Does the Decision to Become an Entrepreneur Differ
Between Males and Females?
University of Amsterdam
Name: Dominique Blom Student Number: 10631380 Date: 29/06/2016
Faculty: Economics and Business
Specialization: Finance and Organization Bachelor Thesis (12 EC)
Supervisor: M. Koudstaal
Abstract
Using a large dataset provided by the General Entrepreneurship Monitor (GEM), this thesis examines if the characteristics that cause females to become entrepreneurs differ from the characteristics that cause males to become entrepreneurs. More specifically, the gender differences in the way fear of failure, confidence in one’s entrepreneurial skills, knowing another entrepreneur and opportunity perception influence the decision to start a business are analyzed. The results suggest that the way in which perceptual variables influence the decision to become self-employed tends to be the same for females and males, with the exception of knowing another entrepreneur. Knowing another entrepreneurs seems to be less important for females than for males in the decision to start a new business. Furthermore, when checking for robustness the gender difference in the way knowing another entrepreneur influences the decision to become self-employed disappears, potentially due to smaller sample
Statement of Originality
This document is written by Dominique Blom who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Inhoudsopgave
1. INTRODUCTION ... 4 2. THEORETICAL BACKGROUND ... 6 2.1FEAR OF FAILURE ... 8 2.2PERCEIVED OPPORTUNITIES ... 10 2.3KNOWING AN ENTREPRENEUR ... 11 2.4PERCEIVED CAPABILITIES ... 12 3. METHODOLOGY ... 14 3.1DATA ... 14 3.2METHODOLOGY ... 15 4. RESULTS ... 17 4.1SUMMARY STATISTICS ... 17 4.2CORRELATIONS ... 18 4.3RESULTS ... 20 4.3.1 Main Results ... 20 4.3.2 Robustness Checks ... 235. CONCLUSION AND DISCUSSION ... 26
1. Introduction
Entrepreneurship has been shown to play an important role in economic growth. For instance, Van Praag and Versloot (2007) examined recent empirical evidence about the economic value of entrepreneurship and found that entrepreneurship generates relatively much employment, productivity growth, innovation and economic growth. Because of this widely known outcome, policy-makers have been promoting entrepreneurship amongst individuals that are underrepresented in the entrepreneurial population (European Commission, 2014). Among these underrepresented individuals, females play a central role.
Several scholars have demonstrated that gender plays an important role in the decision to become an entrepreneur. For example, Delmar and Davidsson (2000) analyze the characteristics of entrepreneurs and find that being male has a strong influence on the probability of becoming an entrepreneur. Moreover Blanchflower (2004), who examines the determinants of being self-employed, also finds that males are significantly more likely to start a business. These findings are in harmony with the fact that the number of female entrepreneurs is significantly lower than the number of male entrepreneurs globally, although it has to be admitted that the amount of females involved in starting a business has increased in the past few years (De Bruin et al., 2006; Langowitz and Minniti, 2007).
Recently, the amount of literature based upon female entrepreneurship has grown. Several studies focused on the relationship between female entrepreneurship and economic growth and found that female entrepreneurship is an important engine of economic growth (Acs et al., 2011; Brush and Cooper, 2012, Minniti, 2010). Female entrepreneurs contribute positively to GNP, innovation, job creation and social welfare globally (Allen et al., 2007). It has been argued that in addition to the number of entrepreneurs, the diversity (for example in terms of gender) in entrepreneurship also plays an important role in economic prosperity (Verheul and Van Stel, 2007; Van der Zwan et al., 2012). But despite the increase in literature and the corresponding findings that females contribute to economic growth, female entrepreneurship is an understudied topic (De Bruin et al., 2006). According to Brush and Cooper (2012), studies about women entrepreneurship comprise less than 10% of all research in the field. Therefore, this paper aims to partially fill this gap. More specifically, it will be assessed if the characteristics that cause females to become entrepreneurs differ from the characteristics that cause males to become entrepreneurs. In other words, are the drivers of the decision to become an entrepreneur different for females than for men?
The focus of this paper will be on Dutch entrepreneurs. Van der Stel et al. (2012) state that female entrepreneurship in the Netherlands increased steadily from 2000 to 2009. In 2011
it was even the case that the number of women participating in early-stage entrepreneurial activity increased with a larger amount than the number of participating men. In 2011, 10.4% of the male population and 6.0% of the female population was involved in entrepreneurial activity in the Netherlands, while in 2010 these percentages were 10.1% and 4.4% respectively (Van der Stel et al., 2012). Span, Van Stel and Van den Berg (2014) expect that this trend will continue and that the gender gap in entrepreneurial activity will eventually decrease in the Netherlands. De Wit and Van Winden (1989) already examined the determinants of becoming self-employed in the Netherlands. They considered family background, personal qualities and sectors as factors influencing the decision to become an entrepreneur. In addition, they also added a gender dummy to see if gender has an impact on the choice of becoming self-employed. De Wit and Van Winden (1989) conclude that among other things Dutch females are less likely to become an entrepreneur. The objective of this thesis is to further explore the mechanism behind this by examining the effect of fear of failure, perceived opportunities, knowing an entrepreneur and perceived capabilities on early-stage entrepreneurial activity. By taking the Netherlands into consideration, this thesis contributes to the scarce research that analyzes female entrepreneurship in Western European countries.
Believing to have sufficient skills to start a business, recognizing good business opportunities, perceiving to have a broad network and risk propensity are considered as subjective perceptions and judgments about the environment. These perceptions are taken into consideration when deciding whether to engage in entrepreneurial activity or not. Several scholars have already accounted for these perceptual variables when analyzing the personal characteristics of entrepreneurs and found that these variables have a significantly greater impact on the decision to start a business than demographic and economic variables (Arenius and Minniti, 2005; Koellinger et al., 2007; Sternberg and Wennekers, 2005).
With data from the General Entrepreneurship Monitor (GEM) an analysis from 2001 to 2012 is conducted. GEM is a research program that describes and analyzes entrepreneurial processes around the globe. By conducting surveys globally, GEM tries to identify those individuals that are involved in entrepreneurial activity. The data provided by the GEM is very well suited for the purpose of this thesis, given that it is one of the first initiatives that includes questions in their survey about individuals’ perceptions towards entrepreneurship.
The results show that perceptual variables have a heterogeneous effect on the decision to become an entrepreneur. It is found that knowing another entrepreneur is less important for females than for males in the decision to start a new business. The relationship between the
likelihood of becoming an entrepreneur and opportunity perception, fear of failure and perceiving to have sufficient capabilities do not have a different impact for females than for males. Furthermore, when checking for robustness the gender difference in the way knowing another entrepreneur influences the decision to become self-employed disappears, potentially due to smaller sample sizes. Overall, the way in which perceptual variables influence the decision to become self-employed tends to be the same for females and males, with the exception of knowing another entrepreneur.
This thesis contributes to the literature by providing a better understanding of the gender differences in the perceptual variables that cause individuals in the Netherlands to become self-employed. Before the GEM was initiated most of the research based on subjective perceptions was restricted to experimental settings because of the lack of data. In addition, a large amount of research only states that perceptual variables explain part of the gender gap in entrepreneurship but do not provide further insights. By adding a gender interaction effect to the perceptual variables, this thesis adds to the previous findings by determining the differences in the way perceptual variables influence females and males in their decision to become an entrepreneur. Considering that this thesis finds a significant gender difference in the way knowing another entrepreneur influences the decision to become self-employed, it is important to account for perceptual variables when analyzing gender differences in entrepreneurship.
The remainder of this paper is structured as follows. In the Chapter 2 relevant literature is reviewed and hypotheses are developed. In Chapter 3 the focus is on the data and methodology. Lastly, the results are discussed in Chapter 4, followed by a discussion and conclusion in Section 5.
2. Theoretical Background
A significant amount of scholars in various fields have analyzed what variables are correlated to an individual’s decision to become an entrepreneur. Entrepreneurship is seen as an employment choice and as a result empirical research considers demographic and economic factors like age, work status, household income and education to be important determinants for the decision to start a business. However, recent literature in the field of entrepreneurship has shifted its focus towards perceptual variables. These variables have been shown to play an important role in the self-employment decision process and should be included when examining entrepreneurial behavior (Arenius and Minniti, 2005; Taylor,
1996). Koellinger, Minniti and Schade (2013) even state that when considering gender differences in the entrepreneurial propensity the explanatory power of demographic and economic characteristics decreases or disappears completely when adding perceptual variables. Considering that the focus of this thesis is on the gender differences in perceptual variables, the existing literature about the perceptual variables included in the model will be reviewed in the following sections. The first variable discussed is fear of failure, followed by perceived opportunities and knowing an entrepreneur. Lastly, perceived capabilities will be discussed. But first a brief discussion of the relevant economic and demographic variables is provided.
Starting with household income, Smallbone and Welter (2001) state that an individual’s income is positively related to the decision to start a business. Individuals with greater (family) wealth face fewer financial constraints, which makes it easier for them to start a business (Evans and Jovanovic, 1989; Blanchflower and Oswald, 1998). Verheul and Thurik (2001) specifically discuss the barriers that Dutch females face in obtaining the required capital to start a business.
Taylor (1996) showed that employed individuals are more likely to become self-employed than individuals without a job, indicating that work status has a positive effect on the entrepreneurial decision. This effect applies for both males and females across countries (Acs et al., 2005). Several scholars show that unemployment rates are negatively correlated to the decision to become an entrepreneur (Blanchflower and Oswald, 1990; Acs and Evans, 1994). This indicates that individuals do not become entrepreneurs out of necessity, but rather that other motivations are at play. Kelley, Singer and Herrington (2016) state that in the Netherlands especially women do not start a business because they have no job.
Age is shown to have an inverted U-shape relationship with the decision to become an entrepreneur (Levesque and Miniti, 2006). After a threshold age is reached, the willingness of an individual to start a business declines. The older an individual becomes, the less time the individual has to earn back the investment that needs to be made to start a new firm (Beugelsdijk and Noorderhaven, 2005). Besides that, income from paid-employment often increases with time, considering that the individual gains more experience and seniority. This reduces the incentives to become self-employed even further (Levesque and Minniti, 2006). According to Kelley, Singer and Herrington (2016) the highest participation rates in entrepreneurship are among the 25-34 and 35-44 year olds. This holds for both males and females.
The effect of education on entrepreneurship is ambiguous. Wagner and Sternberg (2004) state that becoming an entrepreneur requires a high level of education because a certain level of knowledge is needed to succeed. But other scholars concluded that the level of education does not play a significant role in the decision to become an entrepreneur (Blanchflower, 2004; Grilo and Irigoyen, 2006). The ambiguous outcomes hold when differences in gender are taken into consideration. For instance, Dolinsky et al. (1993) find that the likelihood of a female becoming an entrepreneur increases with increasing levels of education. On the other hand, Burke, FitzRoy and Nolan (2002) conclude that education has no significant impact on the decision of females to start a business but has a negative effect on the decision of males.
2.1 Fear of Failure
Fear of failure can be seen as a component of risk aversion (Arenius and Minniti, 2005). The intuition is that an individual is more risk averse when he or she perceives a higher fear of failure rate. Since fear of failure is negatively linked to the decision to become an entrepreneur, less risk averse individuals are more likely to start a business (Kanbur, 1979; Kihlstrom and Laffont, 1979). This believe is supported by Carmer, Hartog, Jonker and Van Praag (2000), who use a Dutch sample of entrepreneurs to empirically test the effect of risk aversion on entrepreneurship. They confirm that entrepreneurship is indeed discouraged by the individual’s level of risk aversion.
Gender plays an important role in an individual’s attitude towards risk as well. Donkers, Melenberg and Van Soest (2001) investigate what factors influence an individual’s risk attitude by using a large Dutch household survey. The survey contains information about the attitudes of respondents towards risk and their background information. With this information the authors were able to determine whether attitudes towards risk vary with the observed characteristic of the participants, such as gender. They find that females are in general more risk averse than males. But also Dohmen, Falk, Hufmann, Sunde, Schupp and Wagner (2011) conclude that females have a more negative attitude towards risk than males. They conducted a survey with questions about the willingness to take risk on an 11-point scale amongst 22,000 individuals living in Germany. These outcomes are in line with the findings of Croson and Gneezy (2009), who review literature on gender differences in risk preferences. They state that most lab and field studies demonstrate that females are more risk averse than males. However, Filippin and Crosetto (2014) reconsider the intuition that females are more risk averse than males. They perform a meta-analysis and find that the results are rather mixed. The likelihood that females are more risk averse than males seems to
depend on the components of the task used to observe the risk attitudes. Hence, gender differences in risk preferences cannot be taken as a fact.
Several studies within the area of entrepreneurship have found that women are more risk averse than men. Sexton and Bowman-Upton (1990) investigate the psychological traits of entrepreneurs and test if there are significant differences between males and females. They find that female entrepreneurs score significantly lower on traits related to risk taking than males. According to them, females are less willing to participate in situations that have uncertain outcomes. The paper of Verhuel, Thurik, Grilo and Van der Zwan (2011) examines why the amount of female entrepreneurs is significantly lower than the amount of male entrepreneurs by investigating the entrepreneurial personality. In their research, they use data from the 2004 Flash Eurobarometer survey, which contains a random sample of the general working-age population from 29 countries. They take risk attitude, self–employed parents, perceived abilities and other socio-demographic factors in consideration to be drivers for an individual to be involved in entrepreneurial activity. They find that risk aversion plays a bigger role in preventing females from becoming entrepreneurs than males. Wagner (2014) analyzes if gender differences in characteristics and attitudes can explain the entrepreneurial gender gap in a Western industrialized country (Germany). The most striking finding of the paper is that fear of failure has a more negative influence on the decision to become an entrepreneurs for females than for males.
The gender difference in the way fear of failure affects the decision to become an entrepreneur is due to several reasons. First, females are more likely to perceive a situation as risky than males and this intuition is likely to hold for entrepreneurship as well (Eckel and Grossman, 2013). Second, females perceive additional risk concerning their home and family. Even though the role of females within society changed severely, women are still more likely to be the primary parent and care takers of the household activities (OECD, 2001). Therefore, females considering engaging in entrepreneurship are more concerned than males that such activity will demand more of their time at the expense of their family (Ascher, 2012; Ivanova Yordanova and Alexandrova-Boshnakova, 2001). Lastly, females tend to consider starting new businesses in riskier sectors, such as service and retail. In general, these sectors are more competitive and show lower survival rates, which leads to higher fear of failure rates for females (Madill et al., 2006; Parker, 2009).
Overall, the intuition is that females perceive more risk than males. Several studies have shown that this also holds in the entrepreneurial field. Since an individual’s fear of
failure rate is higher when he or she is more risk averse, the following hypothesis is established:
H1: The effect of fear of failure on the decision to become an entrepreneur has a more
negative effect for females than for males.
2.2 Perceived Opportunities
Individuals participating in the GEM survey were asked if they perceived good opportunities for starting a business in the area were they lived. Ardichvili, Cardozo and Ray (2003) state that this perception can be seen as the ability of an entrepreneur to recognize an opportunity. Opportunity recognition as a distinctive characteristic is a key part of research in the field of entrepreneurship (Venkataraman, 1997). For instance, Bhave (1994) interviewed 27 entrepreneurs in New York to create a better image of what factors cause individuals to become entrepreneurs. He found that the decision to start a business is preceded by the process of opportunity recognition. Shane, Locke and Collins (2003) investigate the motivations of individuals that are involved in the entrepreneurial process. They state that the recognition of an entrepreneurial opportunity triggers the process to become an entrepreneur. Kirzner (1979) even states that opportunity perception is the most distinctive and fundamental characteristic of an entrepreneur.
Minniti and Nardone (2007) examine gender differences in entrepreneurial behavior. They use a special form of bootstrapping to create equal conditions for each individual. By doing this, Minniti and Nardone (2007) can analyze the decisions that men and women make in an identical environment. They state that males are more likely to start a business when males and females face identical opportunities. This does not necessarily mean that opportunity perception is more important in the decision to start a business for males than for females. Koellinger, Minniti and Schade (2008) state that men and women perceive situations in a different way. This intuition also holds for the perception of opportunities. So even though they face an identical situation, a male might see a good business opportunity while a female does not and vice-versa. Langowitz and Minniti (2007) do find gender differences in the way opportunity perception influences the decision to start a business. They state that the influence of opportunity perception on the decision to become self-employed is higher for males than for females, indicating that opportunity perception is less important for females in the decision to start a business. Unfortunately, Langowitz and Minniti (2007) do not test whether this difference is significant or not. The perception of good opportunities might be less important for females in their decision to start a business because they perceive higher opportunity costs than males from shifting their attention away from other matters in order to
seek good opportunities (Minniti and Naude, 2010). In addition, females generally have less access to valuable business opportunities (Fischer et al., 1993). This thesis might provide a clearer image of the gender differences in the way opportunity perception influences the decision to become an entrepreneur by examining whether this difference is significant.
As stated, several scholars conclude that the decision to become an entrepreneur is preceded by the recognition of an opportunity. In addition, scholars suggest that opportunity perception is less important in the decision to become an entrepreneur for females in comparison to males. From this the following hypothesis is determined:
H2: The effect of perceived opportunities on the decision to become an entrepreneur is
smaller for females than for males.
2.3 Knowing an entrepreneur
Knowing other entrepreneurs has a positive and significant effect on the decision to start a business (Arenius and Minniti, 2005; Koellinger, Minniti and Schade, 2005). Arenius and Minniti (2005) suggest that individuals that know an entrepreneur are 2.4 times more likely to become an entrepreneur themselves. According to them this positive impact of knowing other entrepreneurs might be caused by the fact that entrepreneurs are seen as role models. These role models stimulate and motivate individuals to start a business. A national survey amongst young adults in the USA draws the same conclusion (Development Associates, 1993). The survey finds that the expectation of owning a business in the future is positively influenced by having an inspiring role model. Several other scholars confirm this intuition (Walstad and Kourilsky, 1999; Wagner and Sternberg, 2004).
Another reason for the positive influence of knowing an entrepreneur on the entrepreneurial decision is that it broadens an individual’s network. These networks can help individuals with a lot of ‘how to’ type of questions that are involved with starting a business (Wagner and Sternberg, 2004). Similarly, Aldrich (1999) states that a broad network has a positive influence on an individual’s decision to start a business because it gives individuals access to social, emotional and material support, which enhances the individual’s confidence. Langowitz, Sharpe and Godwyn (2006) found that especially women that are involved in entrepreneurial activity appreciate role models and networks. In their paper, they examine Women’s Business Centres in the US, which are institutions that provide assistance to women that are starting a business. Langowitz, Sharpe and Godwyn (2006) state that females particularly make use of the support services that provide network opportunities and spotlight aspirational role models. Broader networks provide an emotional support system for females
and meeting women that are alike who succeeded in life gives them more confidence, which leads to women that are more comfortable in the business world.
Ascher (2012) discusses the rise of female entrepreneurship and examines the differences between male and female entrepreneurs by exploring several studies. He concludes that females have smaller networks than males and therefore have less support and knowledge to start a new business. Koellinger, Minniti and Schade (2013) also state that women are significantly less likely to personally know someone who started a business than men. They claim that this outcome reduces the likelihood of females to become self-employed.
In general, knowing other entrepreneurs has a positive influence on the decision to become self-employed. This particularly seems to hold for females, for whom role models and network opportunities play an important role in the decision to become an entrepreneur. Scholars even state that the lack of broad networks and aspiring models reduces the likelihood of females to start a business. Therefore, the following hypotheses is formed:
H3: The effect of knowing an entrepreneur on the decision to become an entrepreneur
is more positive for females than for males.
2.4 Perceived Capabilities
An individual’s believe to have the knowledge, skills and experience to start a business is defined as perceived capabilities. Koellinger, Minniti and Schade (2007) use a large cross-county sample obtained from surveys conducted by the GEM to examine what factors significantly influence different stages in the entrepreneurial process. They find that perceptual variables have a significant effect on the formation of businesses and state that an individual’s believe to have the right capabilities to start a business is the most important factor.
Research has shown that men and women differ in terms of self-confidence. Bengtsson, Persson and Willenhag (2005) used a sample of students from Stockholm University to determine if there are gender differences in the self-assessment of students. They conclude that males are more self-confident and optimistic than females. Corell (2001) examines if cultural beliefs about gender influence an individual’s perception of one’s abilities when making career-relevant decisions while controlling for actual ability. She concludes that the perception of one’s abilities is biased by general cultural beliefs about gender and task competence. For example, males are in general perceived to be better at math and because of this males believe that they are more competent in math than females do. This perceived competence of males in their mathematical skills causes more males to choose the
subject mathematics in comparison to females. Considering that entrepreneurship is seen as masculine (Baron et al., 2001), there is a good chance that male entrepreneurs are more confident about their knowledge, skills and experience to start a business than female entrepreneurs.
Verheul, Uhlaner and Thurik (2005) mention that the above intuition is indeed the case. They state that the masculinity associated with entrepreneurship has a negative effect on the entrepreneurial self-image of women. This causes fewer females to be involved in entrepreneurial activities. Another research paper that analyzes the difference in perceived entrepreneurial capabilities of males and females is the paper of Verheul, Thurik, Grilo and Van der Zwan (2011). As discussed before, Verheul et al. (2011) examine why the amount of female entrepreneurs is significantly lower than the amount of male ones by analyzing the entrepreneurial personality. Perceived capabilities is considered as a component of the entrepreneurial personality and according to Verheul et al. (2011) perceived capabilities are significantly lower for female entrepreneurs. This lack of confidence in skills, knowledge and experience partly causes the low amount of females starting a business.
The above findings suggest that perceiving to have sufficient skills to start a business plays an important role in the decision of females to start a business. Kickul, Wilson, Marlino and Barbosa (2004) find evidence for this suggestion. They question teenage boys and girls in the United States to determine whether differences in self-confidence across females and males might explain the gender gap in entrepreneurship. The results suggest that perceiving to have sufficient skills to start a business has a larger impact on the entrepreneurial interest for girls than for boys. This indicates that believing to have the skills, knowledge and experience to start a business is more important in considering entrepreneurship as a future career option for girls than for boys (Kickul et al, 2004). By focusing on individuals that already made the first steps in the entrepreneurial process, this thesis provides further insight to these findings. Lee, Wong and Ping Ho (2004) already examined the gender difference in the way perceived capabilities affects the decision to become an entrepreneur by using a large cross-country dataset provided by the GEM. They find that the perception to have sufficient skills has a stronger effect on the decision to become an entrepreneur for females than for males.
It is clear that believing to have the skills, knowledge, and experience to start a business has a significant influence on the decision to become an entrepreneur. Several scholars argue that females perceive to have less entrepreneurial capabilities and that because of this perception they are less likely to be involved in entrepreneurial activity than males. These suggestions are in line with the findings that perceived capabilities are more important
in the decision to become self-employed for females than for males. By taking the discussed findings into consideration, the following hypothesis is formed:
H4: The effect of perceived capabilities on the decision to become an entrepreneur is
more positive for females than for males.
3. Methodology
3.1 Data
This study uses Dutch individual-level survey data provided by the Global Entrepreneurship Monitor (GEM) for the period 2001-2012. GEM is a research program that describes and analyzes entrepreneurial processes around the globe. Established in 1999, it covered 10 countries while as of today the project has the most resourceful information about entrepreneurship, covering over 85 nations. GEM tracks the entrepreneurial attitudes, activity and aspirations of individuals and it uses a representative sample of at least 2,000 individuals per country for each year. The data collected by the GEM consists of two complementary tools, the Adult Population Survey (APS), which tracks the entrepreneurial traits of individuals, and the National Expert Survey, which monitors institutional factors that are believed to significantly impact entrepreneurship. Considering that this thesis analyzes the difference in the way that perceptions influence females to become entrepreneurs in comparison to males, the individual APS dataset is used.
As stated this thesis focuses on the Netherlands. Most papers that examine the personal characteristics that causes individuals to become entrepreneurs, base their analysis on all the countries that participate in GEM for a specific year. Unfortunately, it is almost impossible to correct for the specific influence of macroeconomic factors on the characteristics and perceptions of individuals that are involved in entrepreneurial activity (Langowitz and Minniti, 2007). Most researchers only control for country effects by adding country dummies to their regressions. But according to Arenius and Minniti (2005) this is not sufficient. This is especially the case when considering females since their employment choices are more dependent on the local environment (Minniti and Nardone, 2007). Because of this the intensity and directions of the outcomes may be influenced by macroeconomic conditions. Therefore, and especially because this paper is based on female entrepreneurship, it would be interesting to analyze just one country over time to see if the general believes about gender differences still hold. In addition, Ahl (2002) states that most studies focusing on female entrepreneurship are based on Anglo-Saxon countries, like the United States, New
Zealand, Australia and Canada. Accordingly, McManus (2000) recommends new research that considers gender differences in entrepreneurship to focus on non-Anglo-Saxon industrial nations. He states that although Western industrial states are similar, the development of female entrepreneurship is likely to differ in Western European countries due to economic, social and cultural influences. By taking the Netherlands into consideration, this thesis contributes to the literature by examining the gender gap in entrepreneurship in a non-Anglo-Saxon developed European country. A few other papers have already studied entrepreneurship in the Netherlands. For example, Verheul and Thurik (2001) analyzed the impact of gender on financial capital using a panel of 2000 Dutch firms. Parker and Van Praag (2006) investigated the extent to which the performance of Dutch entrepreneurial ventures is affected by schooling and capital constraints at the time of startup. And, as discussed, also De Wit and Van Winden (1989) make use of a Dutch dataset to analyze the determinants of the choice between self-employment and paid-employment. Analyzing the difference in perceptual variables between genders will be a useful addition to the above papers.
Furthermore, the dataset that is used in this thesis has a time span of twelve years, namely from 2001 till 2012. By using multiple years of data, the influence of random fluctuations is limited. Bergmann and Sternberg (2007) examined the factors that cause individual’s to start a new business in Germany by using multiple years of GEM data as well. Likewise, Lamotte and Colovic (2013) use several years of data from the GEM to investigate if the age distribution of a population influences entrepreneurial activity. The total sample collected in the years 2001-2012 for the Netherlands consists of 10,175 individuals for whom complete data is available for the purpose of this thesis.
3.2 Methodology
The model that will be used in this thesis is inspired by the model that Arenius and Minniti (2005) used in their research paper. As discussed, Arenius and Minniti (2005) use a cross-country dataset to analyze the influence of demographic, economic and perceptual variables on the decision to become an entrepreneur. According to them, individuals particularly rely on perceptual variables when deciding to start a business. The model used in this thesis differs from the one of Arenius and Minniti (2005) since the main focus of this thesis is on the perceptual variables. More specifically, this thesis analyzes if the way in which perceptual variables influence the decision to become an entrepreneur differs between females and males. The demographic and economic variables are only added as controls. Below, the
As a measure for entrepreneurship, total early-stage entrepreneurial activity (TEA) is used. An individual is classified to be involved in early-stage entrepreneurial activity when he or she is either a nascent entrepreneur or an owner-manager of a new business. A nascent entrepreneur is an individual that is actively involved in creating a business that they will own or co-own and the phase directly thereafter is indicated as owning and managing a new business. An additional requirement to be classified as an nascent entrepreneur is that “the firm has not paid salaries, wages, or any other payments to the owners for more than three months” and to be classified as an owner-manager of a new firm it is required that that “the firm has paid salaries, wages, or any other payments to the owners for more than 3 months, but less than 42 months”.
Four independent dummy variables are used in the regression to see if the effects of subjective perceptions on the decision to become an entrepreneur are different for females than for males. To capture these differences, gender interaction effects are added to the perceptual variables. The first independent dummy is the variable suskill, which analyzes the question if an individual beliefs to have the knowledge, skill and experience to start a new business. Respondents were also asked if they personally knew someone who started a business in the past 2 years, which is specified by the variable knowent. The third independent perceptual variable is opport. Individuals were asked if in the 6 months following the survey, there would be good opportunities for starting a business in the area where they lived. Finally, individuals were asked if fear of failure would prevent them from starting a business, which is indicated by the variable fearfail. As stated, all four variables are interacted with a gender dummy. The gender variable equals 1 if the individual is female and 0 if male.
Several control variables are also included in the probit regression. The first one is age. The respondents were asked to indicate the range that best described their age. The response ranges were “18-24”, “25-34”, 35-44”, “45-54”, “55-64” and “65+”. The individuals falling within the “18-24” range are used as the reference category. Education is the second control variable. Respondents were asked to provide their highest educational attainment. Data was classified into four groups, namely into “Some secondary education”, “Secondary degree”, “Post secondary education” and “Graduate experience”. For the reference category “Some secondary education” is used. The next control variable is work status, which represents an individual’s main occupation. Work status consists of three categories, which are ‘Working full- or part-time”, “Not working” and “Retired or student”. In the probit regressions, “Working full- or part-time” is used as the reference category. The fourth control variable is household income, where the responses were classified into three groups. The
three groups consist of the upper, middle or lower third of the income distribution. For the reference category, the lowest third is used. Lastly, the regression controls for an individual’s previous experience as an entrepreneur. Individuals participating in the survey were asked if they had sold, shut down, quit or discontinued a business in the 12 months preceding the survey. The dummy variable equals 1 if the individual had previous experience as an entrepreneur and 0 if not.
The above-described model leads to the following three probit regressions:
TEA!"#$%&&= Φ (β0+ β1∗ Gender + β2∗ Opport + β3 ∗ Knowent + β4∗ Fearfail + β5 ∗
Suskill+ β6∗ Opport ∗ Gender + β7∗ Knowent ∗ Gender + β8∗ Fearfail ∗ Gender + β9 ∗ Suskill∗ Gender + β10∗ Education + β11∗ Work Status + β12∗ Experience + β13∗ Income+ β14∗ Age)
4. Results
4.1 Summary Statistics
The number of observations, the mean, standard deviation and the minimum and maximum values that the independent and dependent variables can have are included in the descriptive statistics of Table 1. Panel A in Table 1 shows the descriptive statistics for the total sample of female and male respondents and Panel B for the sample of female and male respondents that are involved in total early-stage entrepreneurial activity. As observed in Panel A, there are slightly more females in the total sample. Contrary, Panel B shows that the amount of females involved in early-stage entrepreneurial activity is considerably lower than the amount of males, indicating that there is a gender gap in entrepreneurship. Furthermore, the descriptive statistics show that individuals involved in entrepreneurial activity perceive less fear of failure, know more entrepreneurs, perceive more opportunities and show a notable higher amount of self-confidence. In addition, Table 2 shows the descriptive statistics of the control variables. Panel A shows the educational distribution of the total sample and the sample that only includes individuals that are involved in entrepreneurial activity according to the answer categories provided by the survey. As observed, females and males involved in early-stage entrepreneurial activity are slightly higher educated. Panel B shows that 94% of the individuals involved in entrepreneurial activity works full- or part-time, while only 65% of the total sample belongs to this category. Furthermore, Panel C shows that for both samples the most respondents belong to the 35-44 age category, although the percentage is higher for
activity. Additionally, the respondents who claimed to be involved in early-stage entrepreneurial activity are overrepresented in the higher tail of the income distribution. Contrary, no substantial differences in the income distribution for the total sample of males and females are observed. Lastly, individuals involved in early-stage entrepreneurial activity have more previous experience in entrepreneurship compared to the total sample as can be seen in Panel D of Table 2.
4.2 Correlations
Table 3 contains the correlation coefficients across all variables used in the model for the total sample of males and females. As shown, the predictor variables have a strong correlation (p < 0.001) with total early-stage entrepreneurial activity. Firstly, the table shows that females are less likely to become entrepreneurs (r = -0.09). Secondly and as expected, knowing an entrepreneur personally (r = 0.18), perceiving good opportunities (r = 0.10) and believing to have the skills, knowledge and experience to start a business (r = 0.25) are all positively related to being involved in entrepreneurial activity. Fear of failure on the other hand, is negatively correlated with early-stage entrepreneurial activity (r = -0.06). Furthermore, the Table 1: Descriptive Statistics of the Dependent and Independent Variables
Panel A: Total sample of males and females
Observations Mean SD Minimum Maximum
Total early-stage entrepreneurial activity 10,175 0.07 0.26 0 1
Gender 10,175 0.52 0.50 0 1
Opportunity perception 10,175 0.42 0.49 0 1
Knowing another entrepreneur 10,175 0.33 0.47 0 1
Fear of failure 10,175 0.23 0.42 0 1
Sufficient skill perception 10,175 0.45 0.50 0 1
Panel B: Sample of females and males involved in total early-stage entrepreneurial activity
Observations Mean SD Minimum Maximum
Gender 724 0.36 0.48 0 1
Opportunity perception 724 0.60 0.49 0 1
Knowing another entrepreneur 724 0.64 0.48 0 1
Fear of failure 724 0.14 0.35 0 1
likelihood of being involved in entrepreneurial activity decreases with age (r = -0.13) and unemployment status (r = -0.17). In addition, entrepreneurial activity is positively correlated with previous experience (r = 0.07), having a higher household income (r = 0.09), and a higher education level (r = 0.05).
As can be seen in the correlation matrix, the correlations between the dependent variables are moderate. The only exception is the correlation between age and work status, which is rather high (r = 0.59). Thus, it is expected that problem of multi-collinearity, which occurs when the explanatory variables are highly correlated with each other, will not interfere with the overall results.
Table 2: Descriptive Statistics of the Control Variables
Total sample Total sample
(% of n = 10,175)
Sample of total-early stage entrepreneurs
(% of n = 724)
Panel A: Education Panel A: Education
Some secondary 10 Some secondary 8
Secondary 55 Secondary 48
Post-secondary 29 Post secondary 37
Graduate 6 Graduate 7
Panel B: Work status Panel B: Work status
Full- or part-time 65 Full- or part-time 94
Not working 9 Not working 4
Retired/student 26 Retired/student 2
Panel C: Age Panel C: Age
18-24 years 6 18-24 years 5 25-34 years 14 25-43 years 21 35-44 years 23 35-44 years 36 45-54 years 20 45-54 years 24 55-64 years 17 55-64 years 12 > 65 years 20 > 65 years 2
Panel C: Household income Panel C: Household income
Lowest 33% tile 27 Lowest 33% tile 18
Middle 33% tile 35 Middle 33% tile 27
Upper 33% tile 38 Upper 33% tile 55
Panel D: Experience Panel D: Experience
Yes 2 Yes 6
Table 3: Correlation Matrix
1 2 3 4 5 6 7 8 9 10 11 1 Total entrepreneurial activity — 2 Gender -0.09* — 3 Opportunity perception 0.10* -0.08* — 4 Knowing an entrepreneur 0.18* -0.15* 0.23* — 5 Fear of failure -0.06* 0.05* 0.03* 0.03* —
6 Sufficient skill perception 0.25* -0.27* 0.16* 0.28* -0.09* —
7 Education 0.05* -0.05* 0.10* 0.12* 0.02 0.09* —
8 Work status -0.17* 0.09* -0.17* -0.26* -0.07* -0.22* -0.12* —
9 Experience 0.07* -0.04* 0.01 0.06* -0.01 0.11* 0.02** 0.00 —
10 Household income 0.09* -0.16* 0.13* 0.23* -0.01 0.22* 0.16* -0.31* 0.03* —
11 Age -0.13* 0.02** -0.20* -0.25* -0.09* -0.11* 0.07** 0.59* -0.01* -0.16* —
*Denotes statistical significance at the 1% level. *Denotes statistical significance at the 5% level.
4.3 Results
Overall, the descriptive statistics and the correlation matrix suggest that perceptual variables play a role in the decision to become an entrepreneur. Perceiving to have sufficient skills has the biggest impact on the decision to start a business, followed by knowing another entrepreneur, opportunity perception and lastly fear of failure. To analyze the gender difference in the way perceptual variables influence the decision to become self-employed, probit regressions are run. The main results, which are based on the total early-stage entrepreneurial activity measure, are discussed first. Secondly, the results from additional regressions are reviewed to check for robustness.
4.3.1 Main Results
Table 4 shows the results of the probit regressions on total early-stage entrepreneurial activity. Three different models are estimated, which all contain the same set of variables but differ in the way gender enters the model. In the first model, only males are considered, in the second only females and the third model accounts for both males and females.
Starting with the control variables in Models 1 and 2, it can be seen that males who have a higher level of education are significantly more likely to become self-employed while for females the effect is negative but insignificant. The likelihood of becoming an entrepreneur is higher for both males and females that are working full- or part-time in
comparison with their non-working counterparts. Having previous experience as an entrepreneur has a positive influence on the self-employment decision for both males and females, although the coefficient is only significant for males. Surprisingly, the likelihood of becoming an entrepreneur is lower for the individuals belonging to the middle- and high-income group than for the individuals belonging to the low-high-income group. This could be because these individuals have a higher outside option and thus more to give up when they want to become an entrepreneur, while at low-income levels individuals have not that much to lose and are more willing to take the risk (Kihlstrom and Lafont, 1979). The likelihood of being involved in total early-stage entrepreneurial activity for males increases until the 25-34 age group and decreases thereafter, although only the negative coefficient for the 55-64 age group is significant. For females, the likelihood of becoming self-employed increases until the 35-44 age category and decreases thereafter. However, none of the coefficients for the age categories in the female sample are significant.
Contrary to most of the control variables, all the coefficients of the included perceptual variables in Model 1 and Model 2 are statistically significant. As expected, perceiving to have good opportunities, knowing other entrepreneurs and perceiving to have the capabilities to start a business all have a positive impact on total early-stage entrepreneurial activity for both genders. Fear of failure, on the other hand, has a negative impact on the entrepreneurial activity for males and females. From the first two models it can be concluded that the effects of knowing an entrepreneur, perceiving good opportunities and believing to have the right skills, knowledge and experience are higher for males than for females. While the less positive coefficient for opportunity perception is in line with the existing literature (Langowitz and Minniti, 2007), the lower coefficients for knowing another entrepreneur and the perception to have sufficient skills to start a business are not. Apparently, role models, networks and self-confidence are less important in the decision to become an entrepreneur for females than for males. In addition, the coefficient for fear of failure is more negative for males than for females. This indicates that fear of failure has a higher negative impact on the decision to be involved in entrepreneurial activity for males than for females, while the opposite was expected. It should be noted though, that the differences in the coefficients for males and females are rather small for all perceptual variables apart from knowing another entrepreneur.
To see if the gender differences found in the magnitudes of the perceptual variables in Models 1 and 2 are significant, a gender interaction effect is added to the subjective variables in Model 3. As observed, only the interaction between gender and knowing an entrepreneur is
Table 4: Probit estimates on total-early stage entrepreneurial activity
Model 1 Model 2 Model 3
Variable (Male) (Female) (Both)
Education - secondary 0.09 (0.13) -0.14 (0.15) -0.02 (0.10) Education – post-secondary 0.33* (0.13) -0.08 (0.15) 0.15 (0.10) Education - graduate 0.14 (0.17) -0.23 (0.22) -0.02 (0.13)
Work status – not working -0.10
(0.17)
-0.41* (0.13)
-0.29* (0.10)
Work status – retired/student -1.06*
(0.16) -0.63* (0.19) -0.90* (0.12) Age 25-34 0.05 (0.14) 0.09 (0.18) 0.05 (0.11) Age 35-44 -0.05 (0.13) 0.19 (0.18) 0.04 (0.10) Age 45-54 -0.11 (0.14) -0.06 (0.17) -0.10 (0.11) Age 55-64 -0.31** (0.15) -0.14 (0.18) -0.26** (0.11) Age 65+ -0.26 (0.19) - - -0.45* (0.16) Hh income – middle 33% -0.22** (0.09) -0.12 (0.10) -0.16** (0.07) Hh income – upper 33% -0.19** (0.09) -0.10 (0.10) -0.14** (0.07) Experience - yes 0.51* (0.13) 0.18 (0.21) 0.42* (0.11)
Opportunity perception (opport) - yes 0.13**
(0.06)
0.11** (0.07)
0.14** (0.06) Knowing another entrepreneur (knowent) - yes 0.40*
(0.06)
0.21* (0.07)
0.40* (0.06)
Fear of failure (fearfail) - yes -0.31*
(0.08)
-0.29* (0.09)
-0.31* (0.08) Sufficient skill perception (suskill) - yes 0.95*
(0.09) (0.08) 0.90* (0.08) 0.94*
Gender*Opportunity perception -0.03
(0.09)
Gender*Knowing another entrepreneur -0.19**
(0.09)
Gender*Fear of failure 0.03
(0.12)
Gender*Sufficient skill perception -0.04
(0.11) Gender - female 0.06 (0.12) Constant -2.30* (0.21) -2.08* (0.24) -2.22* (0.17) *Denotes statistical significance at the 1% level.
**Denotes statistical significant at the 5% level.
statistically significant. This coefficient estimate is negative, meaning that knowing an entrepreneur has a smaller positive influence on the decision to become an entrepreneur for females than for males. The interaction effects for opportunity perception, fear of failure and perceived capabilities show very low and insignificant coefficients, indicating that men and women are not affected differently by these variables when making the decision to become self-employed. Overall, the main probit regression supports none of the hypotheses.
It was expected that knowing an entrepreneur would more important in the decision to start a business for females than for males, but the results of this thesis suggest that the opposite holds. This finding might be due to the different networks that males and females have. According to Aldrich (1999) individual’s relationships can be classified according to strength. A person’s relation with a casual acquaintance is considered as weak, while a relation with a personal friend or with family is considered as strong (Aldrich, 1999). The strength of these relationships influences the decision to become an entrepreneur with different magnitudes. Granovetter (1973) argues that weak relationships are often more important for an individual than strong ones. He states that individuals that have a strong relationship with each other often have similar networks and information. On the other hand, acquaintances introduce individuals to larger and more diverse networks, which may provide more unique information and resources (Granovetter, 1973). Renzulli, Aldrich and Moody (2000) state that females tend to have less diverse networks than males. Hence, knowing another entrepreneur might provide less benefits for females, which makes them less important in the decision to become an entrepreneur.
4.3.2 Robustness Checks
As stated, additional probit regressions are run to check for robustness. These additional regressions test to what extend the results remain the same when the motivations to become an early-stage entrepreneur are taken into account. According to the GEM project, there are two main motivations to become an entrepreneur, namely out of opportunity or necessity. An individual is classified as a necessity-driven early-stage entrepreneur if he or she has “no better choices for work” and is classified as an opportunity-driven early-stage entrepreneur if the individual “takes advantage of a business opportunity”.
Table 5 shows the models in which the individuals starting a business to take advantage of an opportunity are considered. Overall, the results from Model 1 and Model 2 are quite similar to the results discussed in Section 4.3.1. Although the magnitudes show some differences, the signs remain the same. The only exception is the coefficient of the post
Table 5: Probit estimates on opportunity-driven total-early stage entrepreneurial acivity
Model 1 Model 2 Model 3
Variable (Male) (Female) (Both)
Education - secondary 0.05 (0.14) -0.06 (0.16) -0.00 (0.10 Education – post-secondary 0.28 (0.14) 0.03 (0.17) 0.17 (0.11) Education - graduate 0.09 (0.18) -0.20 (0.24) -0.02 (0.14)
Work status – not working -0.24
(0.19)
-0.40* (0.15)
-0.34* (0.12)
Work status – retired/student -0.93*
(0.18) -0.61* (0.22) -0.81* (0.14) Age 25-34 0.05 (0.15) 0.15 (0.19) 0.08 (0.11) Age 35-44 -0.07 (0.14) 0.29 (0.18) 0.07 (0.11) Age 45-54 -0.13 (0.14) -0.05 (0.19) -0.11 (0.12) Age 55-64 -0.45* (0.16) -0.13 (0.20) -0.34* (0.12) Age 65+ -0.46** (0.21) - - -0.58* (0.18) Hh income – middle 33% -0.23** (010) -0.09 (0.11) -0.16** (0.07) Hh income – upper 33% -0.16 (0.09) -0.05 (0.11) -0.10 (0.07) Experience - yes 0.56* (0.14) (0.22) 0.14 (0.12) 0.44*
Opportunity perception (opport) - yes 0.15**
(0.06)
0.15** (0.08)
0.16* (0.06) Knowing another entrepreneur (knowent) - yes 0.37*
(0.06)
0.25* (0.08)
0.36* (0.06)
Fear of failure (fearfail) - yes -0.31*
(0.08)
-0.26* (0.09)
-0.31* (0.08) Sufficient skill perception (suskill) - yes 0.97*
(0.09) 0.85* (0.08) 0.96* (0.09) Gender*Opportunity perception -0.01 (0.10)
Gender*Knowing another entrepreneur -0.13
(0.10)
Gender*Fear of failure 0.06
(0.12)
Gender*Sufficient skill perception -0.11
(0.12) Gender - female 0.07 (0.13) Constant -2.33* (0.22) -2.33* (0.27) -2.35* (0.18) *Denotes statistical significance at the 1% level.
**Denotes statistical significance at the 5% level.
Table 6: Probit estimates on necessity-driven total-early stage entrepreneurial activity
Model 1 Model 2 Model 3
Variable (Male) (Female) (Both)
Education - secondary 0.10 (0.26) -0.07 (0.29) 0.04 (0.19) Education – post-secondary 0.19 (0.27) -0.21 (0.33) 0.05 (0.20) Education - graduate 0.05 (0.38) 0.16 (0.42) 0.11 (0.28)
Work status – not working 0.14
(0.31)
-0.53 (0.34)
-0.24 (0.22)
Work status – retired/student -0.42
(0.31) - - -0.62** (0.27) Age 25-34 0.24 (0.42) 0.08 (0.45) 0.43 (0.38) Age 35-44 0.42 (0.40) 0.11 (0.43) 0.52 (0.37) Age 45-54 0.34 (0.40) (0.43) 0.15 (0.37) 0.50 Age 55-64 0.51 (0.39) 0.25 (0.44) 0.63 (0.37) Age 65+ - - - - 0.09 (0.50) Hh income – middle 33% -0.17 (0.19) -0.28 (21) -0.23 (0.14) Hh income – upper 33% 0.48** (0.20) -0.39 (0.21) -0.45* (0.14) Experience - yes 0.10 (0.31) - - -0.03 (0.29)
Opportunity perception (opport) - yes 0.00
(0.13) (0.17) -0.17 (0.13) 0.00 Knowing another entrepreneur (knowent) - yes 0.35**
(0.14)
0.09 (0.18)
0.30** (0.14)
Fear of failure (fearfail) - yes 0.00
(0.16)
-0.11 (0.20)
0.01 (0.16) Sufficient skill perception (suskill) - yes 0.54*
(0.19) 0.55* (0.18) 0.53* (0.19) Gender*Opportunity perception -0.15 (0.21)
Gender*Knowing another entrepreneur -0.20
(0.21)
Gender*Fear of failure -0.11
(0.25)
Gender*Sufficient skill perception 0.04
(0.25) Gender - female 0.09 (0.24) Constant -3.20* (0.52) -2.68* (0.55) -3.24* (0.45) *Denotes statistical significance at the 1% level.
secondary category for the female sample, which turned positive but remains insignificant. The coefficients for perceived capabilities and knowing another entrepreneur are more positive for males than for females regardless of their opportunity-driven motivation. The same holds for the fear of failure rate, which again is more negative for males than for females. The coefficient for opportunity perception is the same for males and males. In contradiction to Section 4.3.1, all of the added interaction effects in Model 3 are insignificant, which indicates that males and females are not affected differently by perceptual variables when they are involved in opportunity-driven early-stage entrepreneurial activity. However, this might be due to the smaller sample size.
The results of the regression performed on necessity-driven early-stage entrepreneurial activity are shown in Table 6. Lee and Kam Wong (2005) argue that demographic, economic and perceptual variables may not have much impact on the decision to become an entrepreneur when the decision is driven by necessity. This suggestion is confirmed by the results presented in Table 6. With the exception of household income for males, none of the coefficients of the socioeconomic variables are statistically significant in the first two models. Looking at the perceptual variables, perceived capabilities has a significant coefficient for both genders, where the coefficient is slightly more positive for females than for males. Knowing an entrepreneur also has a significant positive effect for necessity-driven male entrepreneurs, but only a small and insignificant positive effect for necessity-driven female entrepreneurs. Opportunity perception and the fear of failure do not have an effect on the decision to become an entrepreneur out of necessity for males, but have a relatively large, but insignificant, negative effect on the decision of necessity-driven females. As observed, none of the interaction effects in model 3 are statistically significant. Thus, no significant gender differences are found in the way perceptual variables influence the decision to become an entrepreneur for necessity-driven individuals. These findings are in line with the suggestions of Lee and Kam Wong (2005) but they may also be caused by the small sample size considering that the amount of necessity-driven entrepreneurs in the Netherlands comprises less than 10% within the overall total early-stage entrepreneurship rate.
5. Conclusion and Discussion
This thesis examines the difference in the way perceptual variables influence the decision to become an entrepreneur between females and males. With the use of data from the General Entrepreneurship Monitor, probit regressions are executed to analyze the effect of perceived
capabilities, fear of failure, knowing an entrepreneur and perceived opportunities on total early-stage entrepreneurial activity. In order to capture the gender difference in the way perceptual variables influence the decision to become self-employed, three different variants of the empirical model are estimated. All models contain the same set of variables, but differ in the way gender is introduced to the model. The first model is estimated for males, the second for females and the last one considers both genders. By doing this, the magnitudes in which perceptual variables influence males and females can be determined separately. By adding a gender interaction effect to the perceptual variables in the third model, it can be tested whether the differences in the way perceptual variables influence males and females are significant.
The results show that for both males as females perceptual variables have a significant influence on the decision to become an entrepreneur. Consistent with literature, opportunity perception, knowing an entrepreneur and perceived capabilities have a positive influence on the decision to start a new business for both genders, while the effect of fear of failure is negative. In addition, a significant difference is found in the way knowing an entrepreneur influences males and females. Knowing an entrepreneur has a smaller positive effect on the decision to start a business for females than for males. However, when checking for robustness, this significant difference disappears. Overall, the way in which perceptual variables influence the decision to become self-employed tends to be the same for females and males, with the exception of knowing another entrepreneur.
Although no significant gender differences are found in the way perceived capabilities, opportunity perception and fear of failure influence the decision to become self-employed, the fact that these perceptual variables do significantly influence females to start a business can be an important feature for decreasing the gender gap in entrepreneurship. Perceptual variables are part of an individual’s set of characteristics and are established long before a person makes the decision to start a business. As a result, perceptions about entrepreneurship cannot be changed easily. However, by creating environments that stimulate positive perceptions towards entrepreneurship at early ages, female entrepreneurship can be encouraged. Governments should aim to create environments that strengthen the perception that it’s feasible and desirable to become an entrepreneur.
Despite the richness of the General Entrepreneurship Monitor, some limitations should be mentioned. First, variables not provided by the GEM may contribute to the decision to become self-employed. For example, self-employed parents are found to play a central role in the decision to become an entrepreneur regardless of gender (Matthews and Moser, 1996;
Verheul et al., 2012; Hout and Rosen, 2000). In addition, marital status and the number of children are found to be significant determinants of the choice to start a business and this particularly seems to hold for women (Edwards and Field-Hendrey, 2002; Stevenson, 1986). As a result, omitted variables may bias the estimated coefficients. Second, considering that perceptual variables reflect subjective perceptions instead of objective conditions, the results are likely to be biased. Busenitz and Barney (1997) state that entrepreneurs are less rational, which may lead to distortions in the perceptual variables.
The results of this thesis suggest that more research is required to explain the gender gap in entrepreneurship. As stated, the data provided by the GEM does not contain all relevant variables. Because of this the true relationship between the dependent and independent variables may differ from the ones presented in this thesis. Hopefully, a more complete dataset will become available that will encourage further examination of the gender gap in entrepreneurship. Furthermore, future research should try to use objective measures for the variables knowing an entrepreneurs, sufficient capabilities, fear of failure and opportunity recognition to reduce bias in the outcomes. The results of this thesis do not provide any insights to the question why knowing an entrepreneur is less important for females than for males in their decision to become an entrepreneur. Further research should examine this in more detail by asking more specific questions regarding the importance of the perceptual
Reference List
Acs, Z., Arenius, P., Hay, M., & Minniti, M. (2004). Global Entrepreneurship Monitor
Report. London, UK: London Business School.
Acs, Z.J., Bardasi, E., Estrin, S., & Svejnar, J. (2011). Introduction to special issue of Small Business Economics on female entrepreneurship in developed and developing economies. Small Business Economics, 37(4), 393-396.
Acs, Z.J., & Evans, D. (1994). The determinants of variations in self-employment rates across countries and over time. Working paper.
Aldrich, H. (1999). Organizations Evolving. California, USA: Sage Publications. Allen, E., Elam, E., Langowitz, N., & Dean, M. (2007). The GEM women’s report.
Wellesley, MA: Center for Women’s Leadership, Babson College.
Ardichvili, A., Cardozo R., & Sourav R. (2003). A theory of entrepreneurial opportunity identification and development. Journal of Business Venturing, 18(1), 105-123. Arenius, P., & Minniti, M. (2005). Perceptual Variables and Nascent Entrepreneurship. Small
Business Economics, 24(3), 233-247.
Ascher, J. (2012). Female Entrepreneurship: An Appropriate Response to Gender
Discrimination. Journal of Entrepreneurship, Management and Innovation, 8(2), 97-114.
Baron, R.A., Markman, G.D., Hirsa, A., & Murphy, K.R. (2001). Perceptions of Women and Men as Entrepreneurs: Evidence for Differential Effects of Attributional Augmenting.
Journal of Applied Psychology, 86(5), 923-929.
Bengtsson, C., Persson, M., & Willenhag, P. (2005). Gender and overconfidence. Economics
Letters, 86(2), 199-203.
Bergmann, H., & Sternberg, R. (2007). The changing face of entrepreneurship in Germany.
Small Business Economics, 28(2), 205-221.
Beugeldijk, S., & Noorderhaven, N. (2005). Personality Characteristics of Self-Employed: an Empirical Study. Small Business Economics, 24(2), 159-167.
Bhave, M.P. (1994). A process model of entrepreneurial venture creation. Journal of Business
Venturing, 9(3), 223-242.
Blanchflower, D.G., & Oswald, A.J. (1990). Self-employment and the enterprise culture. In R. Jowell, S. Whiterspoon and L. Brook (eds.), British Social Attitudes: The 1990
Report. Aldershot: Gower Press.
Blanchflower, D.G., & Oswald, A.J. (1998). What Makes an Entrepreneur? Journal of Labor
Economics, 16(1), 26-60.
Brush C.G., & Cooper, S. (2012). Female entrepreneurship and economic development: An international perspective. Entrepreneurship & Regional Development, 24(1), 1-6). Burke, A., FitzRoy, F., & Nolan, M. (2002). Self-Employment Wealth and Job Creation: The
Roles of Gender, Non-Pecuniary Motivation and Entrepreneurial Ability. Small
Business Economics, 19(3), 255-270,
Busentz, L.W., & Barney, J.B. (1997). Differences between entrepreneurs and managers in large organizations: Biases and heuristics in strategic decision-making. Journal of
Business Venturing, 12(1), 9-30.
Cooper, A.C., Woo, C.Y., & Dunkelberg W.C. (1988). Entrepreneurs’ perceived chances for success. Journal of Business Venturing, 3(2), 97-108.
Correll, S. (2001). Gender and the Career Choice Process: The Role of Biased Self- Assesment. American Journal of Sociology, 106(6), 1691-1730.
