THE IMPACT OF CULTURE ON SELF-EMPLOYMENT

40

THE IMPACT OF CULTURE

ON SELF-EMPLOYMENT

3

Second-generation immigrants propensity to be

self-employed, isolating the role of culture by

means of using the epidemiologic approach

FINAL VERSION

Author

Wieb Mulder

s1932241

w.mulder.6@student.rug.nl

0031642012009

Master 1: Small Business and Entrepreneurship

Master 2: International Business and Management

West-Indischekade 29, Groningen

University

University of Groningen

Faculty of Economics and Business

Thesis supervisor Co-assessor

Ms. Dr. M.J. Klasing Ms. Dr. Ir. H. Zhou

Date

January 20

, 2015

Word count

4

ABSTRACT

This study was set out to explore the possible influence of culture on career choices. There exists a high level of unexplained variation in career decisions of individuals coming from different cultural backgrounds, as the traditional literature, involving demographics and traits, has come to a standstill. For as far as known, this study is the first in its kind that tried to isolate the role of culture from the effects of strictly economic factors and institutions. It does so by using a sample of 1837 second-generation immigrants throughout the United States, coming from 29 different countries, over a time period from 1977 until 2012. By merging several databases and by means of using logistic regressions, four cultural dimensions of Hofstede and the level of trust are tested on career choice, while controlling for other variables. Although it has been found that differences between several immigrant groups in the United States exist with regard to their propensity to be self-employed, this difference could not be attributed to the cultural values, presumably because of a small sample size and problems with regard to the amount and magnitude of culture which has been transmitted from parents to child. This study provides more possible explanations for the insignificant findings and further names several future research directions, which could lead to significant results in the near future.

5

ACKNOWLEDGEMENTS

I am using this opportunity to express my gratitude to everyone who supported me throughout the course of this Master thesis project. Although I did not consider myself as a researcher, I was confident at the start of this thesis, because I was always interested in both culture as well as entrepreneurship, which made this subject very appealing to me. Moreover, I had undertaken some quantitative studies before and therefore acquired previous experience with statistical methods. Now, almost six month later, I can conclude that it was much more complicated than initially expected. Especially the fact that you are on your own in the decision making process was something more difficult than imagined. Furthermore, the amount of small decisions I had to take was amazingly high, and transcended what I expected by a large distance, which led to an increasing respect for the profession of the researcher. Last but not least, I always considered myself as a disciplined individual, but throughout the course of writing this Master thesis I can say that this has improved a lot.

First of all, I want to express my warm thanks to my friends, Jan Hessel Veurink and Eizo ter Veer, for their support, valuable suggestions, and guidance at difficult times.

Moreover, I would like to thank my parents for their tremendous confidence in me, for their support and encouragement throughout my entire study, but even more for their sobriety, which manifested itself by asking what topic my thesis involved almost every single month. In particular, I would like to thank Rosa Blansjaar, for listening, thinking about suggestions, and for sharing here view on a number of issues related to the project. Even more so, I want to thank her for being my safe haven, for her patience at times I most needed it, and for her constant encouragement throughout the past months.

I would like to express my special thanks of gratitude to my supervisor Mariko Klasing, as well as my co-assessor Haibo Zhou, as both of them provided me the golden opportunity to write a thesis about this interesting project. I would also like to thank professor Haibo Zhou for her aspiring guidance, invaluably constructive criticism and friendly advice during the project work. I am particularly grateful for the assistance given by professor Mariko Klasing, as her patient guidance, enthusiastic encouragement and useful critiques have been absolutely critical for the end result.

I consider myself as a very lucky individual, as I was provided with this unique opportunity, and sincerely hope that this theme suits you, the reader, just as well as it did to me.

6

TABLE OF CONTENTS

I. INTRODUCTION ... 8

II. THEORETICAL UNDERPINNINGS ... 11

2.1 Theory of Planned Behavior (TPB) ... 11

2.2 Entrepreneurial intention and antecedents ... 12

2.2.1 Individual domain ... 13 2.2.1.1 Demographics ... 13 2.2.1.2 Traits ... 14 2.2.2 Contextual domain ... 15 2.3 Career choice ... 15 2.4 Research gap ... 16

III. CULTURE AND THE DEVELOPMENT OF HYPOTHESES ... 18

3.1 Power distance (POD) ... 19

3.2 Individualism versus Collectivism (IDV) ... 20

3.3 Masculinity (MAS) ... 21

3.4 Uncertainty avoidance (UNA) ... 22

3.5 Trust (TRU) ... 23

IV. METHODOLOGY ... 24

4.1 Data sources ... 24

4.1.1 General Social Survey (GSS) ... 24

4.1.1.1 Sample ... 25

4.1.1.2 Control variables ... 26

4.1.2 Cultural variables ... 27

4.2 Methodological approach ... 28

4.2.1 Data analysis ... 29

V. ANALYSIS AND RESULTS ... 30

5.1 Descriptive statistics ... 30

5.2 Different self-employment levels? ... 32

5.2.1 Absolute differences ... 32

5.2.2 Relative differences... 32

5.3 Explaining differences in self-employment levels ... 34

5.4 Robustness checks ... 40

VI. DISCUSSION ... 44

VII. LIMITATIONS & FUTURE RESEARCH ... 47

VIII. CONCLUSIONS ... 49

REFERENCES ... 50

7

LIST OF TABLES

1: Cultural dimension and their definition 27

8

I. INTRODUCTION

The buzz of entrepreneurship is all around us. The entrepreneur is the baker who bakes bread every morning, but also the small social media company sprouting from nowhere to impact our personal and professional life. Entrepreneurship is often cited as one of the most effective ways to relocate labor and capital in an economy (Luthans, Stajkovic, & Ibrayeva, 2000). University courses, books and academic journals on entrepreneurship exist in abundance (Hébert & Link, 2009), making it one of the most widely cited in the management discipline (Bruton, Ahlstrom, & Obloj, 2008). Scholars recognize entrepreneurs “as engines of growth in the economy” (Dutta & Crossan, 2005). Thus, because entrepreneurs excel in finding and combining resources in new ways, they can be identified as being critical to the long-term health and prosperity of the economy. An entrepreneur is at the same time one of the most fascinating, as one of the most elusive characters in economic analysis. This is what makes entrepreneurship slippery and hard to define, reason why it is not a field known for its consensus. Instead it is argued that entrepreneurship is a multidimensional concept, in which using different standards to measure it can yield radically different results (Wennekers & Uhlaner, 2002).

Entrepreneurs are often equated with the employed, on the grounds that the self-employed fulfill the entrepreneurial function of being risk-bearing residual claimants (Parker, 2004). Some scholars have disputed this assertion, naming it too broad. When considering entrepreneurs as individuals “who create goods and services which are currently non-existent within the market place” (Kruger, 2005) and the self-employed as individuals “who earn no wage or salary but who derive their income by exercising conducting a profession or business on their own account and at their own risk” (Parker, 2004), some problems of this substitution become clear. A striking example is that when self-employed individuals only imitate and replicate, they are not performing true entrepreneurial functions. Furthermore, a problem can occur when someone can be named an employee or a self-employed. As Wennekers and Uhlaner (2002) advocated, the definition of entrepreneurship depends largely on the focus of the research undertaken. On the grounds of Parker, and as most consistent data covering entrepreneurship looks at individual self-employment, this research will equate the self-employed with the entrepreneur, while acknowledging that they are imperfect substitutes. The definition of Parker (2004), which is the definition of self-employment that will be followed in the remainder of this paper, suggests that each individual is faced with two alternatives when choosing a career, either as self-employed or employed in an organization. This choice of employment status was defined by Katz (1992) as “the vocational decision process in terms of the individual's decision to enter an occupation as a wage or salaried individual or as a self-employed one”.

9

1988), such as starting a new business, is more likely to choose self-employment as career. Kolvereid and Isaksen (2006) argued that entrepreneurial intentions are a decisive factor for performing actual entrepreneurial behavior. Ajzen and Fishbein (1980) further stated that intentions provide critical insights into behavioral processes, and robustly predict and explain this deliberate and planned behavior. For most of the current self-employed individuals, the actual career choice therefore descended from entrepreneurial intention earlier on.

Only a small percentage of the working population typically engages in entrepreneurship and thus chooses self-employment over working for an organization (Bosma, Jones, Autio, & Levie, 2008). It is widely known that entrepreneurial activity, for instance expressed by the percentage of managers of incorporated and unincorporated businesses relative to the labor force, differs strongly across countries (van Stel, Carree, & Thurik, 2005). This variation is related to differences in levels of economic development, but can also be accounted to diverging demographic, cultural and institutional characteristics (Blanchflower, 2000; Wennekers, 2006). One of the growing interests within the research field is cross-national entrepreneurship (Hayton, George, & Zahra 2002), in which scholars tried to grasp differences in self-employment levels between different countries and their often distinctive cultures.

In previous research the importance of the cultural context for career choices was argued (e.g. Brown, 2002; Moriano, Gorgievski, Laguna, Stephan, and Zarafshani, 2011). The paper at hand contributes to the growing literature stream involving the importance of cultural values in determining economic outcomes (e.g. Alesina & Giuliano, 2013). This research forms part of a rapidly growing literature stream that has emphasized the role of cultural identity and cultural variation among diverse cultural groups within one country (Leong, 2010), as well as considering that country differences exist in the decision to pursue a career in management (Malach-Pines & Kaspi-Baruch, 2008). The main challenge of any cultural analysis is how to separate the effects of culture from the effects of strictly economic factors and institutions (Fernández, 2010). This separation is very challenging to achieve, which has prevented economists to grasp the possible influence of culture on career choices. The paper at hand will make this critical distinction in an attempt to tackle culture, in order to show in a quantitative fashion, that culture matters.

10

isolates the role of culture. The idea is that individuals from different cultural backgrounds will take different actions despite facing an identical environment, possibly through their cultural backgrounds. This idea of studying second-generation immigrants is called the epidemiological approach, which goes back to the seminal work by Carol, Rhee, and Rhee (1994).

According to Vargas (2005), the large immigrant population in the U.S. can be attributed to globalization, wars, and political problems, which are all likely to convey part of the truth. The typical immigrant therefore mostly migrates from less-developed countries to developed countries in the quest of a better and more stable future. The growing presence of first generation immigrants in the U.S. has logically led to a growth in the percentage of second-generation immigrants. With the major blend of immigrants within the U.S., this is the most obvious country to examine differences in career choice between cultures.

When considering the global entrepreneurial boom that has taken place in the past decades, with the creation of thousands of new companies, it is interesting to consider why individuals from different cultures, make different career choices in comparison to others. In order to get a clearer picture to why this is the case, this paper will relate four cultural dimensions of Hofstede and the concept of trust to second-generation immigrants in the U.S. and their career choices. When significant discrepancies between different cultural backgrounds will be found, this will lead to an improved understanding of the underlying reasons for the career choice of individuals. With the vast amount of research about the crucial role that entrepreneurs have in the long-term prosperity of a country, this knowledge could contribute to the plans of policy-makers in striving to reap the major benefits of entrepreneurship for any economy in the near future. The true contribution of this paper can be found mainly from a technical point-of-view, as the influence of culture on career choices has, for as far as known, never been studied by means of using the epidemiological approach, thereby isolating the role of culture. This leads to the following research question:

To what extent does culture influence and explain the differences in career choice between second-generation immigrants in the United States?

11

II. THEORETICAL UNDERPINNINGS

The aim of this chapter is to provide insight into the complex concepts of career choice, entrepreneurial intention, culture and trust, in order to gain sustainable knowledge essential for the empirical part of this thesis. This research explores the propensity to become self-employed, while drawing on the Theory of Planned Behavior (TPB). This is one of the best established theories to predict intentions, including career intentions. In the psychological literature, intentions have proven the best predictor of planned behavior particularly when that behavior is rare, hard to observe, or involves unpredictable time lags, which is the case for career choices.

2.1 Theory of Planned Behavior (TPB)

This research embraces the Theory of Planned Behavior developed by Ajzen (1991), which looks at personal and social factors in order to explain intentional behaviors, including entrepreneurial intentions. The TPB is an important socio-cognitive theory that has been successfully applied in a wide variety of fields (Beck & Ajzen, 1991). The theory helps us to understand how we can change the behavior of people by predicting deliberate behavior.

Intentions are the single best predictor of any planned behavior and is both dependent on the situation as well as on the person. Understanding the antecedents of intentions increases our understanding of the intended behavior. Because new businesses emerge over time and involve considerable planning (Krueger, Reilly, & Carsrud, 2000) a career choice involving self-employment is exactly the type of planned behavior for which intention models are ideally suited (Bird, 1988; Katz and Gartner, 1988). The TPB explains entrepreneurial intentions more detailed and consistently than alternative models (Krueger et al., 2000; van Gelderen et al., 2008). Although no perfect relationship has been found between intentional behavior and the choice resulting from this, the actual behavior, the intention of the individuals is the best available proxy of behavior. The TPB model has already proven its worth in the past, as it described entrepreneurial intentions and subsequent career choices of students in the United States successfully (Krueger et al., 2000).

12

2.2 Entrepreneurial intention and antecedents

Intentions are the precursor of actual behavior. Therefore, much of what is considered entrepreneurship today, actually is intentional planned behavior (Krueger et al., 2000). Entry into self-employment is behavior determined by intentions, as was concluded by Katz (1992). New ventures can therefore be classified as the result of individual intentions and the subsequent actions (Bird, 1992). Although the TPB was developed initially to encounter and explain individual behavior in general, it has been adopted by entrepreneurship scholars.

As was stated earlier on, only a limited amount of the working population typically engages in entrepreneurship (e.g. Bosma et al., 2008). According to Reynolds et al. (2005), the entrepreneurial activity within a country is the aggregate result of individual decisions to become self-employed. Intentionality and forethought are acknowledged to be core features of human beings (Bandura, 2001), of which intentionality constitutes a representation of the direction of future action. It affects individuals’ choices as well as directs and maintains behavior. The term entrepreneurial intentions has affinity with other frequently used terms with a similar meaning; e.g. entrepreneurial awareness, entrepreneurial potential, aspiring entrepreneurs, entrepreneurial propensity, and entrepreneurial orientation (Urban, 2004). It is defined as the conscious state of mind that precedes action and directs attention toward a goal, such as starting a new business (Bird, 1988), which is accepted as an encompassing concept, as intentions have been found to be immediate antecedents of actual behavior.

Krueger et al., (2000) state that the decision to become self-employed is a deliberate and conscious one. Starting a new business takes time, since it requires considerable planning as well as a high degree of cognitive processing (Baron, 2004). In past research, scholars have tried to use socio-cognitive models and theories to identify possible antecedents of entrepreneurial intention, especially among young people planning their careers (van Gelderen et al., 2008). A variety of antecedents, which are crucial for the formation of entrepreneurial intention, have been examined in the existing literature. Bird (1988) has grouped these antecedents into two macro-categories, identifying both the individual and the contextual domain.

13

2.2.1 Individual domain

2.2.1.1 Demographics

Consistent relationships have been established between certain personal background variables on one hand, and entrepreneurial behavior on the other (Stanworth, Stanworth, Granger, & Blyth, 1989). Most of them have looked at age and gender, while other demographics where included later on.

Age is an important factor for determining a person’s propensity to found a firm (Reynolds, 1995). The relationship is curvilinear and most studies found that the relationship “is concentrated among individuals in mid-career, i.e. between thirty-five and forty-four years of age and then leveling out.” (Parker, 2004). This differs from the conventional belief that entrepreneurs are young professionals who are presumably unrestrained, risk-loving, and are undaunted from finding new ways of combining resources. Another viewpoint is that young people lack experience and access to capital while older workers may lack energy and commitment, as they have higher incomes to give up. With regard to gender there is a substantial overrepresentation of males among business founders in most countries, making self-employment a male dominated career choice. Dealing with gender, Reynolds, Gartner, Greene, Cox, and Carter (2002) show that adult men in the U.S. are twice as likely as women to be in the process of starting a new business. Although female self-employment rates have risen in recent decades (Fairlie, 2005), the prevalence of business ownership among women can be named small. Three possible reasons for this gender gap are that women lack qualities necessary to successfully set up a new business, that women are more risk-averse (Verheul, Thurik, & Grilo, 2008), and that women have significant family responsibilities (Wellington, 2006).

Several other variables have been studied as possible antecedents, one of which is marital status. Evans and Leighton (1989) show that married people are more likely to get engaged in entrepreneurial activities, especially when they are married to a working spouse. A possible explanation is that they can rely on family support, even if things do not work out. Furthermore, recent research claimed overrepresentation of individuals with close role models. A career choice is influenced by positive and consistent reinforcement from observing significant occupational role models, e.g. parents. This becomes clear as 40 percent of small business owner-managers has a self-employed parent (Davidsson, 1995), which shows a clear link with the behaviorist approach discussed earlier. Davidsson (1995) further argued that not only the presence of self-employed parents but also how positively their status or performance is perceived influences the children’s intentions to become self-employed. This may be due to the fact that children of entrepreneurs are more sharply aware of the entrepreneurial option.

14

effect, thereby only considering years, and not what type of education. Findings often remain inconclusive, except for richer countries, where training shows positive effects (Blanchflower, 2004). In conclusion, education could increase skills needed to establish a business. However, higher educated people are more likely to earn more when they are employed.

Aldrich and Cliff (2003) argue that a small family size facilitates self-employment by lowering the perceived family risk of start-up as it reduces the proportion of relatives involved in start-up. Although the self-employment income is riskier, children may enhance the viability of the business in the long haul, making self-employment more likely. Lastly, Evans and Jovanovic (1989) found that entrepreneurial decisions are positively related to the household income of individuals. House ownership and access to funds have an important positive relationship with becoming self-employed, which both are sort of dependent on the household income.

2.2.1.2 Traits

Psychological characteristics or traits are most frequently used in entrepreneurship research. Three of the leading traits are need for achievement, locus of control, and risk-taking propensity, all of which have been found to be positively related to entrepreneurship. An individual with need for achievement seeks to excel and therefore has a desire for significant accomplishment (McClelland, 1961). Those with a high internal locus of control believe that they can change the world by themselves, so achievements comes forward from their own actions (Rotter, 1966). Finally, the average self-employed individual possesses a high risk-taking propensity, because risk taking is so fundamental within this field, entrepreneurs have a higher propensity toward risk. Moreover, as McClelland (1961) stated with need for achievement, they are better able to evaluate risk, particularly when the situation will lead to positive outcomes (Tversky & Kahneman, 1982). There is a range of other potential traits an entrepreneur may possess; they have a desire for autonomy (value freedom), they are over-optimistic, they are innovative and are more likely to have a higher tolerance for ambiguity.

15

2.2.2 Contextual domain

While the traditional literature tends to stress the importance of the individual domain, most recent research has put emphasis on the contextual domain. This domain focuses on the intentional, expectancy-driven nature of the entrepreneurial decision process, rather than following the more deterministic and traditional view (Autio, Keeley, Klofsen, Parker, & Hay, 2001). Theory and recent empirical findings give reason to expect that the strength of relationships among the TPB components might be moderated by culture (Moriano et al., 2011). Lent, Brown and Hackett (2000) suggested that the immediate personal environment (e.g., significant others), as well as the broader sociocultural context (e.g., societal culture) influence an individual’s career choice. Because it is assumed that the broader sociocultural context exert its influence through the immediate personal environment on career choice decisions (Moriano

et al., 2011), it seems plausible that culture influences entrepreneurial intentions through social

norms, which are linked to the immediate personal environment (Lent et al., 2000). Similarly, Krueger (2000) argues that culture influences intentions primarily through the influence on the social component in the TPB model, which is the “subjective norms” component.

A closer look at subjective norms indicates that they are termed as beliefs arising from social pressure (Ajzen, 1991). Subjective norm is the influence of a person’s normative belief that others approve or disapprove a particular behavior. The intention to carry out a particular behavior depends on the subjective norm. In this research, the subjective norm can be considered as the influence from people’s normative belief that their career choice is accepted, encouraged, and maybe even promoted by their circle of influence (Pavlou & Chai, 2002). Simplified, individuals may believe that their family, friends, and peers would favor a certain career choice, and this belief tends to influence their intentions now and their actual behavior in the long haul. In conclusion, subjective norms can be labeled as underexposed in the literature and will therefore be a focus point in the remainder of this study.

2.3 Career choice

16

The first school focuses on general development of careers over time and emphasizes not only the individual but also society as a variable in career choice. Thus, as Vondracek (1990) puts it, “to study career development one has to study a moving target - the developing individual - within a changing and complex context”. Brown (2002) argued that career choice is part of a bigger picture in which a person develops a vocational self-concept in the context of different life roles (e.g. worker, homemaker, leisured) and arenas (e.g. school, work, community). This first school is in accordance with the idea that culture could be an explanatory variable for differences in career choice because the society is included as a possible variable.

The second school of thought is behaviorist-based. An example of this is the social learning theory of Krumboltz, Mitchell, and Jones (1976), in which career-related decisions are based on social learning, for example by positive and consistent reinforcement from observing significant occupational role models (e.g. family). Therefore, the second school is in compliance with regard to the possible role of culture on career decisions too. Both schools appear to be critical in our understanding of the actual career choice as a self-employed individual, which represents a distinctive career choice from day-to-day paid work.

Substantial empirical proof involving both theories shows that people choose work environments that match their personality, values, needs, and interests (Zhao, Seibert, & Lumpkin 2010). Krueger et al. (2000) argue that career-related decisions reflect a cognitive process in which beliefs, attitudes and even intentions evolve as knowledge and experiences are processed. Research has suggested that entrepreneurial careers fit this pattern (Davidsson, 1991; Katz, 1992). From a practical viewpoint, this could imply that second-generation immigrants, with ancestors coming from different national heritages, could have certain cultural values which may eventually lead to more or less self-employment.

2.4 Research gap

When considering the existing literature on entrepreneurial career choices the focus lies on issues such as personality variables, demographics, personal history, and the social context in search of explanations for the preferences and subsequent choices of the individual (Gibb Dyer, 1994). According to Rauch and Frese (2000), these explanations can all be named too general, which implies that they merely apply to broad classes of behavior. This has an impact on the explanatory power, which can be considered as being low, as these broad classes are weak predictors of actual behavior.

17

In other previous research the importance of the cultural context for career decisions is argued. The paper at hand contributes to the growing literature on the importance of cultural values in determining economic outcomes (Alesina & Giuliano, 2013). One of the growing interests within the research field is cross-national entrepreneurship (Hayton et al., 2002), in which researchers try to grasp the major differences in entrepreneurship between different countries. Although the cross-national comparisons often show interesting results, the problem is differentiating cause and effect from simple association, in which it is very hard to provide an explanation for the findings.

In relation to the paper at hand scholars have conducted cultural comparisons based on ethnic background of participants within one country (van Gelderen et al., 2008), but without clearly distinguishing between cultural versus environmental factors. The main challenge of any cultural analysis is how to separate the effects of culture from the effects of strictly economic factors and institutions. Because this separation is very challenging to achieve, this has prevented economists to grasp the possible influence of culture on career choices. The paper at hand tries to fill this literature gap and therefore strives to meet the challenge of showing, in a quantitative fashion, that culture matters for career choices of individuals.

This is why the epidemiologic approach will be followed in the remainder of this paper. This approach attempts to distinguish between cultural versus environmental factors contributing to individual variation. From a practical point of view this implies studying individuals who were born and reside in the same country, implying that they face the same institutional and economic environment, but with different cultural backgrounds. In our case we will study second-generation immigrants in the United States (U.S.), because this group meets all the aforementioned criteria of being born in the same country, facing an identical environment, but having different cultural backgrounds. The underlying idea is that individuals from different cultures will take different actions despite facing identical environments. By doing so, we can see the true influence of culture, because culture is the only factor which varies.

One additional factor can be named regarding the research gap, because this study further aims to contribute to the existing literature by examining the role of trust. Until so far, very few studies involving entrepreneurship have researched the role of trust with regard to career choices (Liao & Welsch, 2005; Zahra, Yavuz, & Ucbasaran, 2006). Most related research has focused on the importance of social networks for venture creation and business growth, although much of this research only assesses the role of trust indirectly.

18

III. CULTURE AND THE DEVELOPMENT OF HYPOTHESES

A magnitude of scholars lend support to arguments relating to the influence of cultural values on entrepreneurial behavior (Gartner & Shane, 1995; McGrath, MacMillan & Scheinberg, 1992). Over more than a century ago, Weber (1904) advocated that differences in entrepreneurial activity can be explained by, among others, cultural variables. In has been customary in recent research that culture has been underexposed, which is contradictory to what may be expected, as differences in entrepreneurship levels between societies points clearly to the role of culture. The belief that there are certain unseen values which shape human behavior is accompanied by the fact that culture derives from the social environment of a person, which implies that culture is learned and not inherited (Hofstede, 1980). In some of the literature, the view is taken that cultural values are programmed into individuals and therefore determined early in life (Hofstede 1980; Barnouw, 1985) and tend to endure over time (Hofstede 1980; Thomas & Mueller 2000). These arguments are reinforced by the findings of the Global Entrepreneurship Monitor reports, in which it was emphasized that cultural and social norms are antecedents of entrepreneurial orientation, as they seem to be the differentiating factors for high levels of entrepreneurial activity in different countries (Minniti & Bygrave, 2003). Even when researchers accounted for various levels of economic development and political support, entrepreneurship specialists often find a significant difference in the level of entrepreneurial activity between different cultures (Uhlaner & Thurik, 2007).

In the last few years the ethnicity or ancestry of entrepreneurs has enriched the debate. Much of these studies related to ethnic entrepreneurs are based on issues of culture, with an increasing body of literature supporting the claim that national culture influences a multitude of economic behaviors (Minniti & Bygrave, 2000; Hofstede, 1980, 2001). According to Hayton, George, and Zahra (2002) a national culture can either support or impede entrepreneurial behavior at the individual level. Culture is hard to grasp but in general attitudes, behaviors, beliefs, values, habits, and traditions are what constitute a national culture. Culture can be understood at many levels of analysis (Hofstede, 2001), of which one possibility is focusing on the national level. This will be the focus of this study, although recognizing that this is somewhat blunt because country borders are imperfect boundaries for classifying different societies.

19

Cultural differences between countries have the potential to influence a variety of individual behaviors, including the decision to become a self-employed individual rather than an employee (Mueller and Thomas, 2000). This paper will therefore utilize four of Hofstede’s cultural dimensions: 1) power distance, 2) individualism versus collectivism, 3) masculinity versus femininity, and 4) uncertainty avoidance, in an effort to understand the possible effect culture has on the level of self-employment of second-generation immigrants. The country scores for these four dimensions are relative. In other words, the cultural measurements can be only meaningful by comparison.

The fifth and last cultural dimension that will be accounted for is trust. Trust has been a relative new researched field over the past decade, but has been studied in relation to entrepreneurship and career choices (Höhmann & Welter, 2005; Welter & Smallbone, 2006). Scholars have dissect the concept of trust as being a crucial element for both starting and growing a new business. Moreover, trust has been found to positively contribute in decreasing the costs and risks inherent in business transactions in turbulent environments (Manolova, Manev, & Gyoshev, 2007; Puffer, McCarthy, & Boisot, 2010). Trust is a socially constructed phenomenon, in which the subjective nature of the concept makes differences across cultures and countries difficult to explain. According to Welter (2012) “the label 'trust' appears to have become something of a catch-all phrase, afforded to every entrepreneurial phenomenon which involves any form of non-contract based collaboration which has no clear explanation”. In this paper, trust will be measured making use of both the European Values Study (EVS) and the World Values Survey (WVS), in which trust is measured by asking people whether they generally believe people can be trusted. Because this research focuses on the national level, the average of these responses resembles the level of personal trust inherent in the society.

3.1 Power distance (POD)

The first cultural dimension is power distance. This is often referred to as being the degree of tolerance for hierarchical or unequal relationships, which implies that inequalities are allowed and accepted. As Hofstede (1980) argued it is “the degree to which a country tolerates relationships between people of unequal power”. Countries with a relative low power distance score generally claim to have equal rights for each individual. In contrast, high power distance exists in a country where power inequalities are magnified (Hofstede, 1980). These countries favor a differential social status (De Luque and Sommer, 2000). Mitchell et al. (2000) found that power distance exerts an influence on arrangement, ability, and willingness cognitions, which in turn affect the decision to become self-employed.

20

the entrepreneur would endorse a greater amount of differentiation between himself and others. Ironically, individuals who are self-employed want control over their employees, a feature of high power distance (Brown, 2003). Having said this, entrepreneurs are less likely to endorse a low power position and “cannot accept authority”, as was indicated by Hagen (1962).

Another notable antecedent is the societal level at which a culture views the value of the entrepreneur, as this can also influence entrepreneurship. Consequently, Thurik and Dejardin (2011) claimed that more self-employment is found there where the entrepreneur is considered to have a high social status, which can be associated with high power distance. In accordance with the research of Hayton et al. (2002), these arguments lead to the following hypothesis:

Hypothesis 1: Second-generation immigrants with a background from a culture with higher

power distance are more likely to be self-employed. 3.2 Individualism versus Collectivism (IDV)

Individualism is an indicator for the strength of the link between individuals in a community. A society's position is typically reflected in whether people’s self-image is defined in terms of “I” or “we” (Hofstede, 1980). The differences between these cultures are obvious in terms of family ties and personality traits. In individualist societies, although the family is important, the degree of emphasis placed on individual accomplishment outranks group benefits (Triandis, 1995). Countries which are more collectivist, characterized by lower scores, show strong group cohesion, loyalty and respect for the members of a particular community. Here, the interest of immediate family is paramount (Triandis, 1995). Furthermore, intrinsic rewards are of significant importance, which implies that harmony prevails over honesty.

Entrepreneurs are often viewed through an individualistic lens, the word lends itself to a somewhat individualistic interpretation. Scholars agree that most entrepreneurs are innovative individuals who follow their own dream in order to found their own business. Sexton and Bowman (1985) stated that self-employed individuals need autonomy and independence and do not need direct support from others or are willing to conform to the norms of others. This is further reinforced by McGrath et al. (1992) who stated that self-employed individuals prefer independent action and separation from group work. When individualistic values are present in a society this has an effect on how much entrepreneurial behavior is displayed. For a given country, the more individuals with individualistic values, the more entrepreneurial behavior is detected (Davidsson, 1995).

21

Although there are some studies that speak in favor of collectivism, the general picture remains unchanged, implying that individualistic societies will lead to more individuals choosing a career as a self-employed. The focus on materialistic achievement and wealth common in these cultures makes them more supportive of entrepreneurial activity (Gupta, Hanges, & Dorfman, 2010). Indeed, Hofstede (1980) argues that collectivistic cultures are generally less likely to exhibit high rates of self-employed individuals. In accordance with the research of Hayton et al. (2002), these arguments lead to the following hypothesis:

Hypothesis 2: Second-generation immigrants with a background from a culture with higher

individualism are more likely to be self-employed. 3.3 Masculinity (MAS)

Hofstede’s third cultural dimension, masculinity versus femininity, refers to the degree of toughness and competitiveness of the members of the society and considers to what extent achievement and aggression are valued. Or as Hofstede (1980) puts it, this dimension distinguishes “the more masculine traits of materialism and success versus the more feminine traits of harmony and relationships”. According to McGrath et al. (1992) masculine cultures can be named aggressive, assertive, and are driven by individual wealth, while feminine cultures are peaceful, thoughtful, and driven by societal needs. In other words, a masculine orientation puts emphasis on things and money while a feminine orientation tends toward people and the quality of life (McGrath et al., 1992).

The fundamental issue with this dimension is how much a society sticks with, and values, traditional male and female roles. There is one problem with regard to this distinction, as it has been named confusing. This is because of the distinction between people’s actual gender and their cultural beliefs. Therefore it is important to understand that women in a masculine culture might well have a more masculine set of beliefs than men in a feminine culture. So, it is possible for a female to possess a masculine of believes, although the name suggests otherwise.

Achievement in a high-masculine oriented society is associated with wealth and position, rather than with human contacts and the living environment. Here, the former shows similarities with self-employment, as these individuals are most probable to value money as a motivator for achievement. In high masculine cultures, gender roles are more differentiated showing “a gap between men’s-values and women’s-values” (Hofstede & McCrae, 2004). In general, qualities considered essential for business tend to be viewed as masculine (Heilman, 2001). This includes entrepreneurship, which is perceived to require traits such as independence, aggressiveness, autonomy, and courage, frequently associated with men (Gupta et al., 2009). It thus seems plausible to assume that masculine individuals are more likely to be self-employed. In accordance with the research of Hayton et al. (2002), these arguments lead to the following hypothesis:

Hypothesis 3: Second-generation immigrants with a background from a culture with higher

22

3.4 Uncertainty avoidance (UNA)

The fourth and last dimension of Hofstede is uncertainty avoidance which refers to the degree of fear that members of a society feel when dealing with unfamiliar and uncertain situations and whether or not they have created beliefs and institutions to try to avoid these occurrences. Hofstede (1980) describe it as “the degree of acceptance for uncertainty or willingness to take risk”. Cultures with low uncertainty avoidance members can be named as being more tolerant. Those people rather easily accept the notion of ambiguity and therefore ‘go with the flow’ when problems arise. A key feature of these societies is that members are encouraged to deal with this ambiguity while further encouraging people to discover their own truth. At high scores uncertainties are avoided through rule, these societies are searching for a collective truth. Here, individuals have a lower tolerance for ambiguity and try to cope with the unknown future by imposing increased stability upon their environment. Tolerance for ambiguity is the degree to which members of the society are open to change and innovation (Kedia & Bhagat, 1988).

Generally, when starting a new business or becoming self-employed, the individual is confronted with a great deal of uncertainty. Sexton and Bowman (1985) concluded that “the willingness of the entrepreneur to participate in an environment that is uncertain, changes frequently, and does not offer clear guidance for action”, while further indicating that entrepreneurs are far more willing to cope with ambiguity in comparison to the less enterprising. Therefore, individuals who are self-employed are considered to excel in low uncertainty avoidant cultures. On the other hand, becoming self-employed may be socially discouraged in high uncertainty avoidant cultures. The ability of entrepreneurs to tolerate relational risk depends on the cultural values of the entrepreneur’s society (Steensma, Marino, & Weaver, 2000). McGrath

et al. (1992) has advocated that entrepreneurs tolerate risk and ambiguity much better in

contrast to non-entrepreneurs.

The literature overwhelmingly links beliefs and values in relation to the self-employed with low uncertainty avoidance. In these cultures members of the society are more open to change and innovation which breeds individuals in becoming self-employed. If a culture wants to encourage self-employment, it must teach its people to perceive uncertainty and change as reasonable aspects of business, not dangers to be avoided (Saffu, 2003). In accordance with the research of Hayton et al. (2002), these arguments lead to the following hypothesis:

Hypothesis 4: Second-generation immigrants with a background from a culture with lower

uncertainty avoidance are more likely to be self-employed.

23

3.5 Trust (TRU)

When we take a closer look at the concept of trust there are several possible links between trust and entrepreneurship. For example, Luhmann (2000) concentrated on the role that trust plays for modern societies, understanding trust as a mechanism that reduces the complexity in everyday life. Luhmann (2000) argues that trust can be a substitute for the risks that is inherent in decisions or situations. For an entrepreneur, this suggests that not all business relationships need to be regulated via contracts, thus allowing to reduce the transaction costs. According to Williamson (1993), “both in personal and commercial relations, trust is often based on a calculated risk, i.e., the person trusting is convinced that trusting is a rational behavior because the risks associated with the bestowed trust that is not being fulfilled are justified by the potential gains if the trust is maintained”.

Trust can be seen as a quasi-automatic outcome of social capital, helping to foster self-employment by individuals (Audretsch, Aldridge, & Sanders, 2011). Aldrich (2000) indicates that successful new entrepreneurs are more likely to be those who can build networks of trust, which assists them in creating legitimacy within the market. He refers to an earlier, unpublished, article by Gartner and Low (1990) who advocated that “organizations emerge when entrepreneurs are successful in achieving an understanding among the trusting parties—potential customers, creditors, suppliers, and other individuals and organizations—that things will work out”. Key characteristics of trustworthy partners include personal characteristics, past behavior and emotions such as demonstrated honesty, loyalty, sympathy and empathy (Nooteboom, 2002).

In conclusion, one can argue that trust is generally believed and shown to be positively related to self-employment. Trust can be seen to assist in lowering the transaction costs and is crucial for creating legitimacy in the market. However, being too trusting may be detrimental if it leads to gullibility and easy exploitation by others. This leads to the following hypothesis:

Hypothesis 5: Second-generation immigrants with a background from a culture with higher trust

levels are more likely to be self-employed.

Figure 1: Conceptual model

In figure 1, the conceptual model can be found. How the five different hypotheses are being tested, as well as other methodological information, can be found in the next chapter.

24

IV. METHODOLOGY

This chapter gives an overview of the method used throughout this research and is divided in two sections; the data sources and the methodological approach. In the former the different sources, the sample and the control variables will be discussed. In addition, the latter outlines the methods that are followed in this research and indicates how the data are analyzed.

4.1 Data sources

4.1.1 General Social Survey (GSS)

Throughout this research, the General Social Survey (GSS) is the main source of data. This sociological survey is used because it offers demographic characteristics and attitudes of residents in the United States. The survey is conducted face-to-face, making use of an in-person interview by the National Opinion Research Center, of adults in randomly selected households. The survey was conducted every year from 1972 to 1994 (except in 1979, 1981, and 1992). Since 1994, the study has been conducted every other year. In order to keep constant markets and institutions, we study the career choices of individuals who were born and reside in the U.S., but who have different national heritages and hence, presumably different cultures. This survey contains several other variables with critical information, which will be discussed hereafter.

First of all, the GSS data provide information on the current employment status of the respondents by asking the question: “Are you self-employed or do you work for someone else?”. For all the hypotheses and models in this research the career choice of individuals will be used as the dependent variable. In order to make this variable suitable for analysis by means of logistic regression a dummy-variable was created with value 1 when a respondent was self-employed and value 0 when the respondent was working for someone else.

Identifying second-generation immigrants was done by considering the question: ‘Were you born in this country?’. When the respondent identified he or she was born in the United States, consideration was given to the question: ‘Were both your parents born in this country?’. Here, the respondent could indicate whether none, one or both parents were born outside the United States. When one parent was native born it was possible to indicate whether this applied to the father or the mother. In order to identify the second-generation immigrants this research excluded those respondents that indicated that both parents were born in the United States.

25

4.1.1.1 Sample

To obtain a sizeable sample, we use observations from 1977 until 2012, as 1977 was the first year the country of origin of the respondent’s forbearers was reported. This gives us 24 different time-frames in which a population of 49471 individuals were randomly selected, among which a multitude of second-generation immigrants from different cultural backgrounds have indicated their career choice.

The first major decision in this research was to determine how second-generation immigrants will be measured. Second-generation immigrants refers to individuals who were born in the U.S. and who have at least one foreign born parent (Suro & Passel, 2003). Suro and Passel (2003) further indicated that although there is some ambiguity in reference to the definition of second-generation immigrants, the definition used here is cited by major research centers such as the United States Census Bureau and the Pew Research Center. We construct our sample by including all second-generation immigrants who indicated whether or not they were self-employed or working for someone else between the age of 18 and 65 years, because this is the working population in the United States. Elderly people will not be considered because they are less likely to take part in self-employment activities, or work whatsoever. Younger people are excluded because the GSS only conducts interviews on adults. This leaves us with a sample of 3008 individuals divided over all the identified countries within the GSS database.

The GSS data file identifies 42 countries of origin, of which some were grouped countries, i.e. Africa, West-Indies, and Arabic. From these grouped countries no cultural measurement could be drawn, as it was unclear exactly which countries were covered within these groups. There were a few countries that needed an adjustment, first of all England and Wales are drawn together with Scotland, jointly forming Great Britain. Although Great Britain technically speaking also includes Northern Ireland we assume that this country has not been investigated, and if it is examined it will fall under the wings of Ireland. Secondly, in this research there is no distinction being made between ‘French Canada’ and ‘Other Canada’, but instead it is decided to merge them into Canada. Thirdly, because the GSS data are published from 1972 onwards, the database still contains old classifications of countries which have been split up in the past decades, such as Czechoslovakia and Yugoslavia. Even though these countries do not actually exist anymore they will be used in this research. When multiple countries exist within one classification used by the GSS the weighted average of each country, for each individual cultural dimension is used. This weight is based on the population of each country. When considering for example Czechoslovakia this implies that 65,6% of the cultural score for Czechoslovakia is based on the cultural score of Czech Republic, with a population of 10.3 million, and 34.4% is based on the cultural score of Slovakia, with a population of 5.4 million. Lastly, Puerto Rico is excluded because no cultural measurements could be obtained for this country.

26

4.1.1.2 Control variables

As was indicated earlier on, certain demographics resulted in more or less self-employment. We will include some of these variables in our analysis as control variables, as these are available in the GSS database. Unfortunately, this does not apply to the traits discussed earlier on, which is the reason this research does not control for this. Each control variable will be discussed below:

Age – The age of the respondents.

Gender – The gender of the respondent. A dummy variable is created in which value 1 is male

and value 0 is a female.

Family size – The family size looks at the number of children a respondent has by asking the

question: “How many children have you ever had? Please count all that were born alive at any time”. One problem is that the maximum number of children that can be indicated without problems is seven. This is due to the fact that when a respondent indicates value 8, this implies that this individual has eight children or more. For convenience reasons, this research assumes that all these individuals have eight children. Although it may be possible that respondents had a child which is deceased or have more than eight children, this is the best estimate of family size.

Marital status – The marital status of the respondent is examined on the basis of the following

question: “Are you currently -- married, widowed, divorced, separated, or have you never been married?”. To make this allocation somewhat more clear it has been decided to create a dummy variable with two options, either you are married, indicated by value 1, or you are not married, indicated by value 0.

Father self-employed – It has been questioned whether or not the father of the respondent was

self-employed. The same holds for the career choice of the mother, but only very recently, which does not provide the opportunity to examine the potential impact of the self-employment status of the mother. Whether or not a respondent had a close role model in the form of the father being employed was examined by asking the following question: “Was he (father) self-employed or did he work for someone else”. Again, a dummy variable was created, with value 1 for a self-employed father, and value 0 for a father working for someone else.

Education level – When examining the level of education from the respondent they were asked

the question: “What is the highest grade in elementary school or high school that you finished and got credit for?”. The respondent was asked to rank one of the options from 0 to 20 years of education.

Household income – This variable looks at the total family income in a given year by asking the

question: “In which of these groups did your total family income, from all sources, fall last year before taxes, that is?”. The respondent was then considered to indicate the category to which they belonged, category 1 being less than $2500, while category 12 indicates a family income of over $25000.

27

4.1.2 Cultural variables

There are five cultural variables which will be used in this analysis, which are the independent variables. The cultural dimensions of Hofstede distinguish country cultures from each other and are categorized into four groups. Furthermore, trust measures the degree of trust inherent in a society. The definitions and identification of these variables can be found in table 1.

The first four cultural dimension are the Hofstede dimensions of national culture, which are measured by applying the widely used method of Hofstede (1980). This method assigns a score between 0 and 100 to each individual country on each individual dimension.

In order to calculate trust for a given country this study uses both the World Value Survey (WVS) as well as the European Value Study (EVS), which investigated socio-cultural variables in 6 different waves from 1981 onwards. As a measure of trust the answer to the binary question: “Generally speaking, would you say that most people can be trusted or that you have to be very careful in dealing with people?” will be used. In order to be able to interpret the outcomes, a dummy variable is created which takes value 1 if the person answered that most people could be trusted and value 0 when an individual believes that you have to be very careful in dealing with people. This research assigns a single trust score to every respondent from the same country in our sample. Therefore, the average of all respondents within a given country are calculated from 1981 until 2009, to construct a value for each country. By means of using rescaling the trust scores range from 0 to 100, as this brings them in line with the Hofstede dimensions. It is decided to exclude wave 6 (2010-2014) as it suffers from methodological issues.

Cultural dimensions Definition Power Distance—degree of tolerance for

hierarchical or unequal relationships

High score — large degree of tolerance for

unequal relationships

Low score — small degree of tolerance for

unequal relationships Individualism—degree of emphasis placed

on individual accomplishment

Individualism (high score) — large degree of

emphasis on individual accomplishment

Collectivism (low score) — large degree of

emphasis on group accomplishment Masculinity—degree of stress placed on

materialism

Masculinity (high score) —large degree of stress

on materialism and wealth

Femininity (low score) —large degree of stress on

harmony and relationships Uncertainty Avoidance—degree of

acceptance for uncertainty or willingness to take risk

Strong (high score) — little acceptance for

uncertainty or risk

Weak (low score) — generally accepting for

uncertainty and risk Trust—degree of trust inherent in society

High score — Most people can be trusted Low score — You have to be very careful in

dealing with people

28

4.2 Methodological approach

Epidemiology originated in the study of patterns, causes, and effects of health and disease conditions in defined populations. Epidemiologists believe these events do not befall the population in a purely chance fashion, but believe in causes as the determiner of who gets sick.

When considering the literature on culture and economics, we find that the epidemiological approach is also applicable and has proven valuable for economic research. We will follow this approach as it studies the variation in outcomes across different immigrant groups residing in the same country. These immigrants presumably differ in their cultures but share a common institutional and economic environment. This allows one to separate the effect of culture from the original economic and institutional environment. In an empirical context this implies using data on individuals that live in one country (in this case U.S.), but whose parents were born in some other country, for which the GSS data are ideally suited. For a long period of time, questions regarding the role of culture in economic outcomes were largely absent in this research field. This was primarily the result of the absence of an empirical method with which it is possible to isolate the effect of culture. With the emergence of the epidemiological approach within economic research, scholars can now isolate and tackle the possible influence of culture, even in a quantitative manner.

This new approach has expelled standard approaches to this question, which are fraught with problems of omitted variables. In past decades a variety of new approaches provided more persuasive evidence that culture matters, which also is the case with respect to the culture’s impact on economic outcomes. A large portion of this evidence stems from following the epidemiological approach, which attempts to separate culture from the environment by studying outcomes of individuals whose cultures are potentially different within a common economic and institutional setting (Fernández, 2010). Fernández (2010) stated that the reasoning underlying this strategy is that “(I) parents transmit their cultural beliefs to their children; (II) cultural beliefs vary across (immigrant) groups in a systematic fashion reflecting culture in the country of origin; (III) individuals who live in the same country or in the same appropriately defined geographical area, face similar economic and formal institutional environments.”

While the second and third line of reasoning are straightforward, the first adoption requires some explanation. Cultural reproduction is an important component, which is the transmission of existing cultural values and norms from generation to generation. Parents prove to be the most influential enculturating forces of cultural reproduction (Gray & McGuigan, 1993). There is abundant evidence that attitudes are transmitted from parents to children. For example, Dohmen et al. (2008) showed that a child’s propensity to trust and attitudes toward risk are strongly positively correlated with parental attitudes.

29

4.2.1 Data analysis

We build up a regression model including the dependent and independent variables as mentioned in the preceding sections. Whether an individual is self-employed or not is a decision which is called dichotomous, in other words, the decision has two possible outcomes: either you are self-employed, or you are not. Because this dummy variable is the dependent variable in this study we cannot use linear regression to encounter possible variations between individuals from different cultural backgrounds, as was argued by Gujarati (2003).

Two of the underlying reasons are that measures, such as R², will not be reliable and secondly, linear regression may give predicted values of the dependent variable that are greater than 1 or less than 0. This will be difficult to interpret as the predicted values are intended to be probabilities, which must lie between 0 and 1. To avoid the limitations we need a model that has the feature that as the explanatory variable increases the outcome should remain within the range between 0 and 1. Thus, the relationship between the independent and dependent variables is necessarily non-linear. These shortcomings led to the development of a more refined and improved model called the logit model. In this model, the cumulative distribution of the error term is logistic, making it more realistic in that it is non-linear (Maddala and Lahiri, 1992). The logit model is estimated using the maximum likelihood approach. By means of using this specific logistic regression this research shows whether there is a specific link of the independent variables on the dependent variable, the career choice. In logistic regression, a complex formula is required because logistic formulas are stated in terms of the probability that an individual is self-employment, which entails value 1, which can be referred to as probability or p. The probability that an individual is working for someone else, which is indicated by value 0 can be interpreted as 1 – p. The model therefore has the following form:

logit (p) = ln ( p

1-p) = β0 + β1 * PDI + β2 * IDV + β3 * MAS + β4 * UAI + β5 * TRU + CONTROLS + ε In this model, ln refers to a natural logarithm, furthermore, PDI is the power distance index, IDV is the individualism versus collectivism index, MAS is the masculinity versus femininity index, UNI is the uncertainty avoidance index, TRU is the trust index, and CONTROLS is the aggregation of all control variables. Subsequently, β0 is the constant, the other β’s are the coefficients of the direction and ε is the residual. In addition to this logistic regression we will also examine whether the independent factors correlate with each other by consulting the correlational matrix.

30

V. ANALYSIS AND RESULTS

In the analysis and results section, we discuss the results of the estimated model of self-employment status, which can be found in the method section. First, however, we look at some notable descriptive results regarding variables used in this research, before we move on to the regression analysis. As was indicated earlier on the total population consist of 3008 second-generation immigrants who were divided over all the countries within the GSS dataset. After classifying the countries, we have access to a total sample of 2332 individuals. When considering the missing values from the control variables, another 495 respondents are dropped, implying that a remaining 1837 second-generation immigrants will be considered in the remainder of this result section until the robustness checks.

5.1 Descriptive statistics

The first result is the overall level of self-employment throughout all the 29 selected countries, in which a rounded percentage of 12%, or 221 out of 1837 respondents, of the second-generation immigrants in the U.S., are currently self-employed.

Secondly, the five cultural and independent variables of this study, which are scaled from 0 to 100, all show diverging results. The averages of the four cultural dimensions of Hofstede all range somewhere between 53 and 67, while trust has an average of only 33.3%, indicating that of all the respondents, one out of three believes that most people can be trusted. The standard deviations for the Hofstede dimensions have an average of 20 which can be named fairly high. This indicates that that the numbers are fairly spread out from the average, in other words, there are large differences between countries and their cultural scores.

The lowest power distance score has been found in Austria, with a score of only 11. For individualism the lowest score is 20, which is the index for the collectivistic China. The country with the lowest masculinity index is Sweden, with a score of only 5, indicating a very feminine culture. The degree to which a culture is uncertainty avoidant has been found lowest for Denmark, with an index of 23. The country in which the majority of people believes that you must be careful in trusting other people is the Philippines, with a percentage of only 7%.