The Changing Nature of
Cultural Values and
Socioeconomic Development:
An Analysis on Global
Generations
ByMarijn Suzanne van der Peijl
Student Number: S2796333
Supervisor: Ms. Ileana Maris
Co-Assessor: Mr. Rudi de Vries
University of Groningen
Faculty of Economics and Business
ABSTRACT
Cross-cultural research has been conducted for many years, highlighting that awareness on cultural differences is essential when doing international business. A recently mentioned development is the changing nature of cultural values and the implications this has for management. Generational subculture theory suggests that a group identity is shaped during one’s impressionable years and that these group identities create different generations with different cultural values. The infant concept of global generations suggests that these group identities can be evident across borders, thereby creating global generations, which are shaped by global events with a big impact. This thesis examines such an event, globalization, and the differences between generations that are exposed and unexposed to this event. Additionally, the socioeconomic development that comes with globalization is researched as a possible explanatory factor for these generational differences. Results indicate that difference between generations exist on the Hofstede dimensions of power distance, individualism and indulgence. While different levels of socioeconomic development are related to different cultural value scores, socioeconomic development is not the best explanation for these differences to exist. The democracy level in a country however shows significant relationships with both power distance and individualism, indicating that the institutional and political environment might be a better explanation for value changes to occur.
TABLE OF CONTENTS
Abstract 2
Table of Contents 3
List of Tables and Figures 5
Acknowledgements 6
1. Introduction and Problem Definition 7
2. Literature Review 10
2.1 National Culture 10
2.2 Cultural Change 10
2.2.1 Generational Subculture Theory 12
2.2.2 National Generations 13
2.2.3 Global Generations 15
2.3 Globalization 16
2.4 Socioeconomic Development 19
3. Hypotheses Development and Theoretical Framework 23
3.1 Cultural Change and Generational Differences 23 3.2 The Moderating Effect of Socioeconomic Development 26
4. Data 28
4.1 Sample 28
4.2 Operationalization of the Variables 28
4.2.1 Generations 28
4.2.2 Cultural Values 29
4.2.3 Socioeconomic Development 32
4.2.4 Control Variables 34
4.3 Estimated Model 35
5. Multiple Regression Analysis 36
5.1 Linearity 36
5.3 Outliers 37 5.4 Correlations 37 5.5 Multicollinearity 38 6. Empirical Results 39 6.1 Power Distance 39 6.2 Individualism 40 6.3 Uncertainty Avoidance 41 6.4 Indulgence 42
6.5 Summary of the Results 43
7. Discussion 45
7.1 Generational Differences 45
7.2 Socioeconomic Development 46
7.3 The Importance of Political Regime 48
7.4 Limitations 48
7.5 Suggestions for Further Research 50
7.6 Implications for Management 51
8. Conclusion 53
References 54
Appendices 61
Appendix A: Sample Countries 61
Appendix B: Multiple Regression Summaries 62
Appendix B1: Power Distance Summaries 62
Appendix B2: Individualism Summaries 64
Appendix B3: Uncertainty Avoidance Summaries 66
List of Tables and Figures
Figures
Figure 1: Different Conceptualizations of Western Generations (Reeves and Oh, 2008) 13 Figure 2: Levels of Culture (Erez and Gati, 2004) 27 Tables
Table 1: Cohort Structure 29
Table 2: Sample Sizes per Dimension 34
Table 3: Correlations Matrix 37
Table 4: Power Distance Model 40
Table 5: Individualism Model 41
Table 6: Uncertainty Avoidance Model 42
Table 7: Indulgence Model 43
Acknowledgements
1. INTRODUCTION AND PROBLEM DEFINITION
Ever since the notion of globalization was mentioned much research has focused on conducting business practices abroad, and the complications that this ‘liability of foreignness’ (Zaheer, 1995) brings for multinational organizations. This liability of foreignness implies that multinational organizations face costs when operating abroad due to unfamiliar environments, economic, cultural and political differences and the need to coordinate these, as well as geographic distance (Zaheer, 1995). McLaughlin and Fitzsimmons (1996) mention that ‘with globalization, the impact of cultural adaptation will need to be central to our study of [many business] topic areas’ (cited from Zhang & Wu, 2014, p. 313)
When organizations want to operate abroad, cultural differences are said to be one of the most prevalent things to take into account. Not only do organizations need to be aware of these differences, they also need to know how to deal with these differences and how to solve dilemmas related to culture. Additionally, Molinsky et al. (2012) mention that, while many managers expect that having a ‘global mindset’ is sufficient when working across borders, this is in fact not true. Apart from knowing that differences between cultures exist, it is, according to Molinsky et al. (2012), important to be able to switch between cultures. This skill, called ‘cultural-code switching’ requires managers to adapt their behavior to their culturally different subordinates. Luc Minguet, CEO of Michelin, describes in Harvard Business Review how his inability to do so created problems for him in managing and communicating with employees when starting to work in the United States. Unawareness and inability to respond to these differences can result in conflict between management and employees (Minguet, 2014). When doing so correctly, organizations can use cultural differences to achieve superior results, according to Finn Majlergaard, founder and CEO of global consultancy company Gugin (2010). As described by Minguet (2014), Molinsky et al. (2012) and many others, it is important to get a good grip of these different cultures and the ways they effect managerial decision making.
studies however, view national culture as a static thing, something that does not change over time and that one can anticipate on.
Cultural change literature suggests that national culture is anything but static. It is argued that differences in cultural values can be caused by being raised at a different point in time and thereby having a different environment to grow up in. These different points in time can be grouped into generations. A generation is defined by Kupperschmidt (2000) (cited in Lu and Gursoy, 2013: 4) as “an identifiable group sharing birth years and significant life events at critical developmental stages such as common historical or social life experiences, the effects of which are relatively stable over their lives”. An area of management theory where these generational differences in values have become increasingly clear is in the area of human resource management. For instance, members of one of the latest generations to enter the workforce, generation Y, see work less as something that is central in their lives and therefore have much lower job satisfaction and loyalty towards their organizations (Lu & Gursoy, 2013). In response to this, Valentine & Powers (2013) state that organizations need to be aware of these changing perceptions of the job and need to anticipate on this in order to engage these workers and avoid employee turnover. This explains why one of the main problems is the frustration of managers that run into difficulties when they use the cross-cultural manuals (Fang, 2005) that assume homogeneity in cross-cultural values.
This indicates that it is important for business to be aware of these changes in values and thereby in national cultures. Businesses need to anticipate upon these kinds of differences to determine the optimal business partners and the way to work with them when internationalizing. Hofstede indicates this with the notion of ‘mutual fit’ (Hofstede, 1984), which implies contingency between national culture and management practices. It is therefore of high value to get a more thorough understanding of the cultural dimensions at current times and especially if there are differences between generations regarding values. A recently published article by Beugelsdijk et al. (2015) takes these differences into account and focuses on the changing nature of cultural values by comparing two generational cohorts. Their results suggest that cultural values indeed change. They do however, not state the causes of these differences.
relating these to the process of economic development. The research objectives are to identify factors that contribute to the changing nature of cultural values, thereby generating a more thorough understanding of the concept of national culture and to support the notion of global generations by indicating differences between global generational cohorts.
This thesis will examine these effects by analyzing cultural change in multiple countries. Collecting all relevant data resulted in a total sample of 35 countries spread over five continents. It will contribute to international business literature by addressing the subject of cultural change in national culture. While previous business research into cultural change has mainly focused on changes within organizational cultures, little is known about the changes in national cultures (Fang, 2005) and what causes these changes. This thesis aims to give more clarity on this subject by focusing on generational subcultures, which states that different generations within a nation have different cultural values and one of the possible factors underlying these changes, socioeconomic development.
Additionally, the results of this research will indicate to what extent multinational companies operating in different countries need to anticipate on changed cultural values. As mentioned earlier by Majlergaard (2010), the correct anticipation can have a significant influence on achieving superior results. The question that guides this research is the following:
2. LITERATURE REVIEW
2.1 National Culture
National culture was first defined by Hofstede (2010 p. 6) as “the collective programming of the mind distinguishing the members of one group or category of people from others” in his book Cultural Consequences. After the publication of this revolutionary book, much research has focused on the effect that differences in cultural values could have on the way business is done. Especially in the field on international business, cross-cultural research has grown to a large body of research. Cultural differences are evident in the choice of entry strategy (Kogut & Singh, 1988), decision-making and hierarchical structures (Daniels & Greguras, 2014), leadership styles (House et al., 2002; Spreitzer et al. 2005), entrepreneurial behavior (Wennberg et al. 2013), relationship building (Griffith et al. 2014), reward systems (Kuhn, 2009; Tsang, 2002), the perception of long-term goals (Carson et al., 2014) and so on. The list in endless and continues to grow.
One of the main arguments against the many models of national culture is the fact that they are assumed to be stable over time (Fang, 2005). According to Kostis and Petrakis (2013) cultural values are seen as psychological stereotypes interpreted in a human’s mind. These characteristics do not change over time and show great resistance against change. This view is more often perceived as being wrong. Schwartz (2006) argues that, while slowly, cultural values do change over time. Social adaption to exogenous factors leads to this change. Cultural stability, according to Leung et al. (2005), is however a notion that is favored by managers, because it reduces ambiguity and enables managers to have a certain degree of control over expected behavior in new environments. Just like Schwartz (2006), Leung et al. (2005) argue that cultural stability can only be assumed as long as there are no environmental changes. Moreover, they state that while cultural change has been recognized, it has hardly been studied at the national level.
2.2 Cultural Change
He defines this as the materialist-postmaterialist framework, developed through a study in political orientation. He explains values are related to the situation in which one was born. In this situation, individuals have certain unsatisfied needs, which are then the main priority in the individual’s life. An individual born at a different time can have other unsatisfied needs, which in turn are his or her priority and shape their values. This changing prioritization creates a change in cultural values (Inglehart, 1971). Inglehart’s (1971) paper gave a start sign to examine the relationship between change in cultural variables and changes in all kinds of other factors. Lesthaeghe (1983) explains the changes in fertility rates and procreation by indicating the increased centrality of individual goal attainment in life and the higher valuation of self-fulfillment. This research has led to a dynamic view of national culture, in which culture is suggested to exist of cognitive structures and process that are affected by changes in the environment (Leung et al. 2005). Moreover, this view suggests that changes in culture are not as scarce as previously assumed.
Variations within national culture are often attributed to differences between the ‘main’ national culture and different subcultures within that culture, or to personal deviations and preferences (Fang, 2009). The existence of differences in national cultural values between generations offers an alternative explanation for the differences of cultural values within one national culture. Much research attributes the changes in cultural values to being born in a different period of time. Inglehart (1971) uses different generational cohorts (born before and after World War II), to explain his view of changing values. Fang (2008) explains cultural change as cultures having a ‘moment’, by which he means that cultures are different at different moments in time. Egri and Ralston (2004) compare Chinese and American managers on values such as openness to change and conservatism and found significant changes both between the generations, as well as between the countries. No evidence was found for the convergence of values. However, they do state that generation gaps are equally relevant for both countries. Beugelsdijk et al. (2015) support their claims on cultural change by examining different generational cohorts (born before and born after 1958). They discovered that indeed generations differ in their cultural values. In general, they state that societies have become more individualistic, less power distant and more indulgent (Beugelsdijk et al. 2015). The cultural distance between countries has however, not changed.
address this issue by listing a number of factors that could possible moderate the relationship between cultural value change and generational differences. It is however, first necessary to get a view of the generations that are currently in the working force and are therefore most relevant to International Business.
2.2.1 Generational Subculture Theory.
In their study on the influence of cultural values on moral philosophy, Robertson et al. (2012) notice that the differences in moral behavior are not only caused by the difference in personal values, but also by the differences in the generation that they were born in. They state that “generational subculture theory is grounded in the notion that macro-level social, political, and economic events that occurred during a generational group’s impressionable years shape a group identity. This group identity is comprised of a distinctive set of values and beliefs that remain relatively stable over time (Robertson et al. 2012, p. 100)”. Not only do they state that differences exist between different generations, but also that these differences remain stable over time. Arsenault (2004) noticed that generational differences are often underappreciated, since the formation of norms and values during the influential years of a generation is often questioned. In his study however, he validated the fact that generations develop a unique mind-set, translating into preferences for emotions, attitudes and values.
Applying Robertson et al. (2012) model to cultural differences in general, it is likely that the circumstances in society at the point of influence, the impressionable years, are a main determinant on the values they develop. This notion is supported by the system view of Kitayama (2002), which states that an individual organizes his of her psychological processes through actively trying to coordinate these with cultural systems and practices. These psychological processes develop flexibly and are subject to different environmental circumstances, which results in differences across different groups.
or category of people from others” (Hofstede et al. 2010, p.6). Combining these two definitions shows that different generations can be programmed differently, thereby having different cultural values. In the current literature, no uniform conceptualization of generations exists. This thesis will therefore review the most frequently used terminologies of generations, as well as the notion of ‘global generations’.
2.2.2 National generations
In current generation research, much attention has been focused on defining generations within national boundaries. A distinction can be made between three generations that are currently under a lot of attention in the Western management literature: Baby Boomers, Generation X and Generation Y. As shown by Reeves and Oh (2008), many different conceptualizations of generations exist (Figure 1).
Figure 1: Different Conceptualizations of Western Generations (Reeves and Oh, 2008)
Generation X includes individuals raised in period where both parents started to work outside the home and divorce rates were higher than ever before, many generation X-ers have turned to friends for comfort and support, resulting in work coming on the second place after family and friends (Young et al. 2013). Generation X-ers are therefore characterized by a high degree of independence. Moreover, they do not seem to attach a lot of value to authority, they work smart and efficient rather than hard and find a balance between work-life and recreational activities (Lamm and Meeks, 2009).
Generation Y is the latest generation entering the workforce. This generation has grown up in a ‘wired’ world, having connection with everyone at any time (Ryan, 2000). Additionally, they were raised with strong legislations and high educational and safety standards. They are characterized by having strong ambition, lots of confidence and high achievement standards (Lamm and Meeks, 2009). They like to have responsibility and value having input in the decision-making process (Wong et al. 2008).
Not only Western management literature distinguishes different generations. Schewe and Meredith (2004) define Russian generational cohorts: the Stagnation generation (1953-1968), the Perestoika generation (1969-1974) and the Post-Soviet generation (1975 and later) being the three generations currently in the workforce. Latin-American countries, such as Brazil also have different generations: The Iron Years generation (1951-1962), The Lost Decade (1963-1974) and the ‘Be on your own’-generation (1975 an later) (Meredith and Schewe, 2004).
capital, communication, technologies and others across borders, which makes the concept of generations within nation states out of date (Beck and Beck-Gernsheim, 2009).
Since one of the objectives of this thesis is to move away from generations defined along national boundaries and to take a global perspective, it will take the notion of ‘global generations’ into account. In the next section, this notion is explained and arguments for using it in this study are given.
2.2.3 Global Generations
Karl Mannheim, who’s work was published in 1997, was one of the first to look into the concept of generations. Between 1920 and 1933, he looked at generational differences, thereby holding on to generations within national boundaries. He could have never guessed that from the end on the twentieth century on, generations could also develop across national boundaries. According to Edmunds and Turner (2005), it is important to start thinking about generations in terms of ‘globalism’, where global events facilitate the development of global generations. These events define their values preferences and attitudes (Schewe & Meredith, 2004). These shared experiences are embedded in these values and they do not change throughout the life. Beck and Beck-Gernsheim (2009) mention that the idea of generations within national boundaries is therefore outdated and that it is methodologically important to look at global generations.
While research regarding global generations is in its infancy, some have attempted to distinguish global generations based on events that made a global impact. Edmunds and Turner (2005) make a first attempts at this. The first global generation that they mention is the 1960s generation, which came into existence due to high levels of social capital and solidarity, resulting from a common goals to protest against the Vietnam War which caused threats of nuclear action and the use of public resources to fund the war (Edmunds & Turner, 2005; Schewe & Meredith, 2004). While these protests were mainly visible in the USA and Europe, they were the motivation for many other protests in Asia, Africa and the Middle-East. Faber (1994) states that the 1960s generation has been essential in changing culture and politics around the world.
opened up new possibilities for communication and for involvement in a ‘global civic society’, which was highlighted by politicians. The increasing internationalization of trade has lead governments to give specific attention to related issues, such as green politics and ethics of care. The Internet made it possible for individuals to witness these consequences of global trade through the spread of images and texts, which lead to awareness of people’s living conditions and poverty. The attack on the World Trade Centre on 9/11 has been marked as the event that gave rise to the third global generation (Edmunds & Turner, 2005). While Edmunds and Turner have made a first attempt at actually indicating global generations, other researchers have paid attention to the notion of global generations. An important remark regarding global generations comes from Schewe and Meredith (2004). They mention that the influence of certain historic events shaped the values and believes of people who already were of age during that event. This supports the notion made earlier by Robertson et al. (2012) who mentions that events must have occurred during a persons impressionable years.
As is indicated by Edmunds and Turner (2005), global events that make an impact shape people’s values. Beck and Beck-Gersheim (2009) argue that this spreading of events not only happens to the distribution of Western values through the world, but also the other way around. Third World values spread amongst the Westerners as well, increasing the realization of inequality in the world as well as the motivation to act upon it. Moreover, there is a larger appreciation of the own situation when being exposed to situations from across the globe (Beck and Beck-Gersheim, 2009). They indicate that these movements of thoughts, but also of capital, technologies and the establishment of global movements as well as global terrorist networks are part of the process of globalization, which influences they way people perceive the world and shape their values. Globalization as we know it today is therefore likely to be one of the ‘influential events’, mentioned by Edmunds and Turner (2005) that contributes to the development of global generations.
2.3 Globalization
Baldwin (2006) argues that this is mainly caused by two phases on unbundling. The first phase of this unbundling process consist of two waves of which the first started in the 1850s ad was initiated by the industrial revolution. The second wave struck by the end of the 1960s and made a separation of production and consumption of goods and products possible due to a decrease in transportation costs. The second phase, according to Baldwin (2006), enables a decrease in communication costs as well as coordination costs, which resulted in the possibilities for outsourcing processes and tax to lower income countries. This led to large networks of firms operating across borders.
Apart from enabling cross-country operations, Baddeley (2006) argues that globalization promotes trade, allows easy access to capital markets, infrastructure development and export in developing countries. Additionally, globalization allows for the transfer of values from developed countries to developing countries, thereby contributing to their level of development. Leung et al. (2005) state that globalization has a significant effect on the distribution of economic rewards in a non-uniform way. According to them, globalization has, especially in the early 2000s, led to political and economic changes. They state that adaption is the key to change and that values have changed as a result of this adaption process (Leung et al. 2005). Moreover, they discuss the convergence of cultures. Especially in the domain of consumer lifestyle a shift towards more converging values can be noted due to the embracement of Anglo values. Since the process of globalization is largely dependent upon multinational corporations, Gupta and Govindarajan (2000) argue that the values that start dominating the global context are mainly based on Western values, which include democracy, free markets, diversity, freedom of choice and individual rights. Additionally, Beugelsdijk et al. (2015) mention that global exposure to Western media, management practices, education systems, and consumer goods will likely lead to a convergence of cultures. But not only does this lead to a convergence of cultures, it also helps developing economies to increase their level of development, by anticipating upon the information transferred from the ‘developed West’, and using their local skills to support the markets by providing products related to the wishes of the global consumer.
line of thought and different reasoning than being born in a world without these levels of globalization. The globalized generation compares their own reality to the one of others in different parts of the world, without the limitation of their direct surroundings, but thinking without borders and realizing the environments in other geographical locations.
Not only does the globalized generation learn to think without borders, they also have different standards. Migration, travelling and the possibility of reaching other continents open up possibilities and change peoples’ mindsets. Large distances are no longer an inhibiting factor to change the own situation and the globalized generation realizes that changing their own situation is possible since they know that a better place exists elsewhere. Beck and Beck-Gernsheim (2009) argue that globalization has informed people about the inequalities in living standards and makes them less acceptable of their own (bad) situations. Meyer (2000) additionally argues that globalization leads to intensified cultural changes, since both people as well as organizations are influenced by standard operating models, generally accepted laws and regulations, educative ideologies, but also the transfer of music, jokes and many others. The event distinguishing the two generations in this thesis is globalization and the change in values that globalization has brought. It is therefore necessary to determine the moment at which globalization has begun. While an exact year is hard to indicate, research has tried to come close. While the use of the word ‘globalization’ only started in the 1980s, Robertson (1992) mentions that in fact, the process began around the mid 1960s. On behalf on the United Nations, Shangquan (2000) supports this claim, stating that governments started to emphasize the benefits of global trade around 1970. Additionally, the second wave of the first phase of unbundling as mentioned by Baldwin (1960) occurred by then end of the 1960s. Globalization as we know it today therefore started approximately between 1965 and 1975. In order to compare the globalized and the unglobalized generation, a line has to be drawn here. Based on the information mentioned above, as well as limitations on the available data, 1970 has been indicated as the first year of globalization as we know it today.
Glass (2007), who also uses the age of 18 as the end year of the impressionable years. Therefore, this paper will follow most research by ending the impressionable period at the age of 18.
For the individuals born before 1970, a number of them have been exposed to globalization for a part of their impressionable years. These individuals inhibit examining the absolute effect that globalization has on changing values. The unglobalized generation will therefore consist of all individuals who have not been exposed to globalization during their impressionable years. This includes those individuals who where older than 18 by 1970. The second generational cohort will therefore include all individuals born before 1953.
As argued in this section, there is a difference in cultural values between those born in a globalized society (the globalized generation) and those who where not born in a globalized society (the unglobalized generation). As mentioned in the introduction of this thesis, the aim is to explain how globalization has changed these cultural values and how the differences between these two generations can be explained. Since globalization has led to higher levels of economic development, a possible explanation for these differences between the two generations is socioeconomic development. This will be discussed in the next section.
2.4 Socioeconomic Development
causes of poverty, and low incomes, thereby focusing on giving suggestions on how to improve these conditions (Acemoglu, 2010).
According to Ingelhart et al. (1996) cultural values are, next to institutional factors, an important determinant of socioeconomic development. This vision dates back to Weber (1904), who explains the emergence of capitalism through the protestant church. According to Weber, capitalism is based on the foundations of Protestantism. Protestantism had very specific values. The job was seen as a ‘calling’, and the benefits of working (money) were seen as the ultimate proof of your love for God. Additionally, Protestantism favors investing over consuming, so money earned was invested, which was the start of capitalism. This example from Weber (1904) shows that there is indeed a relation between the values people have and the market.
In current times, theories on economic development can be classified under the notion of ‘modernization theory’. According to Maseland and Beugelsdijk (2011), modernization theory does not only involve the questions of how countries have become rich, but a larger view of economic development, which consist also of social and cultural change. Citing Nash (1964), Beugelsdijk and Maseland (2011, p. 53) state that “modernization is the process of making societies, cultures, and individuals receptive to growth of tested knowledge and its employment in the ordinary business of daily living”. Inglehart (1997) follows up on modernization theory, by mentioning that modernization brings new values. Their central statement is that modernization has cultural and political consequences, which indicates the effect of economic development (modernization) on the cultural values of people. He states that economic development leads to individuals redirecting their priorities and thereby, changing their norms and values. This implies that the globalized generation has a different set of cultural values, caused by the redirection of their priorities due to the socioeconomic situation at the time shaped their values.
In the area of institutional theory, Chang (2011) describes how economic development influences institutions. Increased wealth leads to an increase in demand for institutions of higher quality. Second, the increase in wealth increases the affordability of institutions. Third, increases in wealth change the requirements people have, thereby demanding new institutions. This view by Chang (2011) complements Inglehart’s (1997) view, indicating that economic development changes the requirements and priorities people have in societies.
Not only Inglehart (1971) has investigated the relationship between culture and economic development. Research on the dimensions by Hofstede (1980) shows evidence of a relationship between cultural values and economic development. As explained by Diener et al. (1995), social progress and thereby increasing levels of development is closely related to subjective wellbeing. Based on the Weighted Index of Social progress (WISP) they show that power distance negatively influences social progress and thereby wellbeing. This may be caused by the strict formalization in high power distance societies, in which rules and regulations are enforced to maintain the current state of affairs, thereby inhibiting social progress. Yamada et al. (2011) mention the positive effect of wealth and health perception on life satisfaction. This implies that an increase in wealth, health and education can increase life satisfaction.
Another highlighted relationship is the one between socioeconomic development and individualism. Individualism enables innovative activity by allowing people to fully express their creative minds and create competition amongst individuals that fosters innovative thinking (Taylor and Wilson, 2012). These notions suggest that individualism is related to the level of progress and thereby, socioeconomic development.
which one grows up determines their priorities and cultural values. Gaygisiz (2013) contributes to these statements by indicating that globalization indeed influences cultural values such as indulgence vs. restraint by correlating positively with the World Governance Indicators, thereby stating that a higher level of indulgence will lead to more voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption, which are indicators of societies that are perceived as ‘developed’.
It is evident that there is a relationship between socioeconomic development and individual’s cultural values. Globalization has brought changes to societies all around the world by increasing exposure to different cultural values as well as bringing new standards, increasing levels of development and changing social situations. It is therefore expected that socioeconomic development explains the differences between the globalized generation and the unglobalized generation. In this research, the focus will lie on four Hofstede dimensions in order to examine differences between the two generations and the possible explanatory effect of socioeconomic development. These cultural values are power distance, uncertainty avoidance, individualism and indulgence. The next section will describe these cultural values, as well as the hypotheses supporting the expectations.
3. HYPOTHESES DEVELOPMENT AND THEORETICAL FRAMEWORK
3.1 Cultural Change and Generational Differences
One of the main objections on research into cultural values is that they are assumed to be stable over time (Fang, 2005). However, current research has suggested that cultural values do change, but very slowly. Following Inglehart (1997) on the notion of materialism and postmaterialism, it is clear that conditions at the time of birth determine the priorities set by individuals in life. Baddeley (2006) notes that globalization is one of the main drivers of changing cultural values. Robertson et al. (2012) suggests, that differences can be attributed to generational subculture theory, which indicates that a group identity is shaped during the impressionable years of an individual’s life and that this identity remain stable over time. It is therefore likely that individuals who have been exposed to globalization have different values from individuals who have not been exposed to globalization. The first hypotheses easily flows from this:
Hypotheses 1: The globalized generation and the unglobalized generation have different cultural values.
correlation is found (for the degree to which the respondent agrees that a wife must always obey her husband). This item does not contradict the meaning of the Hofstede dimension, but scores on this item can only be found for 12 countries and therefore, Beugelsdijk et al. (2015) decide to exclude this dimension from their replication. This approach will be followed in this paper and no reconstruction of this dimension will be used here.
The exclusion of masculinity and long-term orientation from the following of this paper means that the analysis will consist of four replicated dimensions. Therefore, hypotheses will be formulated for these four dimensions.
Power distance is explained as the degree to which power in a society is distributed unequally and the degree to which this is accepted (Hofstede, 1980). This not only involves determining who is the most powerful person, but it deals with the extent to which employees feel the need to obey this person. Individuals high in power distance feel that authority figures are the ‘real’ boss and that these people should be respected and obeyed. Employees high in power distance are therefore expected to trust their superiors and be loyal and submissive towards them. Individuals who are low in power distance on the other hand, do not perceive distinctions based on authority positions or status. Power distance therefore influences the extent of centralized decision-making and formal hierarchy within an organization (Daniels and Greguras, 2014).
Diener et al. (1995) mention that high power distance is an inhibitor of fast social progress. As globalization is expected to change people’s values due to bringing new values and bringing progress to the economic situation, it is expected that people who are exposed to globalization are capable of deviating from strict social standards and hierarchies in order to embrace the globalization, thereby having a lower acceptance of power inequalities. Additionally, globalization is argued to distribute Western values across the globe due to the implementation of Western practices and standards. In general, developed, Western countries have lower levels of power distance. This suggests that individuals born in a globalized world are used to lower power distance societies than individuals who were not exposed to globalization and the Western values that come with this movement.
Individualism vs. collectivism is the dimension developed by Hofstede (1980), concerning itself with the relationship between individuals and the group. It determines the extent to which individuals are autonomous and to what extent they are part of a group. Individualistic individuals mainly focus on individual achievements and personal accomplishments. Collectivistic individuals on the other hand, value group loyalty and the development of close relationships in communal ways, both inside and outside the family circle (Browaeys and Price, 2008).
Pinillos and Reyes (2011) explain that the level of individualism is higher in more developed countries and lower in poorer countries. Globalization transfers the values of Western, more developed countries, to the other countries in the world, as mentioned by Beugelsdijk et al. (2015). Additionally, Gupta and Govindarajan (2000) state that these values include democracy, free markets, diversity, freedom of choice and individual rights. This implies that the globalized generation has been exposed to higher levels of individualism than the unglobalized generation. Therefore, it is expected that:
Hypotheses 1b: Individualism is higher for the globalized generation then for the unglobalized generation.
Uncertainty avoidance deals with the extent to which unstructured and surprising situations will make people feel uncomfortable (Hofstede, 1980). Just like power distance, the dimension of uncertainty avoidance is perceived as one of the most important cultural dimensions and is mentioned not only in Hofstede’s dimensions, but also in the work of Robert House (2002). Individuals with a high level of uncertainty avoidance like to be aware of what is expected from them and how these expectations need to be fulfilled (Zhang and Zhou, 2014). They feel a need for highly defined and structured situations in which everything is explicitly stated.
they are used to a dynamic environment in which new signals are common. It is therefore expected that:
Hypotheses 1c: Uncertainty Avoidance is lower for the globalized generation than for the unglobalized generation.
The final dimension in the Hofstede index is fairly new and developed in 2010 and measures ‘the extent to which people try to control their desires and impulses’ (Hofstede & Minkov, 2010). Indulgence and restraint are two extremes of a spectrum of allowance of freely expressing natural human drives related to enjoying life and having fun, whereas indulgence desire gratification of these drives, restraint individuals follow strict norms that inhibit such gratifications (Griffith & Rubera, 2014).
As shown by Gaygisiz (2013), indulgence vs. restraint is positively correlated to the Human Development Index, which consists of having a long and healthy life, having access to knowledge and having a decent standard of living, measured by using health, education and income respectively. Additionally, a higher level of indulgence will lead to more voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption (Gaygisiz, 2013). As mentioned by Gupta and Govindarajan (2000), these are the Western values that spread throughout the world due to globalization. It is therefore likely that exposure to globalization will result in higher levels of indulgence (and lower levels of restraint):
Hypotheses 1d: Indulgence is higher for globalized generation than for unglobalized generation.
3.2 The moderating effect of socioeconomic development.
either a top-down or a bottom-up process. In this way, it is argued that changes in the global culture lead to changes in a country’s national culture. This top-down perspective can, according to Leung et al. (2005) be applied to the notion by Inglehart (1997), describing that changes in the socioeconomic situation, which are caused by globalization as explained earlier, can lead to a shift in cultural values on the national level.
Being raised in a certain socioeconomic situation therefore determines your mindset and the values that come with this mindset. However, as soon as the situation in a country changes, individuals born in a different environment will create a different mindset, this implies the influence of the socioeconomic situation on the cultural values. It is therefore expected that:
Figure 2: Levels of Culture (Erez and Gati, 2004)
4. DATA
4.1 Sample
In this research, a number of different data sources will be used. Data from the World Values Survey (WVS) will be used to measure cultural values and generational differences. The latest wave of the WVS data (Wave 6) was collected from 2010-2014. A longitudinal data file will be used, which is an integrated file of all World Values Survey waves (1-6). In each wave, different respondents are used. Using the longitudinal file including Wave 6 will therefore result in the highest number of respondents. The World Value Survey collects data at the individual level. In this thesis, the respondents will be grouped in generational cohorts according to their year of birth. Then, country averages on the items used will be calculated and country scores will be calculated for each cohort.
In order to measure economic development, data from Clio Infra will be used. Clio Infra is a set of interconnected databases, collecting information on economic, social and institutional indicators for over 500 years. The main focus of Clio Infra is to collect information on indicators that might explain varying levels of development in the world, which complements this thesis. The Clio Infra database includes information on most countries covered in the World Values Survey database. However, since not all countries have always existed (the split of the Russian Federation etc.), not all countries are covered in the Clio Infra database. Therefore, not all information can be collected for all countries, which results in a smaller sample of countries than calculated from the World Values Survey Data.
4.2 Operationalization of the variables 4.2.1 Generations
generations instead of simple age as a measure has a main benefit. While someone born in 1980 may not act absolutely different from one born in 1981, studies from psychology argue that differences between individuals can be attributed to the mean cohort level. This allows us to make generalizations and predictions about certain tendencies of individuals (Lamm & Meeks, 2009). Therefore, in this paper, generations will be used.
This results in the following two cohorts: Generation Cohort Number of
respondents
Percentage Minimum Maximum Mean
Unglobalized 0 108797 46,8% 1886 1952 1940
Globalized 1 123759 53,2% 1970 1999 1980
Total 232556 100% 1886 1999 1962
Table 1: Cohort Structure 4.2.2 Cultural Values
reached the level of influence of Hofstede’s model. Leung et al. (2005) refer to Hofstede’s model as ‘classic’ and mention that his work has revolutionized research in both the fields of culture and International Business. Based on these notion and the fact that extensive research has been done using the work of Hofstede, such as developing an index of cultural distance (Kogut and Singh, 1988), the dimensions of developed by Hofstede will be used in this research.
In order to measure cultural values, we follow the method develop by Beugelsdijk et al. (2015). They reconstructed most of Hofstede’s dimensions based on the items of the WVS, because reconstructing a survey the size of that of Hofstede did is a too expensive activity at this time. Based on a reconstruction of the original sample used by Hofstede for his IBM study, Beugelsdijk et al. (2015) are able to develop a set of items correlating highly to the original Hofstede dimensions. Using this method, Beugelsdijk et al. (2015) are able to replicate three out of the four Hofstede dimensions.
Power distance will be reconstructed using three WVS items. The first one is the materialist-postmaterialist index, as constructed by Inglehart (1971). This index is based on four WVS items: The preference for maintaining order in the nation, giving people more to say, fighting rising prices and protecting freedom of speech. The second item to indicate power distance is the extent to which one feels that they should always love and respect one’s parents. The last item is agreement with the statement ‘when jobs are scarce, employers should give priority to people of this country over immigrants’. The replicated dimension of power distance has a Cronbach’s Alpha of 0.83 and correlates 0.72 with Hofstede’s original dimension (Beugelsdijk et al. 2015). The following formula is provided by Beugelsdijk et al. (2015) to calculate country scores on the dimension of power distance:
PD: 134 + 60.4 x (country mean item 2 - country mean item 1 - country mean item 3)
which people in a country find homosexuality justifiable (Beugelsdijk et al. 2015). The last item concern the degree to which people feel that private ownership of business should be increased. The replicated dimension of individualism has a Cronbach’s Alpha of 0.81 and correlates 0.77 with Hofstede’s original dimension. The following formula is provided by Beugelsdijk et al. (2015) to calculate country scores on the dimension of individualism: IVD: 17.3 + 7 x (country mean item 1 - country mean item 2 + country mean item 3 + country mean item 4)
In this thesis, individualism has been replicated for 97 countries for Cohort 0 with a Cronbach’s Alpha of 0,715 and for 96 countries for Cohort 1 with a Cronbach’s Alpha of 0,757.
Uncertainty avoidance is reconstructed using four items from the WVS dataset. The first item measures the lack of trust in other people. The second and third item measures the confidence that people in a country have in political parties and the justice system and the fourth item measures the degree to which people think the country is run by few big interests vs. for all people’s benefit. The replicated dimension has a Crohnbach’s Alpha of 0.76 and correlates 0.67 with Hofstede’s original dimension. The following formula is provided by Beugelsdijk et al. (2015) to calculate country scores on the dimension of uncertainty avoidance
UA: - 11.3 + 21.67 x (country mean item 1 + country mean item 2 - country mean item 3 - country mean item 4)
In this thesis, uncertainty avoidance has been replicated for 65 countries for Cohort 0 with a Cronbach’s Alpha of 0,770 and for 63 countries for Cohort 1 with a Cronbach’s Alpha of 0,797.
scores on the dimension of indulgence vs. restraint:
IND: 6.79 + 18.4 x (country mean item 3 - country mean item 1 - country mean item 2)
In this thesis, indulgence has been replicated for 99 countries for Cohort 0 with a Cronbach’s Alpha of 0,687 and for 98 countries for Cohort 0 with a Cronbach’s Alpha of 0,623.
All reliability scores exceed the minimum threshold of 0.6 for reliability. Since this scale is in its early development, there is no need to increase the value of alpha by deleting any items. This would lead to a change in the formulas for cultural scores, which is beyond the scope of this research.
4.2.3 Socioeconomic development
No universal measure of socio-economic development exists. However, a number of variables are generally accepted as indicators of economic growth. Capital, labor and human capital are referred to as the ‘basic indicators’ of economic development and will therefore be used in this paper (Petrakis and Kostis, 2013). While GDP is seen as the main indicator of economic development, Takamori and Yamashita (1973) argue that economic development cannot be regarded as the only indicator of socioeconomic development, due to its interconnectedness with social, cultural and political factors. It is therefore important to not only measure the level of purely economic development, but to also use other indicators for socioeconomic development. Human capital increases peoples’ level of knowledge and information. Moreover, it allows people to be more productive and contribute to higher levels of diversified assets and activities (Burchi, 2006). Little information regarding working populations in the early 1900s is available. Therefore only two of the basic indicators will be used, capital and human capital. Respectively this is translated into the level of GDP per capital and the average years of education.
cohort, the average GDP per capita for that birth cohort will be used. Scores for GDP are country averages based on the cohort years. Not for all countries in the data file, information regarding GDP for the unglobalized generation is available, which will reduce the number of countries to be used in the analysis. For some countries no data is available for the first years of the cohort. A minimum limit has been set a 1920, since the number of respondents in the cohorts starts to increase from 1920 on. When data is available only later than 1920, no GDP averages for the unglobalized generation can be computed, which excludes the country from the analysis
Human capital in a country will be measured using the indicator Average years of Education.
This indicator measures the rate of the children who should be enrolled in primary school that actually completes this. Education is a known indicator of economic development, since children who are able to complete school are usually more developed since they do not have to work in order to provide for their families. The average years of education are measured by Clio Infra every 10 years. Cohort averages for the globalized generation are based on the measures from 1900-1950. Those for the globalized generation are the ones from 1970-2000. Once more some countries had limited data available for cohort 0. In order to construct a sufficient number of countries for which data is available, data has to be available from 1930 on. If not, the country is excluded from the analysis.
Additionally a third measure is added: ‘Life expectancy at birth’ which indicates the expected age that one will reach at the time they are born (World Bank, 2015). The possibility of getting older is related to higher health and safety levels, which are related to higher levels of development. This indicator is especially important in developing countries, since the level of life expectancy in develop countries in generally much higher and fairly stable (Mondal and Shitan, 2014). Life expectancy is measured yearly for some countries and periodically for others. Life expectancy faced the same problem as GDP, limited data availability for a number of countries for the unglobalized generation. 1930 Was once again set as a minimum for availability.
Dimension Power Distance Individualism Uncertainty Avoidance Indulgence Number of countries 26 33 27 34
Table 2: Sample Sizes per Dimension
4.2.4 Control Variables
Since it unlikely that the changes in cultural values can be solely attributed to changes in economic development, a number of variables will be controlled for that can contribute to the changes in values.
The first control variable will be continent. Countries in continents can be clustered in groups due to their geographic proximity, mass migrations and social capital and their religious and linguistic commonalities (Gupta et al. 2002). Since globalization is argued to transfer Western values across the globe, it is likely that non-Western cultures are more heavily impacted by globalization. A continental-based control will therefore be conducted. Continent Dummy’s will therefore be constructed for all continents (North America, South America, Asia and Europe). No countries from Africa are included in the analysis, so no dummy will be constructed for that continent. No dummy for Oceania will be constructed, since this continent will serve as the reference country in the analysis. Data on continents will be gathered from the United Nations.
towards a more democratic regime a possible facilitator of cultural change. The influence of the political regime will be measured by examining whether the political regime at the end of the first cohort (1952) differs from that at the beginning of the second cohort (1970). If so, these changes might have caused changes in values.
4.3 Estimated Model
In order to test the hypotheses, a regression model is constructed introducing the variables. As mentioned earlier, it is assumed that the differences between generations in cultural values are explained by changes in their levels of socioeconomic development. This means that the generational cohort, the socioeconomic development index, the interaction effects between cohort and socioeconomic development, as well as the control variables continent and political regime influence the score on a cultural dimension.
This results in four estimated regression models, one for each dimension:
PD = β + β GEN + β GDP + β LIFE + β EDU + β CO*GDP + β CO*LIFE + β CO*EDU + βAsia + βEurope + βNorthAmerica + βSouthAmerica + βPoliticalRegime + ε
IND = β + β GEN + β GDP + β LIFE + β EDU + β CO*GDP + β CO*LIFE + β CO*EDU + βAsia + βEurope + βNorthAmerica + βSouthAmerica + βPoliticalRegime + ε
UA = β + β GEN + β GDP + β LIFE + β EDU + β CO*GDP + β CO*LIFE + β CO*EDU + βAsia + βEurope + βNorthAmerica + βSouthAmerica + βPoliticalRegime + ε
5. MULTIPLE REGRESSION ANALYSIS
This thesis aims to analyze the mentioned relationships with a multiple regression analysis. According to Osborne and Waters (2002) there are a number of assumptions that need to be checked before conducting multiple regression analysis. When these assumptions are not properly discussed, over- or underestimation of the significance of certain effects can occur, called Type I or Type II error (Osborne & Waters, 2002). Laerd (2015) mentions another assumption that needs to be tested for, which is multicollinearity.
5.1 Linearity
When conducting a multiple linear regression, one of the main assumptions is a linear relationship between the IVs and the DV. According to Osborne and Waters (2002) the most preferred method of examining linearity is by examining the plots. Laerd (2015) mentions the creation of scatterplots between the IV and the DV for both categorical variables, in this thesis cohort 0 and cohort 1.
In this paper, linearity is assessed between all IVs (GDP, Education, Life expectancy) and all DVs (Power distance, individualism, uncertainty voidance and indulgence). Scatter plots are visually inspected. After examining the scatterplots, linearity can be assumed for all IVs and power distance, individualism and indulgence. Uncertainty avoidance shows no linearity with the independent variables. This however, can also mean that there is no significant relationship between this DV and the IVs.
5.2 Sample Size
5.3 Outliers
Outliers can severely influence a multiple regression analysis. This dataset has been visually inspected for outliers. Outliers are identified using SPSS. For all variables, 5 cases were identified to be extreme highs and 5 cases were identified as extreme lows. While these cases were the highest and lowest in the range, no impossible values were found based on the formulas for the dimension scores. Country dimension scores did not range from 0 to 100, as in the original sample developed by Beugelsdijk et al. (2015), since the formula has been develop based on this sample and country scores differ when the sample changes. As for the independent variables, no significant outliers were found for average years of education and life expectancy. The GDP boxplot shows outliers, in the higher range. However these outliers are GDP levels of developed countries, which in general have a higher level of GDP. It can therefore be said that no complicating outliers can be found and no cases have to be erased from the sample.
5.4 Correlations
Table 2 shows the correlations between the independent variables. While correlations between the IVs are undesirable, since they inhibit the explanation of effects of IVs on the DV, correlations between the IVs and the DVs are desirable. As becomes clear in the correlation table, GDP, Average Years of Education and Life Expectance have a high and significant correlation. Correlations between GDP and Average Years of Education and Life Expectancy are 0,402 and 0,746 respectively and are both significant at the 0.05 level. According to Pallant (2010), correlations of 0.8 or higher between IVs can result in the inability to explain which independent variable caused changes in the dependent variable.
Table 3: Correlations Matrix Correlation Matrix Mean 1 2 3 4 5 6 7 8 1 Cohort 0,5 1 2 GDP 5947,67 0,594** 1 3 EDU 8,2 0 0,402** 1 4 LIFE 60,32 0,79** 0,746** 0,099 1 5 AsiaDummy 0,21 0 -0,300* -0,005 -0,255* 1 6 EuropeDummy 0,41 0 0,287* 0,212 0,334** -0,426** 1 7 NorthAmericaDummy 0,06 0 0,290* 0,029 0,146 -0,127 -0,209 1 8 SouthAmericaDummy 0,29 0 -0,249* -0,274 -0,261* -0.329** -0,504** -0,161 1 9 Political Regime 2,57 0,07 0,465* 0,069 0,264* -0,005 0,133 0,251* -0,334**
5.5 Multicollinearity
Multicollinearity occurs when independent variables yield significant effects when they are regressed on one another. It constitutes the problem created by high correlations between independent variables, which is the inability to explain changes in the dependent variable. Collinear variables provide similar information. In order to check for multicollinearity the Tolerance and Variance Inflation Factor can be assessed. Multicollinearity is not a problem when the Tolerance is > 0.1 and the VIF < 10 (Pallant, 2010). When Tolerance and VIF scores are inaccurate, multicollinearity exists. According to Frost (2013), multicollinearity is not always a problem. Multicollinearity does not change the overall fit of the model. Allison (2012) support this notion, stating that multicollinearity caused by the inclusion of interaction variables is not a problem, since there is a good chance that the interacted variables correlate highly with the interaction term. In their paper on multicollinearity problems in moderation analysis, Disatnik and Sivan (2014) explain that multicollinearity indeed does not create a problem when it occurs between an independent variable and a moderation term.
6. EMPIRICAL RESULTS
This chapter provides the results of the multiple regression analysis. A multiple regression model was tested in order to examine whether the cultural dimension scores depend on generation and socioeconomic development. After standardizing the dependent and independent variables, the independent variables and the control were entered into a regression model. In Model I, the cohort effect is measured. In Model II, the socioeconomic indicators are added. In Model III, the interaction effects between the socioeconomic indicators and the cohort are added, whereas in model IV, the control variables are added. Sections one through four respectively discuss the results for each dimension of national culture.
6.1 Power Distance
From Model I it can be concluded that cohort has a significant influence on the level of power distance (β=-0.575, p<.05), implying that the level of power distance is lower for the globalized then for the unglobalized generation. These results support Hypotheses 1a. Moreover, the adjusted R-square for this model is 0,084, which means that 8,4% of the variance in power distance is explained by cohort differences. In Model II, GDP (β=-0.755, p<.01), Life Expectancy at Birth (β=0.213, p<.1) and Average Years of Education (β=0.398, p<.05) are added to the model. All three are significant indicators of power distance scores and contribute to a significant change in the overall model fit (ΔR²=0.507, F=19.441, p=0.000), which now explains 59,1% of the variance in power distance.
Table 4: Power Distance Model
6.2 Individualism
Model I shows a significant effect of both the constant (β =-0.283, p <.1) and the cohort effect (β=0.567, p <.05), indicating a rise in individualism scores for the globalized generation compared to the unglobalized generation. This supports Hypotheses 1b. Model I explains 8.1% of the variance in individualism. Model II shows the effects of adding the socioeconomic indicators to the model. GDP (β=0.584, p = 0.000) and Life Expectancy at Birth (β=0.699, p = 0.000) turn out to significantly influence individualism levels. The overall model fit has changed significantly (ΔR²=0.636, F=45.703, p = 0.000) and now explains 71,7% of the variance in individualism.
In Model III, Hypotheses 2 is tested by adding the interaction effects to the model. The interaction effect between average years of education and cohort turn out to be a significant indicator of individualism scores (β=0.649, p<.05), indicating that average years of education explains generational differences in power distance. This shows partial support for hypotheses
Dependent Variable Power Distance
Model I Model II Model III Model IV
Constant 0,287 - 0,496*** -0.661* -0.398 Cohort -0,575** 0.925*** 0.990 0.655 GDP -0.755*** -1.203* -0.305 EDU 0.213* 0.148 0.126 LIFE -0.398** -0.305 -0.281 GDP_CO 0.494 0.137 EDU_CO -0.027 -0.119 LIFE_CO 0.040 -0.512 Region: Asia 0.496 Region: Europe 0.144
Region: North America 0.368
Region: South America 0.045
Political Regime -0.488***
R² 0.084 0.591 0.596 0.724
R² - Change 0.084 0.507 0.005 0.128
F-Statistic 4.593 19.441 0.179 3.624
Significance 0.037 0.000 0.910 0.009
2. The inclusion of the interaction effect to the model changes to overall model significantly (ΔR²=0.037, F=2.877, p <.05). Control variables are added in Model IV and equal to power distance, political regime significantly influences the level of individualism (β=0.182, p<.05). The overall model explains 81.1% of the variance in individualism scores.
Table 5: Individualism Model
6.3 Uncertainty Avoidance
Uncertainty Avoidance turns out not to be significantly influenced by neither a constant nor a cohort effect, as shown in Model I, thereby rejecting Hypotheses 1c. There are no significant differences in uncertainty avoidance scores between the two generations. Adding the socioeconomic indicators in Model II does not result in significant changes in the level of uncertainty avoidance, nor does adding the interaction effects, as shown in Model III. The overall models turn out to be insignificant. No support is found for hypotheses 2 on the dimension of uncertainty avoidance. Adding the control variables in Model IV results in a significant change in the overall model fit (ΔR²=0.389, F=6.410, p = 0.000), explaining
Dependent Variable Individualism
Model I Model II Model III Model IV
Constant -0.283* 0.619*** 0.577** 0.338 Cohort 0.567** -1.199*** -0.845* -0.599 GDP 0.584*** 0.620 0.508 EDU 0.003 -0.109 -0.113 LIFE 0.699*** 0.618*** 0.431** GDP_CO -0.449 -0.450 EDU_CO 0.649** 0.587** LIFE_CO -0.006 0.180 Region: Asia 0.002 Region: Europe 0.284
Region: North America -0.230
Region: South America -0.183
Political Regime 0.182**
R² 0.081 0.717 0.754 0.811
R² - Change 0.081 0.636 0.037 0.057
F-Statistic 5.677 45.703 2.877 3.197
Significance 0.020 0.000 0.044 0.014
50,2% of the variance in uncertainty avoidance. However, none of the control variables shows a significant effect.
Table 6: Uncertainty Avoidance Model
6.4 Indulgence
As shown in Model I, both the constant (β=-0.304, p<.1) and the cohort effect (β=0.608, p <.05), significantly influence the level of indulgence, whereas this level is higher for the globalized than it is for the unglobalized generation. This supports hypotheses 1d. Model I explains 9,4% of the variance in the level of indulgence. Adding the socioeconomic indicators in Model II changes this level significantly (ΔR²=0.178, F=5.135, p <.01), now explaining 27,2% of the variance in indulgence. This significant change is caused by the inclusion of GDP (β=0.623, p<.01) and Average Years of Education (β=-0.343, p<.01).
Dependent Variable Uncertainty Avoidance
Model I Model II Model III Model IV
Constant -0.095 -0.181 -0.510 -0.282 Cohort 0.190 0.364 0.693 1.241 GDP -0.189 -0.849 -0.827 EDU -0.162 -0.361 -0.175 LIFE 0.022 0.093 0.176 GDP_CO 0.350 0.446 EDU_CO 0.534 0.629 LIFE_CO 0.131 -0.775 Region: Asia -1.034 Region: Europe -0.344
Region: North America -0.119
Region: South America 0.573
Political Regime -0.171
R² 0.009 0.080 0.112 0.502
R² - Change 0.009 0.071 0.032 0.389
F-Statistic 0.485 1.264 0.551 6.410
Significance 0.489 0.297 0.650 0.000
