needs with particular forms of SWB were consistent across all eight sociocultural regions.

New research consistently suggests that income, conveniences, and satisfaction with standard of living consistently predict evaluations of life satisfaction but that psychosocial factors more strongly influence the components of positive and negative feelings (Diener et al. 2009a). In Table 8.1, we present correlations from the Gallup World Poll Wave 1, between income, household conveniences, psychosocial resources, and life evaluations and positive feelings. As can be seen throughout the world and in all world sociocultural regions, life satisfaction was related to material aspects of life, that is, household income, the ownership of household conveniences, and the meeting of basic needs for food and shelter. Positive feelings, in contrast, were most related to respect, mastery, and social support, and these correlations tended to be consistent in all regions of the globe. The weakest associations were found in the Northern Europe/Anglo region, probably because of ceiling effects, with most individuals owning conveniences and having their basic needs met. The correlations of life satisfaction and psychosocial resources, and of positive feelings and national measures, were not as strong. Thus, across sociocultural regions, there are several commonalities in the causes of life satisfaction, positive feelings, and negative feelings.

Table 8.1 Correlations with two types of well-being
Full size table

Nations

While there are some universal societal factors, such as democratic governance, human rights, and longevity, which are related to SWB across nations (Diener et al. 1995a), there are also cross-national similarities in terms of personality characteristics and social bonds that are associated with happiness. For instance, Lucas et al. (2000) found that extroversion was related to positive feelings in all 39 nations examined. Kuppens et al. (2008) also found that, across nations, positive emotions were more strongly associated with life satisfaction than was the absence of negative emotions. Fulmer et al. (2010) found that there is a congruity effect such that people who have personality characteristics that match the dominant traits in that cultures tend to report higher levels of well-being.

In terms of social resources, longitudinal studies and studies of large representative samples reveal that married people are, on average, happier than non-married people (Glenn 1975; Lee et al. 1991; Lucas et al. 2003). This correlation holds true across nations. Diener et al. (2000) found that being married was associated with high SWB almost universally across cultures, although there were some cultural differences, such as lesser negative affect among divorced persons in collectivist cultures. Therefore, strong social bonds such as those provided by supportive marriages seem to be universally beneficial to SWB, along with extraversion and positive feelings. However, the issue of reverse causality—high well-being leading to marriage better relationships and extraversion—cannot be ruled out.

Small Homogeneous Groups

Even when examining small cultural units, some universal causes of happiness emerge. In a study of homeless individuals from Calcutta in India, and from both California and Oregon, Biswas-Diener and Diener (2006) found that all three groups reported high levels of negative affect and low satisfaction with material resources, such as income and housing. Likewise, all three groups reported similar levels of high satisfaction with domains related to the self (e.g., morality, intelligence, and physical appearance), which indicates the resiliency of self-appraisal despite the negative effects of poverty. In each locale, homelessness was predictably associated with lowered well-being, which suggests that having one’s basic needs unmet is a strong predictor of lowered life satisfaction around the globe. These results were replicated in another Calcutta study that surveyed those living in slum housing, sex workers, and pavement dwellers and found that life satisfaction decreased incrementally as the fulfillment of basic needs diminished (Biswas-Diener and Diener 2001). Specifically, pavement dwellers (i.e., those who live without housing) had much lower life satisfaction than both sex workers and those living in the slums. Thus, even among impoverished groups, the fulfillment of a basic need such as housing can drastically affect the way individuals evaluate their lives. Thus, satisfaction with material resources is strongly linked to overall life satisfaction across diverse groups.

Similarly, strong social relationships have been shown to universally aid well-being. Homeless groups in Calcutta showed greater satisfaction with social relationships than did homeless groups in the United States. Moreover, sharing living quarters with others was also related to increased life satisfaction in Calcutta, whereas homeless respondents in the United States commonly showed a lack of trust in other homeless people around them (Biswas-Diener and Diener 2006). Interestingly, respondents in Calcutta also reported higher satisfaction with material domains and income, despite having better access to food, clean water, medical care, employment, and adequate shelter available to them in the two US locations (Biswas-Diener and Diener 2006). Although respondents in India may display higher levels of satisfaction because the country, in general, shows much more sympathy and less social stigma for the poor, social support seems to play a key role in the differences between the homeless groups surveyed. Because homelessness in Calcutta is associated less with pathology or personal fault and more with economic conditions, the homeless in Calcutta tend to stay close to their family units, whereas their counterparts in the United States are often estranged from relatives, children, and spouses. In this way, the social support provided to the Indian homeless may be the reason why they show levels of life satisfaction not only higher than their American counterpart but which fall in the positive range.

Cultural Differences in the Causes of Well-Being

Cultural Dimensions

Although there are similarities across cultures in terms of the causes of well-being, cultural differences also exist. A few of the areas where there are substantial differences are in the way certain emotions are valued across cultures, in how social approval affects judgments of life satisfaction, and in the perceived importance of individual achievement versus communal harmony. For example, the value of self-esteem to SWB is moderated by individualism, as indicated by the fact that self-esteem is a strong predictor of life satisfaction in individualistic cultures, but not in collectivist cultures (Diener and Diener 1995). Similarly, satisfaction with one’s freedom was more strongly associated with life satisfaction in individualistic cultures than in collectivistic cultures (Oishi et al. 1999).

In the area of emotion, Suh et al. (1998) found that emotional experiences were more strongly associated with life satisfaction in individualistic cultures than in collectivistic cultures. Suh et al. (2008) extended these findings by using a priming procedure. Specifically, American and Korean participants in their study were primed either with the concepts associated with individualism or collectivism before making their life satisfaction judgments. When participants were primed with individualism, their life satisfaction judgments were based primarily on their own emotional experiences. When they were primed with collectivism, however, their life satisfaction judgments were more strongly influenced by social appraisals (how parents and friends viewed participants’ lives) than their own emotional experiences. These findings suggest that many cultural differences are not absolute but depend on the focus of attention and salience of information at the time of reporting (see Oyserman and Lee 2008 for review). Because people are more likely to attend to information that is seen as important in their culture, differences in the correlates of life satisfaction arise because of the differential accessibility of relevant information (Oishi et al. 2000).

Sociocultural Regions

We could not locate previously published findings on regional differences in the patterns associated with well-being. Thus, we analyzed Wave 3 of the Gallup World Poll for well-being of marital status groups across a few regions of the world. Table 8.2 shows the life evaluation scores (0–10 scale) and positive feelings scores (0–1.0 scale) for each region.

Table 8.2 Life satisfaction and positive feelings of marital status groups in four world regions
Full size table

Several important conclusions emerge from our examination of the table. First, the patterns are different for the two types of well-being. These pattern differences point to the absolute necessity of measuring different types of well-being in cross-cultural research and of not making blanket statements about happiness. For instance, widows in the Middle East are relatively high in life evaluations but very low in positive feelings. Across marital groups, Northern Europe is clearly highest in life evaluations, but South America is highest in positive feelings.

In terms of specific marital categories, widows are the worst off in life evaluation in two of the regions, but the divorced are lowest in the two other regions. In terms of positive feelings, widows are the worst off in all regions, but this deficit is most dramatic in the Middle East, where widows are also relatively highly in life evaluation. In every region, the married are highest in positive feelings, but singles score highest on life evaluations in two other regions. The finding that unmarried individuals are relatively happy (DePaulo and Morris 2005) replicates across regions and measures, as does the finding that divorced individuals are low in well-being. At the same time, the findings reveal that the relative standing of marital groups can differ depending on cultural region as well as type of well-being. Thus, we need to take a more careful look at where and why patterns replicate. Clearly, the processes leading to life evaluation versus positive feelings are distinct.

Nations

As stated above, Lucas et al. (2000) showed that the association between positive affect and extraversion was similar across 39 nations. Schimmack et al. (2002b) replicated this finding, showing that the latent link between extraversion and hedonic balance (PA–NA) was positive and the link between neuroticism and hedonic balance was negative in all five nations examined, namely, the United States, Germany, Mexico, Japan, and Ghana. Interestingly, however, consistent with Suh et al. (1998), the latent association between hedonic balance and life satisfaction was larger among Americans and Germans than among Mexicans, Japanese, and Ghanaians.

When specific types of pleasant emotions were examined, several cross-national differences also emerged. For instance, happiness was strongly associated with interpersonally disengaging positive emotions (e.g., pride) among Americans, whereas it was strongly associated with interpersonally engaging positive emotions among Japanese (e.g., fureai, Kitayama et al. 2000, 2006). Similarly, pride loaded on the positive mood factor along with other positive emotions among European-Americans and Hispanic Americans, whereas it loaded on both positive and negative moods among Asian-Americans, Japanese, and Indians (Scollon et al. 2005). More recently, Tsai and colleagues found that the ideal positive emotion entailed high activation (e.g., excitement) among Americans, whereas it entailed low activation (e.g., calm, peaceful) among Chinese (Tsai et al. 2007a, b).

In a similar vein, relationship harmony was more strongly associated with life satisfaction in Hong Kong than in the United States (Kwan et al. 1997). Likewise, among Japanese and Filipinos, social support had a direct effect on subjective well-being above and beyond self-esteem, whereas social support did not predict subjective well-being among Americans beyond self-esteem (Uchida et al. 2008). Namely, social support was associated with subjective well-being among Americans to the extent that it was associated with self-esteem.

There are also cross-national differences in motivational processes involving subjective well-being. For instance, Americans are more motivated to view themselves as consistent across different roles and situations than Koreans (Suh 2002). Furthermore, self-concept consistency was more strongly associated with life satisfaction in the United States than in Korea. There are cross-national differences in terms of the relation between goal attainment and subjective well-being. For example, Japanese college students who were pursuing their goals to make their family and friends happy became more satisfied with their lives over time as they achieved their goals than did those Japanese who were not pursuing their goals to make their family and friends happy (Oishi and Diener 2001). In contrast, American college students who were pursuing their goals for themselves became more satisfied with their lives over time as they achieved their goals than did other Americans who were not pursuing their goals for themselves. Likewise, among Americans, pursuing goals with an avoidant mindset was negatively associated with life satisfaction, whereas it was not negatively associated with life satisfaction among Koreans and Russians, who view avoiding negative evaluations as important (Elliot et al. 2001).

Other studies have shown cross-national differences in the role of specific situations or life events in subjective well-being. For instance, Oishi et al. (2004) conducted a cross-national experience-sampling study in which participants were beeped at random moments and asked to record the situations they were in, and their moods, for 1 week. For instance, they found that both Americans and Japanese were happier when they were with friends and their romantic partners than when they were alone. However, this effect of friend and romantic partner, respectively, was significantly stronger among Japanese than among Americans. In the same study, they found that Indians felt more negative emotions when they were with strangers than they were alone. American participants did not feel any more negative emotions when they were with strangers than when they were alone. Overall, Japanese and Indian negative affective experiences varied, to a greater degree, across different interpersonal contexts than did Americans’. In another study, Oishi et al. (2007) conducted a 3-week daily diary study in the United States, Korea, and Japan, in which participants were asked to record their daily life events and well-being. On average, American participants experienced more positive events per day than Koreans and Japanese. Interestingly, it took Americans about two positive daily events to mitigate the effect of one negative event on their daily satisfaction. In contrast, it took about one positive event to counteract the effect of one negative event among Japanese and Koreans. Because American participants experienced more positive events (e.g., compliments) than Koreans and Japanese in general, the value of a positive event among Americans was lower than among Koreans and Japanese (see Diener et al. 2003; Tov and Diener 2007 for more comprehensive review on cross-national differences in well-being).

Small Homogeneous Groups

Biswas-Diener et al. (2005) found that the Kenyan Maasai, the US Amish, and the Greenlandic Inughuit all reported above-average levels of life satisfaction, domain satisfaction, and affect balance. However, the Amish reported lower satisfaction with self-related domains, whereas the Maasai and Inughuit reported lower satisfaction with material domains. While all three groups reported high satisfaction with social domains, the Maasai were the only group that reported frequently feeling substantial pride. Thus, subjective well-being may differ in a more intricate pattern, even when the overall level is similar across groups.

More recently, Cox (2009) examined the subjective well-being of sex workers, dump dwellers, urban poor, rural peasants, and university students in Nicaragua. Among urban poor, income was significantly positively associated with global life satisfaction. In contrast, however, among sex workers, dump dwellers, rural poor, and university students, income was unrelated to global life satisfaction. One important issue here is the measurement of material wealth or resources among these groups. While income may not affect well-being in cultures such as the Maasai, where money is hardly used, the ownership of cattle, for instance—which is an important material asset—is correlated with life satisfaction.

Cultural Universals in the Structure of Well-Being

Cultural Dimensions

Although there has been debate regarding the universality of emotions, research shows that while individual emotions and the situations that cause them may differ from culture to culture, positive and negative affect can be validly assessed across individualistic and collectivistic cultures. A number of studies show that countries around the world use emotion words that cluster into the two main categories of positive or negative emotions. Positive and negative affect clusters emerged from studies examining emotion words in Japan (Watson et al. 1984), as well as in the United States, Italy, and China (Shaver et al. 1992). These studies provide evidence that positive and negative feelings are perceived across nations and that emotions, such as joy, anger, and sadness, appear to be universally experienced as positive or negative.

In addition to positive and negative affect, there is also support for the similarity of life satisfaction across cultural dimensions. Vittersø et al. (2002) found that a one-factor model of the Satisfaction with Life Scale (SWLS; Diener et al. 1985) fit reported life satisfaction in 41 nations. This study not only suggests that the SWLS measures a single construct but also suggests that cultures around the world have similar notions of life satisfaction regardless of what conditions they think contribute to it.

Sociocultural Regions

Diener et al. (2004) studied the frequency of 12 emotions and found that they resulted in positive and negative clusters for all seven regions of the world, including Africa, Latin America, East Asia, Southeast Asia, West Asia, Eastern Europe, and Western Europe. Importantly, in virtually all regions, certain emotions were consistently placed into the positive or negative clusters: positive emotions included pleasant, cheerful, and happy, while negative emotions included unpleasant, sad, and angry.

Similarly, Tay et al. found that across six cultural–political regions of the world, emotions fell along a positive–negative valence continuum. Words such as cheerful, pleasant, happy, and love were viewed as positive in all regions, and words such as sad, worry, unpleasant, and anger were seen as negative in all regions.

Nations

Nation-level studies have provided much support for the validity of measuring positive and negative affect across cultures. Based on 9,300 respondents from 48 nations, Kuppens et al. (2006) concluded that there are two broad universal affect factors—pleasant and unpleasant feelings—that characterize people across the globe. In addition, Scollon et al. (2004) found that positive feelings and negative feelings each formed similar clusters across nations. In this study, it is noteworthy that emotions for which there are no equivalent English-language emotions, which they called indigenous emotions, also clustered as would be expected in the pleasant or unpleasant clusters. Moreover, these results also extend to the within-person level analysis, as evident by Scollon et al. (2005) finding that positive and negative emotions formed separate and strong factors in all of the five cultures they studied. In other words, the feelings of the same valence tend to be experienced together.

Small Homogeneous Groups

Russell (1983) asked Indians from the Gujarat region of India (who spoke Gujarati only) to sort 28 emotion terms in terms of similarity and dissimilarity. Much like the two-dimensional structure of emotions in English, the similarity judgments of emotions in Gujarati resulted in the two-dimensional circumplex of emotion with the pleasant/unpleasant and arousal as the major dimensions.

Cultural Differences in the Structure of Well-Being

Cultural Dimensions

Despite cultural universals, interesting structural differences in SWB originate from cultural–emotional norms, which play an important role in the frequency of reported emotions, and thus provide insight into the measurement of well-being across the globe. Emotions which conflict with societal norms may be deemphasized and, consequently, experienced less frequently. This, in turn, may explain why some cultures report certain emotions as negative which are normally viewed as positive in the majority of nations.

Cultural dimensions may affect the way the separate components of SWB are related to each other. Kuppens et al. (2008) found that individualism moderated the association of negative emotions with life satisfaction, with people in individualistic nations being more adversely affected by unpleasant feelings. In cultures that value self-expression, positive feelings had a stronger relation with life satisfaction.

Kuppens et al. (2006) found that beyond the individual pleasant and unpleasant factors that strongly describe people’s feelings across cultures, there were two nation-level factors that described less, but non-negligible, amounts of variance in the reports of feelings: positivity and interpersonal emotions. One nation factor consisted of feelings related to other people rather to than unpleasant feelings. This factor included high loadings on guilt, shame, gratitude, and jealousy. The positive-nation factor was related to the life satisfaction of nations and to the cultural appropriateness of expressing positive emotions. Importantly, the interpersonal dimension was inversely related to the individualism of nations, suggesting that in collectivistic nations, people pay more attention to interpersonal feelings. This attention to the collective may be explained by Kitayama et al. (2009) finding that individualistic cultures tend to favor emotions associated with independence, whereas collectivistic cultures tend to favor emotions associated with interdependence.

Sociocultural Regions

Schimmack et al. (2002a) found that pleasant and unpleasant emotions were inversely correlated in many sociocultural regions (e.g., Europe, Latin America). However, pleasant and unpleasant emotions were not inversely related in the East Asian region. Tay et al. (2011) used latent class analysis to examine the structure of affect in persons in greater detail. Using the Gallup World Poll, they did find a class with low levels of positive affect and high levels of several types of positive affect. Conversely, they found another class with moderate levels of positive affect and relatively low level of negative affect but also high stress. These two classes represent the positive–negative dimension that characterized most structural work on the emotions, with the exception of stress, which went with the positive emotions for the second latent class. However, there was also a class reporting relatively high negative and positive emotions found frequently in Latin America. Finally, there was a class with relatively light positive emotions, except pride, and very low levels of sadness and worry. Thus, a valence dimension was somewhat universal across the globe, but the patterns were, in fact, more intricate than values alone would suggest.

Although the Tay et al. study cited earlier found a positive–negative valence dimension in all regions of the world, there were certain terms that were more ambiguous in specific regions. The positive terms pride and gratitude and the negative terms jealousy and guilt did not have a clearly delineated valence value in some cultures, suggesting some experienced ambiguity in low positive versus negative valence dimensions.

Nations

Eid and Diener (2001) examined the desirability and appropriateness of pleasant and unpleasant affect in the United States, Australia, Taiwan, and China. Roughly 83% of Americans and Australians belonged to the latent class, in which all positive emotions joy, affection, pride, and contentment were perceived as appropriate. In contrast, China and Taiwan proved to be very heterogeneous, with people falling more evenly across several latent classes based on the desirability of various combinations of emotions. Pride was the outlier emotion in this study, with 57% of Taiwanese having mixed feelings about this emotion and 32% of Chinese feeling that pride was clearly inappropriate. Although it is important to recognize cross-national differences in emotional norms, it is also important to note that emotional norms might not always strongly influence emotional experience. For instance, Tsai et al. (2006) found that the emotions people value are not necessarily the emotions they experience most often.

For life satisfaction, Oishi (2006) used item response theory (IRT) to examine measurement equivalence of the Satisfaction with Life Scale (SWLS) between the United States and China. Item 2, “The conditions of my life are excellent,” was equivalent across the two nations, whereas items 1, 3, 4, and 5 were significantly different (significant DIF). In particular, item 4, “So far I have gotten the important things I want in my life,” and item 5, “If I could live my life over, I would change almost nothing,” showed large DIF. These items did not measure life satisfaction of Chinese very well. In other words, Chinese who endorsed item 2, “The conditions of my life are excellent,” did not necessarily endorse items 4 or 5. In contrast, Americans who endorsed item 2 also endorsed items 4 and 5. This differential item functioning (DIF) is disconcerting in interpreting the mean differences of life satisfaction across nations. Interestingly, however, the latent life satisfaction scores that were estimated based on the differential item functioning (more equivalent items get more weights than less equivalent items) did not substantially change the estimation of mean differences in the SWLS between the USA and China. With the raw scores, Cohen’s D was 1.18, whereas with the IRT-based score, it was 0.71 (still the large effect). Thus, although life satisfaction has some common core across societies, there can also be culture-specific item differences at the national level, as well as reliability differences.

Small Homogeneous Groups

Although we were unable to find the studies that directly test the structure of emotions in small homogenous groups, there are many pieces of evidence that suggest cultural variations in the structure of well-being. Russell (1991), for instance, identified cultures that do not have the corresponding words for so-called basic emotions. There is no word for disgust in Polish, Ifaluk, and Chewong. Additionally, there is no word for sadness in Tahitian and Chewong, no word for fear in Ifaluk, Utku, and Pintupi, and no word for surprise in Fore, Dani, Malay, and Ifaluk.

In addition, anthropologists have identified many indigenous concepts associated with well-being. For example, Menon and Shweder (1994) identified the indigenous concept, “lajya,” in the Orissa region of India. The best English translation of lajya is shame, though it also means shyness and embarrassment. Oriyas believe that lajya (shame) is a feminine virtue that is both powerful and good. Menon and Shweder asked participants in Orissa and the United States to pick one emotion that is most different from the rest. The target emotions were raga (anger), sukha (happy), and lajya (shame). Indians in the Orissa region viewed anger as the most different from shame and happy, whereas many Americans viewed happy as most different from anger and shame. For Oriyas, shame is instrumental to maintaining harmonious relationships with others, whereas anger is detrimental to harmonious relationships. Anger is detrimental to social relationships in the United States, as well. The critical difference, however, was that the instrumental value of emotion in social relationships did not come to mind naturally among Americans whereas that was the first dimension that came to mind among Oriyas. Instead, the pleasant versus unpleasant dimension was the first to come to mind for many Americans.

Mean Levels of Subjective Well-Being Around the World

Diener and Diener (1996) suggested that most people in the world are happy because humans are predisposed to experience mild positive emotions when they are in positive or neutral circumstances. Research has shown that in neutral, ambiguous, and positive circumstances, people show a slight positive offset, which engenders more sociable, approach-oriented, and optimistic behavior (Ito et al. 1998). This is consistent with the Pollyanna Principle advanced by Matlin and Stang (1978). Studies show that people are prone to dampen negative emotions quickly (Taylor 1991). Importantly, this pattern may be adaptive in that it causes most people to experience low-intensity positive feelings of well-being the majority of the time, as long as nothing bad is happening. These feelings of well-being yield a state in which individuals confidently explore their environment, approach new goals, and gain important personal resources (Fredrickson 1998). However, negative events can quickly draw people’s attention to dangers. The person reacts quickly to the negative event, and thereafter, the unpleasant emotions are dampened and the person returns to a positive state.

Diener and Diener (1996) reviewed evidence from around the world and found support for the idea that most people are, in fact, happy, but since that time, the emerging picture has grown more intricate. For example, Diener et al. (2009) reported that although virtually all nations reported more positive feelings than negative feelings (or a positive affect balance), in a large number of nations, respondents reported average life satisfaction that was below neutral.

In this chapter, we have examined the first three waves of the Gallup World Poll, which included about 360,000 respondents representatively sampled from 146 nations. The results were compelling in showing a positive offset for feelings, but not for life evaluations. At the individual level, 54% of respondents were at or below neutral in life evaluations, and in 39% of nations, the average respondent was at or below neutrality on evaluations of life. In contrast, 62% of respondents felt more enjoyment than sadness, and in every one of the 146 nations, the average respondent felt more enjoyment than sadness. Thus, the positive offset applies to feelings, but apparently not to evaluations of life. Apparently, people can be dissatisfied in terms of having what they desire but nevertheless be prone to feel somewhat positive feelings. Thus, people might have, on average, a positive set point in their emotional lives yet be capable of judgments of their lives that are below the midpoint of the scale. Clearly, this is a very important question for future research.

The general predisposition to positivity offset does not mean, however, that all cultures are the same when it comes to livability and high SWB. There are pronounced differences between societies in each type of subjective well-being. For example, in the Gallup World Poll, life satisfaction on a 0–10 scale varied from an average of 3.1 in Togo to 7.9 in Denmark. Having frequently felt enjoyment yesterday varied from 89% of respondents (Canada, Denmark, and New Zealand) to 56 % in Georgia. Affect balance between enjoyment and sadness varied from a high of 87 in Laos, meaning 87% more people felt much enjoyment than felt much sadness, to a low of 49 in Armenia, where approximately the same number of people felt sadness and enjoyment.

The mean levels of SWB in cultures are related to circumstances such as income and lack of corruption (Diener et al. 2009) and also to cultural dimensions such as femininity, power distance, and uncertainty avoidance (Basabe et al. 2000). Basabe et al. (2002) found that the Human Development Index of the United Nations (assessing the income, education, and life expectancy of nations) correlated substantially with the life satisfaction of nations but only weakly with affect. In contrast, cultural dimensions such as masculinity, power distance, and individualism were substantially associated with affect balance.

Diener and Lucas (2000) presented evaluation theory to integrate the diverse findings on the causes of SWB and found that relative standards, cultural differences, and the fulfillment of both needs and goals can all influence SWB. They suggest that culture can direct people’s attention toward certain factors, just as inborn needs can, and that this focus can lead to a greater impact for the factor receiving attention.

Integrating the Insights

Several broad conclusions are now clear regarding culture and the experiences of well-being. One finding is that variation exists in the causes of different types of SWB across cultures, for example, life satisfaction versus positive affect. Money is an important goal in the modern world, and obtaining it correlates with life satisfaction around the world. For cultures that rarely use money, such as the traditional Maasai, a proxy for money which gives access to resources (e.g., the ownership of cattle) can be associated with well-being. In contrast, positive affect, unlike life satisfaction, is strongly related to social support cross-culturally, supporting the notion that the positive affect system and social relationships are intimately tied.

Another important conclusion is that there are several cultural universals in the structure of SWB, for example, that positive emotions tend to be frequently experienced by the same individuals, as do negative emotions. Furthermore, there is much agreement about the values of most of the major emotions. However, there is variability across cultures in the desirability of high-arousal emotions, as well as specific emotions, such as pride. Therefore, researchers can compare cultures on pleasant and unpleasant feelings but need to be cautious about the meaning of specific feelings, such as pride, which can differ in their valence. Some caution is also needed in aggregating emotions of the same valence because there can be latent classes of individuals who experience different ensembles of emotions. Moreover, researchers should also pay careful attention to the ways in which the associations between positive and negative emotions can differ across cultures.

In general, our review gives reason for optimism in terms of our ability to study SWB across cultures. Data show that SWB generally can be validly measured across cultures. There are similarities in the structure of feelings so that comparisons across cultures are possible. Even where individuals within groups differ in the structure of their SWB, latent class analysis provides a method for comparing similar groups across cultures (Eid and Diener 2001). Tay et al. (2011) show how latent class analysis can be used to identify similar classes of people in each culture and then compare the frequency of these latent classes across cultures.

There is reason to believe that we can, to some degree, describe desirable societies in terms of certain universals that produce high SWB, for example, those with adequate material and social resources. At the same time, cultures, to some degree, also must be judged based on their own criteria, as some causes of happiness differ across cultures. In some cases, there are differences between cultures at the specific and concrete causes of well-being but consistency at a more abstract level. For example, Oishi and Diener (2001) found that different types of goals most enhanced SWB in different cultures. However, it is likely that achieving goals leads to well-being in all cultures but that the specific goals can differ to some extent.

Future Directions

Despite the growing amount of data across cultures gathered at the dimensional, sociocultural, regional, national, and small homogenous group levels, our review indicates important future directions for research. One of the most important is the study of the outcomes of well-being across various groups and cultures. A growing body of evidence suggests that high levels of SWB provide benefits of effective functioning not only at the individual level but on the societal level as well. For example, happy people tend to engage more frequently than unhappy people in altruistic, prosocial activities such as volunteering, (Tov and Diener 2008); happy people tend to have better relationships, better health and longevity, and higher incomes (Pressman and Cohen 2005; Diener and Biswas-Diener 2008; Lyubomirsky et al. 2005). In addition, happy people report higher levels of trust, cooperation, and propeace attitudes, and they are generally more confident in the government and more democratic and less intolerant of immigrants or other racial groups (Tov and Diener 2008; Diener and Tov 2007). Similarly, Tov et al. (2009) found that high SWB led to greater societal trust and more confidence in parliament. Importantly, these patterns of association were stronger in societies that were objectively better, that is, in nations higher in equality and income and lower in war-related deaths. Thus, although happy people tend to perceive societal conditions in a more positive light, they did so most in the desirable conditions that justified these positive evaluations.

Despite the promising findings on the benefit of high subjective well-being, little research has been conducted comparing the majority of these outcomes across cultures. The finding of Diener et al. (2002) that happy students later earned higher income was replicated by Graham and her colleagues in Russia (Graham et al. 2004),, but there have been few longitudinal studies replicating results on the outcomes of well-being. The important question is whether the benefits of well-being generalize across cultures or are restricted to the westernized nations where most of the research has been conducted.

One area where more research is required is on measurement artifacts. To avoid these effects, multiple measures should be utilized in more studies. Diener et al. (1995b) found little evidence that measurement artifacts caused cross-nation differences in SWB. However, Scollon et al. (2004) found that the cultural groups they studied did not differ in negative emotions when online recording of emotions was used to measure them but that they did differ in retrospective global reports of emotions. Similarly, Oishi (2002) found that European-Americans and Asian-Americans did not differ in online positive emotions but did differ when they retrospectively recalled their emotions for the period of the online recordings. These findings indicate that online experience-sampling methods ought to be used more frequently, and careful attention ought to be paid to the patterns produced by online measures versus retrospective measures of feelings. Although some researchers might perceive differences between measurement methods as a threat to the field, in fact, they are a golden opportunity for understanding the origin of cultural differences.

Finally, the effects of language should be studied further in order to refine surveys of well-being used across cultures. Boroditsky (2001) demonstrated how language influences the way we think, and therefore, the connotation of words—the most frequent medium of measurement of SWB—could vary across cultures. In addition, it could be that for bilingual speakers, the language of the measure could prime different thoughts. For bicultural individuals, they can switch between different cultural frames. For example, we performed an initial analysis of language with a large cross-national college student sample and did not find strong effects. Namely, individuals within a nation using different languages appeared similar, whereas individuals in different nations who used a common language did not appear similar. However, this initial examination was far from definitive, and much more research is needed. For example, people may report different levels of life satisfaction depending on the language used or the measurement setting, for example, if the measure is administered at home versus at work. Thus, although we have made substantial progress in our understanding of subjective well-being across cultures, there are still many exciting and promising topics for future research.