1 Introduction

This paper uses the four waves of the World Values Survey to examine how an individual’s values relate to that individual’s life satisfaction. People rank happiness-seeking as one of their most cherished goals in life (Diener and Oishi 2000). In previous studies, life circumstances and demographics, traits and dispositions, as well as behavioral intentions have been identified as three general predictors of subjective well-being (Lyubomirsky et al. 2005). However, little research has investigated personal values as a predictor of life satisfaction. Furthermore, those studies only focused on single cultural groups or several cultural groups. Whether the relationship of values and life satisfaction stands across different cultures is still not answered. In addition, some studies showed that modern values are positively correlated with subject well-beings, while some studies showed a negative relationship between secularism and subject well-being. The present study, therefore, aims to study the relationship of secular value and life satisfaction across time periods.

1.1 Values and Life Satisfaction

Schwartz (1992) defined values as “transituational goals, varying in importance, and serving as guiding principles in life” (p. 21). In the present study, a fundamental dimension of value, secularism versus traditionalism, will be examined. A person with traditional values is characterized as having “deference to the authority of God, Fatherland and family” (Inglehart and Baker 2000, p. 23). They emphasize the “importance of social conformity and have high levels of national pride” (p. 25). Societies and citizens with secular-rational values emphasize the opposite (Inglehart and Baker 2000).

Some studies have suggested a positive relationship of some modern values and subjective well-being. Strupp (1980) identified autonomy, responsibility, and fairness to others as good for mental health; Jensen and Bergin (1988) proposed that the values of self-direction, benevolence, and universalism promoted psychological health. Conversely, value types of conformity, tradition, security, and power are regarded as detrimental to well-being (Gat 1997, unpublished manuscript, cited in Sagiv and Schwartz 2000). Using the Schwartz Value Survey (1992), a study of college students in Germany, Nepal, and Fiji found that not only self-enhancement values, including power/achievement, and hedonism, but also values that emphasize maintenance of status-quo, including security and conformity, related to personal worries that was related to poor mental health (Boehnke et al. 1998). Moreover, the study of Sagiv and Schwartz (2000) found that the value types of achievement, self-direction, and stimulation associated positively, whereas tradition value associated negatively with affective well-being among students and adults from Israel and former East and West Germany. Furthermore, Lai (2004, unpublished manuscript) found that life satisfaction measured across three occasions was negatively associated with the endorsement of Schwartz’s (1992) self-enhancement goals in Chinese undergraduates.

As reported above, it has been found that the value type of tradition correlated with affective well-being negatively (Sagiv and Schwartz 2000), while autonomy and self-direction were regarded as beneficial to mental health (Strupp 1980). Secularism would be similar to the value of autonomy and self-direction, regarded as modern values, so it should thus be positively correlated with happiness. In contrast, tradition and the values conceptually related to tradition (e.g., conformity and security) were found to be negatively correlated with different well-being indicators, so traditionalism should be thus negatively correlated with life satisfaction.

On the other hand, there are ample studies showing a negative relationship of secularism and life satisfaction in the field of the psychology of religion. The concept of “secularization” in religious study refers to “the process in which religion gradually loses the encompassing and important role which it had in traditional society” (Halman and Draulans 2006, p. 265). A meta-analysis of values and religiosity found that religious people tend to like the values that were found to be negatively associated with well-being (such as tradition, conformity, and security) and dislike the values that were found to be associated positively with well-being in previous studies (such as self-direction and stimulation; Saroglou et al. 2004).

There is, however, great variation in the secularization process across cultures and nations, even in well-developed countries. For example, in the sociological literature, Europe is categorized as an exceptional case in secularization process (Berger 1999). Many indicators related to religious development, including lower attendance rates, poor financial support for religious institutions, and a shortage of clergy, reflect this process of secularization (Berger et al. 2008). The term “Eurosecularity” has been coined to describe how fast secular values have spread across the European continent (Berger et al. 2008). These authors even suggested that, “Western and central Europe is the most secularized area in the world” (p. 11, 2008); Davie (1999) has described Western Europeans as “unchurched” populations. Compared with Europe, another representative of well-developed countries, the United States, is still religious, despite the modernization process.

Ample studies consistently indicate a positive relationship between religiosity, the opposite side of secularization, and different indicators of well-being. Wulff (1997) suggests that, “Intrinsic religious orientation has proved to be positively associated with life satisfaction, psychological adjustment, self-control, better personality functioning, self-esteem, and purpose-in-life, while negative association with intrinsic orientation has been found for anxiety, death anxiety, neuroticism, depression, impulsivity, etc.” (p. 248).

There are several explanations for the positive relationship between psychological well-being and religiosity. Sole membership in a religious group is associated with increased levels of individual well-being (Pargament 2002). This religious belonging provides opportunity for spiritual experience (Schwenka 2000), social support, and cognitive framework for responding to existential questions (Ellison 1991) contributed to well-being. Thus, secularism in value orientation should be negatively associated with psychological well-being, as religiosity was found to be positively associated with well-being.

Thus, the relationship between secularism and life satisfaction may be positive or negative. This variation may depend on the way in which secularism and its putative opposite, traditionalism, is assessed. Alternatively or additionally, it may be that the cultural origin of respondents exercises some impact on the direction and strength of that relationship. Thus, we will adopt an exploratory approach to this relationship in our analysis of data from the World Values Survey which includes data collected from many cultural settings at four time periods.

1.2 The Role of Human Development

Neither country-level characteristics nor individual-level characteristics can solely explain individual well-being. Bonini (2008) found that region, national wealth, human development, and environmental sustainability accounted for 19% of the variance in individual well-being, while individual characteristics (and error) accounted for 81% of the variance. Such findings suggest that it is important to understand how country-level characteristics interact with individual-level characteristics to shape the level of well-beings felt by individuals. In light of such findings, the second purpose of this study is to investigate whether the association of life satisfaction and a secular versus a traditional value orientation is moderated by the respondent’s national culture.

There is considerable evidence showing a significant relationship of socio-economic development and well-being (e.g., Salmela-Aro and Schoon 2005), such that modernization and economic development are associated with higher levels of life satisfaction and quality of life in a given society. In his cross-sectional analysis of 90 nations, Veenhoven (2005) found that people live longer and happier in today’s most modern societies. Furthermore, he found that happiness and longevity have increased in ten modern nations over the last 30 years. Based on these findings, Veenhoven suggests that human beings live longer and happier than even before, since societal evolution is positive for quality of human life in the more recent transformation from less free and less healthy agrarian to freer and more affluent industrial societies. So then, does societal development moderate the link between values and life satisfaction, as has been found for other measures of well-being discussed above?

The human development index (HDI) was created by the United Nations in the early 1990s. The HDI is a composite index based on life expectancy at birth, adult literacy rate, the combined gross enrollment ratios for primary, secondary and tertiary schools, and GDP per capita in terms of purchasing power parity (UNDP 2000). The HDI is regarded as a more robust alternative to GDP per capita for measuring the relative developmental progress of nations, and has an international status and recognition as a United Nations-endorsed indicator (Bonini 2008).

In the present study, HDI will be treated as societal factor for investigating the moderating effect of societal variation on the relationship between a secular versus a traditional value orientation on life satisfaction. Countries with high HDI are more highly developed, modern countries. Much previous research has provided evidence supporting the position that socio-economic conditions are significant for determining life satisfaction at the individual level. Therefore, it is predicted that increased HDI should have a positive impact on the strength of association between traditional/secular values and happiness at the individual level.

On the other hand, socio-economic development may lead to changes in value endorsements. Inglehart and Baker (2000) stated that people’s attention would be influenced by societal development and that modern people emphasize personal values, such as self-expression more. Therefore, they found that countries with higher GNP per capita tend to be more secular. Besides, Allen et al. (2007) found that persons from richer countries tended to endorse secular values, such as Dionysianism, Autonomy, and Egalitarianism; whereas poorer countries tended to endorse traditional values, such as Submission, Embeddedness, and Hierarchy. In other words, countries with higher HDI tend to have a more secular societal context of values while countries with lower HDI tend to have a more traditional societal context.

Discrepancies between personal and prevailing environmental value priorities have maladaptive consequences for individuals. People are likely to experience a positive sense of well-being when they hold values more similar to those prevailing in their social environment. When they strive to attain the goals to which their values are directed in that society, their activities are more compatible with the societal Zeitgeist. Conversely, subjective well-being is likely to be undermined when there is discrepancy between personal values and those generally endorsed by one’s fellow citizens. Sagiv and Schwartz (2000) found that business and psychology students had greater well-being when their values were congruent with the values promoted by their respective academic departments. Similarly, when there are discrepancies between personal and societal value priorities, people may feel estranged from their cultural group, their personal goals are more difficult to realize, and this estrangement may cause people feel distressed and unhappy (Bernard et al. 2006).

Consequently, we hypothesize that the correlation between traditional/secular values and life satisfaction should be positive among people from higher HDI countries, whereas that relationship should be negative among people from lower HDI countries. In high-HDI countries, the common culture is individualism and egalitarianism, which is congruent with secularism. People who are more secular should have smaller discrepancies with societal common values, and in consequence they should be happier. In contrast, the common culture in low-HDI country is collectivism and hierarchy. People who are more secular should have larger discrepancies with societal common values; as a result, they should be less happy.

2 Method

2.1 Participants

The individual data used in the present study were extracted from the World Values Survey (1984–2004), which is the largest investigation ever carried out of attitudes, values, and beliefs around the world. This study has been conducted across four time periods using representative national samples in an increasingly large number of nations. Each wave will be used to examine whether any patterns detected are consistent across waves. There are 16, 36, 47, and 43 nations in Waves 1, 2, 3 and 4, respectively.

2.2 Measures

2.2.1 Refining the Measure of Values

A culturally equivalent measure of secularism was extracted from the items tapping traditional vs. secular values and survival vs. self expressive values originally used by Inglehart (1997). However, of the ten items labeled as “values” by Inglehart (1997), “Abortion is very justifiable” and “Homosexuality is never justifiable” are prescriptive norms; the item, “Respondent has not signed and would not sign a petition” is a report of a behavioral intention; the items, “You have to be very careful about trusting people” and “Respondent describes self as not very happy” are attitudes and self-assessments, respectively. These five items, therefore, are not values or important goals as typically defined in the values literature, although they may be conceptually related to values; indeed, they may be predicted by an individual’s values. Thus, in exploring the dimensions of value underlying the ten “values” of the World Values Survey, one should focus on the five, “pure” value items, viz., the importance of God, respect for authority, pride in one’s own country, autonomy index (which is a composite of several single items), and post-materialism index (which is a composite of several, single items).

The sample sizes from the different countries in the World Values survey were unequal. To avoid biasing the outcome by differences in national sample size, the procedure recommended by Becker (1996) for the meta-analysis of factor structures was used. This procedure weights each country’s correlation matrix equally and assumes that each sample from the different constituent countries provides the best approximation available of the correlations in its population.

The Inglehart (1997) autonomy index was expanded, and a one-factor solution was judged as obvious and compelling. This was the case at all four time periods, and therefore represents a robust, cross-culturally equivalent solution, accounting for 14, 15, 14, and 14% of the matrix variance across the four time periods, respectively. After recoding the scores for the items with negative loadings, the score for the autonomy value of any given respondent was computed by adding the scores for the 8 items measuring independence, determination, imagination, responsibility, obedience, religious faith, good manners, and thrift. The range of scores for the autonomy index was thus from 0 to 8. Table 1 shows the factor loading of each variable in a one-factor solution at each of the four time periods.

Table 1 Factor loadings of items tapping the autonomy index in different waves
Full size table

The procedure for finding the factor structure underlying the five values of traditional/secular value endorsement was the same. Only one factor showed an eigenvalue larger than 1 for data from any of the four time periods and it accounted for 37, 34, 29, and 30% of the total matrix variance, respectively, across the four waves. The reliabilities of the factor in Wave 1, 2, 3, and 4 were α = 0.62; α = 0.58; α = 0.52; and α = 0.57, at individual levels in four waves, respectively (Table 2).

Table 2 Factor loadings of items tapping traditionalism-secularism in different waves
Full size table

The scores for the items with negative loadings were first recoded. The score of each person on the resulting factor, secularism, was computed by adding items with a positive factor loading to the recoded scores for items with a negative factor loading. This score ranged from 0 to 5. And we called this factor as traditional-secular value. The higher the score, the more secular a given respondent is.

2.2.2 Life Satisfaction

The measure of life satisfaction in the WVS was taken by using a 10-point scale. The question posed was, “All things considered, how satisfied are you with your life as a whole these days?” where “1” refers to feeling dissatisfied and “10” refers to feeling satisfied.

2.2.3 Human Development Index

The dimensions of the human development index (HDI), including life expectancy, educational development, and GDP, are weighted equally. The index in 1985 1990, 1995 and 2000 were used to analyze the data in Wave 1, 2, 3, 4, respectively (UNDP).

3 Results

In order to present a comprehensive picture across time periods, we have performed the individual-level analysis for all participants in four waves. In order to have a culturally balanced assessment of the correlation between secularism and happiness across the many countries involved, the same procedure, i.e., a Fisher transformation, averaged and transformed back into a correlation, was applied when the bivariate correlation between secularism and happiness was calculated.

These correlations in each of the four waves were significantly negative, but weak [r = −0.07; r = −0.07; r = 0.02; r = 0.01, respectively, p’s < 0.01]. There is a directional change in the relationship of secularism and life satisfaction. In Wave 1 and Wave 2, higher individual life satisfaction was associated with a lower individual endorsement of secularism. The more secular the person, the lower life satisfaction he or she reported. However, in Wave 3 and Wave 4, higher individual life satisfaction was associated with a higher individual endorsement of secularism. The more secular the person, the higher the level of life satisfaction he or she reported.

3.1 Multiple Regression Analysis

In order to test whether HDI moderates the association between secularism and happiness at the individual level, multiple regression analysis was conducted. The main-effect variables were mean-centered before conducting the analysis in order to reduce possible multicollinearity and to facilitate model estimation when main effects and interactive effects were both present (Aiken and West 1991). Interaction terms were computed by multiplying the standardized values of measures (Aiken and West 1991).

The overall models were all significant in Waves 1, 2, 3 and 4 [F (3, 15,149) = 75.88, p < 0.001; F (3, 42,441) = 225.1, p < 0.001; F (3, 45,992) = 692.13, p < 0.001; F (3, 46,168) = 1,158.35, p < 0.001, respectively]. A main effect for HDI was consistently found such that those who live in a country with higher HDI were happier [t (15,151) = 14.13, p < 0.001, b = 0.12; t (42,443) = 25.56, p < 0.001, b = 0.14; t (45,994) = 35.88, p < 0.001, b = 0.19; t (42,170) = 50.25, p < 0.001, b = 0.27, respectively]. These findings are consistent with previous research, e.g., Veenhoven (2005).

In addition, a significant two-way interaction was found between an individual’s endorsement of secularism and HDI in predicting individual life satisfaction for all 4 waves [t (15,151) = 2.91, p = 0.004, b = 2.85; t (42,443) = 4.44, p < 0.001, b = 0.63; t (45,994) = 29.8, p < 0.001, b = 4.66; t (42,170) = 20.4, p < 0.001, b = 3.49 from Wave 1 to Wave 4, respectively].

To further explore the moderating effect of HDI, regression lines for those scoring around the mean in HDI, one standard deviation above and one standard deviation below the mean of the HDI were plotted (Aiken and West 1991). Figure 1 represents the interaction between HDI and traditional versus secular values for each wave. The graphs show that the effect of secularism on happiness varies as HDI changes in all waves.

Fig. 1
figure 1

The interactive effect of traditional/secular value and HDI in predicting happiness at the individual level in each wave of the WVS

Full size image

In the interaction graphs of Wave 1 and Wave 2, HDI moderated the strength of negative association between secularism and life satisfaction. A less negative association was found among people in countries with higher HDI in these first two waves; on the contrary, a positive association was found between secularism and life satisfaction among people in countries with higher HDI in Wave 3 and Wave 4. In countries with a lower HDI, secularism was consistently associated with lower reported happiness in all four waves.

4 Discussion

The purpose of this study was to investigate the relationship of traditionalism versus secularism with life satisfaction, and the consistency of this relationship across different temporal contexts.

The relationship between secularism and life satisfaction was not consistent across temporal contexts. Before the 90s, people with higher secularism of values appear to have lower life satisfaction across all societal contexts. After the 90s, people with higher secularism of values appear to have higher life satisfaction if they resided in countries higher in HDI. More studies should be carried out for understanding the directional change in this relationship. This study, however, may provide one plausible explanation for this change, i.e., the influence of societal development on value and life satisfaction.

Significant two-way interactions of traditional/secular values and HDI were found among individuals in all waves of the WVS. The results provide indirect evidence to support the notion that discrepancies between personal and societal value priorities have maladaptive consequences for individuals. This result is consistent with those of previous studies, e.g., Sagiv and Schwartz (2000), viz., when there are discrepancies between personal and societal values, people may feel estranged from their culture and this estrangement may cause people feel distressed and unhappy (Bernard et al. 2006), and thus experience lower life satisfaction.

Most highly developed countries, such as Sweden, Germany, and Canada, are characterized as individualistic cultures. In those countries, individualistic values and their pursuit are encouraged interpersonally and institutionally, so that secularism is congruent with culturally-held values. Estrangement and frustration with one’s society may be higher when individuals in individualistic cultures endorse and strive to achieve traditional values that are less compatible with widely held individualistic goals. Therefore, a positive relationship between secularism and happiness was found among people from high-HDI countries in Waves 3 and 4.

Conversely, most less-developed countries, such as Pakistan, Nigeria, and Morocco, are relatively collectivistic cultures. The societal culture in these nations is more hierarchical and authoritarian, so people who are more secular would experience larger discrepancies between personal and societal values. The larger the discrepancy, the more estranged they would feel, especially in more authoritarian, conformity-demanding cultures. Eventually, those people would encounter greater and greater frustration while striving to achieve their goals, and report less happiness. Thus, a negative relationship between secular values and satisfaction with life was found consistently across waves among people from low-HDI countries.

The use of imbalanced samples is one limitation to this study. Most countries selected are in Europe and nations from other continents are less well represented—34 out of 72 countries and regions countries were European, and more secular in values; in contrast, only three countries were from Africa and only 14 countries were from Asia, locations where traditionalism should be higher. Thus, cultures from the European continent are relatively over-represented. More non-European countries should be included in future studies in order to provide a more culturally representative data pool, thereby providing a better test of the culture fit hypothesis involving values.

Moreover, only a single question was used for tapping the individual’s level of life satisfaction. Although the psychometric properties of this single question tend to be strong, they provide only one approach to assessing subjective well-being (SWB; Diener 2000). This kind of scale can be easily contaminated by biases, including differing cultural conceptions of what constitutes happiness, e.g., Lu and Gilmour (2004). Participants are easily affected by temporary mood or situational factors, so that the overall happiness rating might not truly reflect participants’ happiness levels. For instance, a series of studies showed that global measure of life satisfaction can be affected by the respondent’s mood at the moment of responding to the scale and by other situational factors (Schwarz and Strack 1999). In the same study, it was also reported that the sequences of items in the survey and other artifacts can also influence reports of SWB.

Another potential problem is acquiescent response bias. Smith (2004) investigated acquiescent response bias by comparing large-scale cross-cultural surveys and found acquiescent response bias which may vary across countries. For example, people from the countries high on both family collectivism and preference for increased uncertainty avoidance tend to have high bias in responses to personally relevant items. Thus, a culturally comprehensive and counter-balanced measure of life satisfaction should be used in future research.