Women’s Entrepreneurial Propensity: The Role of Perception and Gender Equality

Women’s Entrepreneurial Propensity:

The Role of Perception and Gender Equality

Marlene Victoria Guthke

S3572803

Supervisor: Dr. Samuele Murtinu

Co-Assessor: Dr. Maryse Brand

MSc Small Business & Entrepreneurship

Faculty of Economics & Business

University of Groningen

January 2019

Abstract

Table of Content

1. Introduction ... 1

2. Theoretical Framework ... 3

2.1 Behavioral Economics ... 3

2.2 Individuals’ Propensity to Start a Business ... 4

2.2.1 Environmental Factors ... 5

2.2.2 Demographic Factors ... 6

2.2.3 Social Capital Factors ... 7

2.2.4 Perceptional Factors ... 8

2.3 Role of Gender Equality ... 11

3. Data and Methodology ... 14

3.1 Data Sources ... 14 3.2 Variables ... 15 3.3 Methodology ... 18 4. Results ... 20 4.1 Descriptive Statistics ... 20 4.2 Mediation Results ... 21 4.3 Moderation Results ... 25 4.4 Additional Results ... 25 5. Discussion ... 27 5.1 Theoretical Implications ... 27 5.2 Policy Implications ... 29

5.3 Limitations & Future Research ... 29

6. Conclusion ... 30

References ... 32

Appendices ... 40

I. Joint Test of Significance ... 40

II. Risk Attitude as Control Variable and as Proxy for Gender ... 41

III. Gender Equality Index as Control Variable ... 43

1

1. Introduction

Women are worldwide on average less likely than men to be entrepreneurs. Based on recent data, women report equal or higher entrepreneurship rates than men in only three out of fifty-four economies. Europe shows worldwide the lowest rates for female early-stage entrepreneurial activity (GEM Consortium, 2017).

Therefore, improving the number of female entrepreneurs is a crucial equality goal of the European Union with regard to its important contribution to equal economic independence of women (European Commission, 20181).

However, in terms of economic and social wealth, a general increase of entrepreneurs is not necessarily beneficial. Thus, it needs to be clearly understood whether and how more female entrepreneurship contributes to overall economic and social wealth. Up front, by considering how entrepreneurship strengthens economic and social wealth. Empirical evidence shows that entrepreneurs that recognize opportunities and establish long-term existing ventures mainly drive economic development and employment growth (Van Praag & Versloot, 2007; Vivarelli, 2013; Fritsch & Noseleit, 2013). Recalling the above gender issue: why should less gender difference in entrepreneurship stimulate a rise in qualified and valuable ventures?

Overall, women appear to be more risk averse and less overconfident than men (Charness & Gneezy, 2012; Huang & Kisgen, 2013). Therefore, scholars emphasize that female entrepreneurs desire to grow in a controlled fashion, which may result in ventures that are able to ‘out-survive’ those headed by entrepreneurs pursuing risky and high-growth strategies (Cliff, 1998). The distinct characteristics of female entrepreneurs may lead to more careful decision-making concerning starting and developing a business, and pursuing a more sustainable growth. While self-employed men and women differ very little in their skills, distinct differences appear in business goals and management styles (Brush, 1992). In this respect, recent evidence found women to be more likely than men to not prioritize economic goals but to use their businesses as a vehicle for social and environmental change (Hechavarría et al., 2017). Concluding, the different goals driving women in their business activities might lead to more businesses pursuing a social or environmental cause. As such, enable possible improvements in social disparities and towards sustainability. The encouragement of women to start qualified ventures is, subsequently, also able to secure women’s economic independence, thus, further strengthening the aforementioned EU goal in gender equality.

2 However, while there is a reasonably good understanding of the factors influencing men to start a business, these factors show a limited ability to predict women’s likelihood to start a business (Delmar & Davidsson, 2000). Yet, there is no clear consensus on how the factors influencing the decision to pursue an entrepreneurial career differ for men and women. Recently, scholars identified perceptional factors to account for substantial difference in entrepreneurial activity between men and women (Arenius & Minniti, 2005; Langowitz & Minniti, 2007).

Following this approach, this study investigates if the perception towards entrepreneurship is able to explain the differences in entrepreneurial propensity between men and women. I posit that a better understanding of the factors that drive women’s entrepreneurial activity enables to increase their successful participation in entrepreneurship. As such, the findings contribute to the field of study by also investigating a mediation effect of perception on women’s entrepreneurial propensity. Thereby, I address the limitations of the previous research by examining institutional effects and considering that women’s greater risk aversion as possible driver of their negative perception.

Perception is subjective and individual-specific but highly influenced by the environmental conditions surrounding individuals (Kahneman, 2003). In this respect, gender inequality describes an essential factor within the environment of individuals that is inherently different across genders. In particular, as stereotypical expectations for gender and occupational roles have been shown to assimilate into individuals’ perceptions and behavior (Eagly & Carli, 2003; Gupta, Turban & Pareek, 2013). Nevertheless, the relation of gender equality and female entrepreneurship is quite understudied.

3 independent experts developed the indices and gender gaps within countries were adjusted to comparable levels of achievement.

Precisely investigating the influence of gender equality for female entrepreneurs may enable better understanding of reasons and mechanisms behind the different entrepreneurial behavior of women. Summarizing, I propose the following research question:

Does gender equality create an environment for improved perceptions by women, and is therefore an important determinant of women‘s entrepreneurial propensity?

The findings may offer implications for the development of more efficient policies and successful incentives to encourage qualified nascent female entrepreneurs. The last section will particularly emphasize this contribution.

Within the study, the terms entrepreneurship and self-employment are used interchangeable. One might argue that the two contain different degrees of risk, but as the focus is on the factors that drive individuals to start a business, only the decision to take any kind of business risk is relevant.

The paper is structured as follows: Section two provides an overview of the theoretical background on which the hypotheses that drive the analysis are grounded. The subsequent section outlines the data and elaborates on the research methodology. Next, the results are presented. This is followed by a discussion on the findings including theoretical and policy implications, and the study’s limitations resulting in future research directions. Lastly, the research contributions are summarized in the conclusion.

2. Theoretical Framework

2.1 Behavioral Economics

4 economics recognizes that humans’ decision-making deviate from standard economic models due to cognitive, psychological, cultural and social factors (Mullainathan & Thaler, 2000). Furthermore, it seems able to capture those concepts as it combines economic theory with approaches from psychology (Mullainathan & Thaler, 2000, Kahneman, 2003).

The core of behavioral economics is the concept of bounded rationality, which explains that individuals are limited by taking decisions. Thus, individuals’ subjective belief and choice will most likely be different from the optimal belief and choice assumed in rational-agent models (Kahneman, 2003). Indeed, behavioral economics recognizes that imperfect alternatives may nonetheless appeal to individuals due to their perception and intuition although it may not be in line with rational reasoning. This is especially important as this research builds on evidence that perceptional variables are decisive factors of entrepreneurial propensity (Langowitz & Minniti, 2007). Astebro et al. (2014) confirm that the approach enables to understand individuals’ perceptions (and misperceptions) on pursuing an entrepreneurial career. Therefore, behavioral economics is identified as the most suitable approach for this study.

2.2 Individuals’ Propensity to Start a Business

Gartner’s (1985) early investigation into entrepreneurship demonstrates the complexity of new venture creation. In particular, the great diversity among entrepreneurs is recognized that leads to inherently different factors driving the process of starting a business. Additional to the complexity of entrepreneurial characteristics, there is no clear consensus if and in what sense male and female entrepreneurs differ. Therefore, also the factors influencing the propensity to pursue an entrepreneurial career are indistinct.

5 In contrast, social feminist theory considers men and women to differ inherently. Still neither men nor women are regarded as superior. Instead, the differences between both are shaped by different experiences throughout the whole life determining different behavior, knowledge, and viewpoints. In comparison to liberal feminism, women’s discrimination is not overt. Instead, the experiential differences are deep embedded in culture. In terms of entrepreneurship, result these experienced differences in distinct entrepreneurial characteristics and preferences (Fischer, Reuber & Dyke, 1993).

Both theories outline important implications on the drivers of men and women towards entrepreneurship. While social feminism may explain the different entrepreneurial propensity of men and women rather through stereotypes shaping different experiences, resulting in different traits, liberal feminism may concentrate more on the barriers drawing women away from entrepreneurship

2.2.1 Environmental Factors

Literature recognizes the importance of macro-environmental factors that influence entrepreneurial behavior and contributes to the heterogeneous nature of entrepreneurship across countries. Empirical studies are consistent about significant differences in the extent and level of new firm creation across countries (Blanchflower, 2004; GEM Consortium, 2017). In particular, the cultural, legal, political, and social environment in which the business is started needs to be considered when investigating entrepreneurial propensity. Additionally, the available resources, human capital and technological development in the respective countries potentially influence overall entrepreneurial propensity (Welter & Smallbone, 2011; Terjesen, Hessels & Li, 2016).

In general, entrepreneurial activity decreases with a higher level of economic development. On the one hand, is this found to be regardless of gender, as both women and men seem to prefer secured jobs rather than starting their own businesses when diverse employment opportunities are available. On the other hand, diminishes the gender gap in entrepreneurial activity with the increasing economic development (Kelley et al., 2017).

6 Several scholars address that cross-cultural differences may strongly affect female entrepreneurship (Steensma, Marino, Weaver & Dickson, 2000; Del Junco & Brás-dos-Santos, 2009; Kreiser, Marino, Dickson & Weaver, 2010). Furthermore, entrepreneurial characteristics are found to vary across countries, which suggests possible differences also for women’s entrepreneurial characteristics across countries (Terjesen, Hessels & Li’s, 2016).

Concluding, research should recognize macro-environmental differences that drive new venture creation, and especially women entrepreneurship. Otherwise, the complexities of entrepreneurial process may be not understood (De Bruin, Brush & Welter, 2006).

2.2.2 Demographic Factors

The decision to start a business shows correlation to several demographic variables such as age, work status, education, as well as household income (Blanchflower, 2004).

The age of individuals and entrepreneurial behavior describes an inverted u-shaped relationship across gender, with the highest tendency to engage in entrepreneurship between 25 and 34 years (Levesque & Minniti 2006; GEM Consortium, 2017).

Besides, several researchers proofed the importance of work status in the decision-making for entrepreneurship. Overall, employed individuals show a higher likelihood to become entrepreneurs. The desire for greater independence through entrepreneurship is hereby recognized as relevant driver (Taylor, 1996; Blanchflower & Oswald, 1998).

7 The decision to start a business is also related to individuals’ income and wealth. Low income may cause insufficient capital to start a business (Evans & Jovanovic, 1989). Reversely, the availability of income weakens financial constraints. Especially small and medium enterprises are the most constrained. Compared to large enterprises, they have less access to formal sources of finance (Beck, Demirgüç‐Kunt & Maksimovic, 2005; Beck & Demirgüç‐Kunt, 2006). As SMEs form a large part of the private sector in many countries, the constraints of financing a new business are applicable to numerous individuals considering entrepreneurship.

The role of financial constraints is especially discussed for female entrepreneurs (Carter, Williams & Reynolds, 1997; Carter & Rosa, 1998; Blanchflower, 2004). Likewise, recent evidence proves that female start-ups experience barriers and gender specific bias when trying to acquire start-up capital (Verheul & Thurik, 2001; Kanze, Huang, Conley & Higgins, 2018). It is also worth noting, that marital status and having children are factors with stronger influence for women than for men. Koellinger, Minniti & Schade (2013) argue based on multiple studies of developed and developing countries that women face higher opportunity costs than men due to their still more pronounced family responsibility.

2.2.3 Social Capital Factors

Strong and consistent predictor for nascent entrepreneurship are found in certain aspects of individuals’ network. Moreover, individuals involved in entrepreneurship possess a higher social capital. Especially, relationships described as strong ties in a network have important influence. Family members and close friends, who are business owners, as well as their active encouragement, increase the likelihood to favor an entrepreneurial career (Davidsson & Honig, 2003). In particular, the importance of parental role models is supported by multiple evidence (Scherer, Adams, Carley & Wiebe, 1989; Chlosta, Patzelt, Klein & Dormann, 2012; Bosma, Hessels, Schutjens, Van Praag & Verheul, 2012).

8 Particularly for women networks seem to be highly important. In the context of organizational settings, one of the most frequently reported issue faced by women is their limited access to or even exclusion from informal networks (Ibarra, 1993). In this respect, investigations on women business centers in the US emphasizing networks for women, are found to be highly successful (Langowitz, Sharpe & Godwyn, 2006).

2.2.4 Perceptional Factors

The concept of subjective perceptions as influence factors is strongly linked to behavioral economics, which claims that it is incorrect to assume that decisions are entirely determined by the assessed utility of a final outcome. Instead, a decision is dependent on the perception of all involved consequences (Kahneman, 2003).

Perceptual variables are not unknown in economic literature and have been considered before for new business creation (Gatewood, Shaver, & Gartner, 1995). Acs & Szerb (2006) reviewed scholar’s policy implications based on GEM data from the years 2001 to 2005; they found a shared view of the investigated papers expressing the value of perceptual variables as relevant factor to foster entrepreneurship and economic growth.

Several researchers claim that perceptual variables are crucial for female entrepreneurs, as they found that demographic and socioeconomic factors are not sufficient in explaining entrepreneurial differences across genders (Arenius & Minniti, 2005; Langowitz & Minniti, 2007; Minniti & Nardone, 2007).

Opportunity Recognition

9 experience as well as less experience of being involved in a business venture than male business founders (Kalleberg & Leicht, 1991; Boden & Nucci, 2000). However, DeTienne & Chandler (2007) suggest that women and men utilize different processes of opportunity identification due to their different human capital, but neither process is superior.

Still women seem to be less likely than men to perceive opportunities. Although there is no evidence of any objective reason for a lower ability of women to recognize opportunities. These gender differences are differently pronounced across countries, and therefore assumed to be dependent on cultural factors (Arenius & De Clercq, 2005).

Interestingly, recent findings indicate that women report greater opportunity evaluation when business opportunities are described with more feminine than masculine stereotypical attributes. Simply changing the content of information according to gender stereotypes has decisive impact of the respondents’ evaluation. Consequently, one may assume that gender stereotypes, present in society and business environment, importantly influence women’s opportunity recognition (Gupta, Turban & Pareek, 2013).

Self-Efficacy

The concept of self-efficacy can be summarized as someone’s perceived ability of performing a certain task (Bandura, Freeman & Lightsey, 1999). In context of entrepreneurship, Chen, Greene & Crick (1998) defined self-efficacy as an individual’s confidence in his or her ability to successfully perform entrepreneurial roles and tasks. Their empirical evidence showed that students’ entrepreneurial intention is positively related to their entrepreneurial self-efficacy. The relationship was further proofed by demonstrating that entrepreneurial self-efficacy distinguished small business executives from managers. Overall, there is general agreement that entrepreneurs show significantly higher levels of self-efficacy than non-entrepreneurs (Markman, Balkin & Baron, 2002).

Considering gender difference, men are found to have equal efficacy for occupations traditionally male- and female-dominated, while women show weaker self-efficacy for traditionally male-dominated occupations. This is especially striking, as both genders did not differ in their abilities (Bandura, Barbaranelli, Caprara & Pastorelli, 2001). The same evidence is found to be true for entrepreneurial self-efficacy (Scherer, Brodzinski & Wiebe, 1990; Chowdhury & Endres, 2005).

10 they believe they do. In addition, further reasons are discussed, such as less career experiences, less role models, or lower social support (Scherer, Brodzinski & Wiebe, 1990; Dyer, 1994). Zhao, Seibert & Hills (2005) found entrepreneurship-related education, previous entrepreneurial experience, and risk propensity as fully mediated by self-efficacy. However, they did not find gender to be mediated. Instead, gender showed a direct effect on intention, which indicates that other factors are involved than solely being man or woman.

Furthermore, Chen, Greene & Crick (1998) argue that positive entrepreneurial self-efficacy depends on the fact that an individual can identify itself with the associated behavior and characteristics of entrepreneurship. This may suggest again that gender stereotypes associations are closely related to entrepreneurship (Gupta, Turban, Wasti & Sikdar, 2009; Gupta, Turban & Pareek, 2013). In other words, women may identify themselves less with entrepreneurial traits as there are traditionally masculine stereotyped, resulting in a lower entrepreneurial self-efficacy.

Fear of Failure

Early research built the theory that in a competitive equilibrium less risk averse individuals choose entrepreneurship over non-risky employment, due to the higher risk involved in being a business owner (Kihlstrom & Laffont, 1979). Several recent studies confirm that the probability of entrepreneurship increases with risk tolerance (Stewart & Roth, 2001; Hvide & Panos, 2014). Weber & Milliman (1997) established the importance of perceived risk for the process of making risky decisions. In this light, direct effects for predicting the outcome of a risky decision are found to be rather inaccurate. Instead these effects seem to be mediated by individuals’ risk perception (Sitkin & Weingart, 1995). The differences between men’s and women’s behavior by dealing with risk is an extensively discussed research topic. Diverse evidence indicates that women are more risk averse than men. Non-business related evidence showed, for instance, that women are more risk averse when gambling (Levin, Snyder & Chapman, 1988). Further evidence from economic literature proofed women’s lower risk preference and lower competitive preference (Charness & Gneezy, 2012; Croson & Gneezy, 2009); both are considered as relevant characteristics of potential entrepreneurs.

growth-11 oriented female and male entrepreneurs are more similar than they are different. This suggests the same ability of women to manage a growing company, while their significantly lower risk preference is still present. One may conclude that fear of failure when starting a business might be more pronounced for women due to greater risk aversion. Nevertheless, the cause seems to be in a negative perception instead of the actual higher likelihood of failure.

Based on the reviewed literature, I identified four groups that influence entrepreneurship: demographic, social capital, (macro-) environmental and perceptual factors. Focusing on female entrepreneurship in developed countries, the minor differences in demographic factors of men and women may be negligible (Kelley et al., 2017). In addition, Lefkowitz’s (1994) study revealed how traditional differences of men and women in job preferences disappeared after controlling for the spurious effects of gender-related differences in perceived job characteristics, occupational level, education, and income. Furthermore, recent literature observed that women’s subjective perceptions have a crucial influence on entrepreneurial propensity (Arenius & Minniti, 2005; Koellinger & Minniti, 2006; Langowitz & Minniti, 2007). In fact, even when controlling factors in the social environment women perceive themselves and their business environment in a less positive light than men do (Langowitz & Minniti, 2007). Additionally, due to the approach of behavioral economics, I consider individuals’ limited rationality as central and consequently the possible gap between subjective beliefs and actual reality. This leads towards the assumptions that women’s perception may be the key driver for their entrepreneurial propensity. More precisely, the perception of women may reflect rather their negative subjective beliefs than their actual abilities. In other words, being a woman may have a direct effect on entrepreneurial propensity, but this effect may actually be explained by (i.e. be mediated by) women’s perception. Therefore, the following hypothesis is derived:

Hypothesis 1: Women are more likely than men to have a negative perception towards

entrepreneurship, and are therefore less likely to be inclined to start a business.

2.3 Role of Gender Equality

Inequality as Barriers for Female Entrepreneurship

12 no direct predictor for the proportion of female entrepreneurs in a country (Baughn, Chua & Neupert, 2006; Safaraz, Faghih & Majd, 2014).

In countries with rather low economic development, women’s entry into self-employment may be easier than overcoming barriers to enter formal employed jobs (Carter & Marlow, 2003; Minniti & Arenius, 2005). Hence, labor market discrimination or “glass ceiling” career inhibitors may result in increasing female entrepreneurship. In the same light, it is argued that women experiencing inequality are eager to achieve economic parity by pursuing entrepreneurship (Lerner, Brush & Hisrich, 1997), resulting again in less gender differences in entrepreneurship.

Nevertheless, scholars also discuss reasons how gender equality enhances female entrepreneurship. From the perspective of liberal feminist theory, high gender equality should lead to the disappearance of society’s discouragement of women. Moreover, structural barriers, such as prior discussed disadvantages in access to finance or the exclusion from networks, should vanish with gender equality (Fischer, Reuber & Dyke, 1993). Indeed, evidence indicates that gender equality enables a greater extent of assets in female ownership, and positively affects higher educational levels of women as parents invest more into their daughters’ education (Kantor, 2002). Furthermore, increased gender equality is found to enhance the freedom of women from domestic responsibilities (Cliff, 1998).

Inequality as Preserver of Gender Stereotypes

13 Baughn, Chua & Neupert (2006) argue that support for women entrepreneurs is embedded in a society’s larger socioeconomic role system, which is expected to be more equal when the country’s governmental system exhibit gender equality. The mentioned role system may also reflect gender stereotypes. Hence, higher gender equality seems to be accompanied by a less stereotyped view on entrepreneurs. The result may be women that identify themselves more with entrepreneurs, and thus, have more confidence in owning entrepreneurial traits and being able to recognize and pursue entrepreneurial opportunities. In this perspective, research on female leadership proved that due to the incongruity of the traditional female role and many leader roles, women are discouraged to express their leadership competency (Eagly & Carli, 2003). To sum up, with increased gender equality women may relate differently to entrepreneurship and consequently have a more positive perception towards starting a business. Therefore, I posit the following hypothesis:

Hypothesis 2.1: Gender equality positively moderates the relation between being a woman and

perception towards entrepreneurship.

However, the turning point seem to be when entrepreneurship is viewed as gender neutral. Since Gupta, Turban & Bhawe‘s (2008) evidence shows that men and women report similar intentions if entrepreneurship is presented as gender neutral (Gupta, Turban & Bhawe, 2008). Summarizing, if a country reflects gender equality, entrepreneurship may be seen as gender neutral resulting in equal entrepreneurial propensity of men and women. Also taking into account the aforementioned arguments of liberal feminist theory, which emphasizes that it is the turn of the public sphere to enable women’s advancement (Marlow & Patton, 2005), I suggest the following hypothesis:

Hypothesis 2.2: Gender equality positively moderates the relation between being a women and

14

3. Data and Methodology

3.1 Data Sources

The study builds on data from four different sources. Firstly, the data set 'Adult Population Survey' (APS) provided by the Global Entrepreneurship Monitor is the base of the study. The used data was collected in 2014. Standardized surveys were conducted from a sample of at least 2000 participants per country.

Secondly, a further data set realized by the Global Entrepreneurship Monitor is used. The ‘National Expert Survey’ (NES) looks at the national context linked to conditions that enhance (or hinder) new business creation, so-called Entrepreneurial Framework Conditions (EFCs). Its main methodological difference to similar surveys is that it focuses not on general economic factors but on EFCs that directly affect entrepreneurial activities. This makes the data especially suitable for studies in the field of entrepreneurship. As for the previous data set, I use the year 2014.

Thirdly, data from OECD and Eurostat are used to measure men’s and women’s attitude towards entrepreneurial risk in the member countries. Only data from the year 2013 was available.

15 complete gender equality, and one for full gender inequality2_{. The latest edition is used from}

the year 2017.

After matching the four data sources the study sample consists of 69.138 observations from 20 countries (Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Luxembourg, Netherlands, Poland, Portugal, Spain, Slovakia, Slovenia, Sweden, United Kingdom).

3.2 Variables

Dependent variable. In order to measure entrepreneurial propensity, I construct the variable

describing nascent entrepreneurs as proposed by the GEM consortium and used before in the same context by Langowitz & Minniti (2007). The respondents of the GEM APS are coded as

nascent entrepreneur if they were alone or with others in the process of starting a business,

have committed resources to it, and expect to own at least part of it. This definition is required to separate individuals who only considered entrepreneurship from those that were committed to it and further to separate possible managers from entrepreneurs.

In addition, I exclude those respondents that are coded as nascent entrepreneurs but answered that they were involved in start-up activities because of no better choices of work (necessity entrepreneurs). This additional construct is used to only include individuals that engage into new venture creation to take advantages of a business opportunity (opportunity entrepreneurs). Hereby, the different motivations of these two types of entrepreneurs are taken into account. Specifically, the perception towards entrepreneurship plays a less dominant role for necessity entrepreneurs since they have little or no choice (Langowitz & Minniti, 2007).

The propensity to entrepreneurship (EntrProp) is thus a dichotomous variable (nascent, opportunity entrepreneur = 1; no entrepreneur = 0).

Independent variable. The predictor variable Gender is a dummy variable that equals one if the

respondent is a woman and zero if the respondent is a man. Women and men are equally represented in the sample.

Mediation variable. I use three variables to measure individuals’ perception towards

entrepreneurship. Respondents were asked to evaluate themselves and their entrepreneurial environment. In particular, the respondents had to assess if (1) they personally see opportunities

2 _{For more detailed information please see: https://eige.europa.eu/gender-equality-index/about, in particular}

16 for starting a business, (2) if they see themselves as having the sufficient ability, knowledge and skills to start a business, and (3) if their fear of failure affects the decision to start a business. The GEM consortium describes these questions as perceptions towards entrepreneurship, and Arenius & Minniti (2005) as well as Langowitz & Minniti (2007) have used them in the same context by measuring individuals’, in particular women’s, perception. Answers to all the three questions are dichotomous: yes or no. The three dummy variables are computed into one categorical variable (Percep) describing individuals’ perception towards entrepreneurship (0=negative perception, 1=rather negative perception, 2=rather positive perception, 3=positive perception).

Moderation variable. The Gender Equality Index (GEI) measures gender equality in six areas

(work, money, knowledge, time, power, and health) using 31 indicators. All indicators are combined into a single measure for each EU member state, forming the Gender Equality Index. The variable representing GEI is continuous at the country-level.

Control variables. Considering the diverse influence factors on the decision of starting a

business as discussed previously, the study takes three groups of control variables into account. In all analyses, I control for a set of demographic factors, namely age (Age), work status (WorkStatus), income (Income), and education (EducLevel). All variables are categorical. Further I control for social capital factors by using the dummy variable KnowEnt, which equals one if the respondent indicated “yes” to the question if “he/she knows another entrepreneur”, and zero otherwise.

Moreover, I control for country-specific effects by introducing dummy variables for each country in the sample (Name of Country), which gives an initial investigation of the influences of macro-environmental factors.

17 growing firms on the national and local government level (ranging from completely false to completely true).

Additionally to the three previous groups of controls, I apply individuals’ attitude towards entrepreneurial risk (RiskAtt) based on OECD and Eurostat Data as control variable. Furthermore, the dependent variable Gender is substituted by RiskAtt, accordingly I assume that gender is a proxy for risk attitude. This is reasonable, as there is general agreement among researchers for the close correlation between individuals’ gender and their risk attitude, whereby women show a substantially lower risk attitude than men Charness & Gneezy (2012). While Gender is a binary variable only differentiating men and women, the newly introduced variable RiskAtt is numerical with various outcomes for men and women at the country-level.

Table 1 Description of Variables

Variable Definition Possible values

EntrProp Dummy that equals one if respondents are (1) currently in the process of starting a business, (2) have committed resources to it, (3) expect to own at least part of it, and (4) are involved in this start-up to take advantage of a business opportunity.

Yes, No

Gender Dummy that equals one if respondent is a women. Yes, No

Percep Categorical that summarizes if respondents believe that, (1) in the next six months, good business opportunities will exist in the area where they live, (2) they have the knowledge, skill and experience required to start a new business, and (3) fear of failure would prevent them from starting a new business. Categories were coded as followed:

Negative,

rather negative, rather positive, positive

0, if respondents answered no to all questions 1, if respondents answered yes to one questions 2, if respondents answered yes to two question 3, if respondents answered yes to all questions

GEI Continuous variable indicating Gender Equality Index of country 0-100

Age Categorical indicating the age group of respondents Below 18

18–24 years old 24–34 years old 35–44 years old 45–54 years old 55–64 years old Above 65

WorkStatus Categorical indicating respondents’ current occupational status Not working Student Homemaker Retired / disabled Part time only Full: full or part time

Income Categorical indicating respondents’ household income Lowest 33%

18 Gender Entrepreneurial Propensity Path c / c’ Entrepreneurial Perception Path a Path b

EducLevel Categorical indicating highest educational degree obtained by respondents None Some Secondary Secondary Degree Post Secondary Grad Exp GuvSupPol National experts indicated the degree to which their home-country’s

policies consistently favor new firms, and prioritize support for new and growing firms on the national and local government level.

9 point scale ranging from completely false to completely true RiskAtt Percentage of women and of men declaring that they would rather take

a risk and start a new business than work for someone else.

0-100

3.3 Methodology

Mediation Analysis

Considering the above suggested hypothesis, it becomes clear that two analyses have to be carried out. Firstly, I use the methodology put forward by Baron & Kenny (1986), in order to test whether the relationship between gender and entrepreneurial propensity is mediated by the perception towards entrepreneurship (Hypothesis 1). The first step is to identify whether a significant relationship exists between gender and entrepreneurial propensity, which corresponds to path c in Figure X. Secondly, there may be a significant relationship between gender and perception towards entrepreneurship (path a in Figure X). Thirdly, perception may be significantly related to entrepreneurial propensity when included as independent variable next to gender (path b in Figure X). During this last step, the effect of gender on entrepreneurial propensity might turn insignificant when perception is included in the model (path c’). This would indicate a full mediation: gender is related to entrepreneurial propensity but only through perception. However, if the effect of gender on entrepreneurial propensity becomes weaker, (but still significant) there is partial mediation. This would indicate that gender is directly related to entrepreneurial propensity, but also indirectly through perception.

The described mediation paths together with the prior derived hypothesis are graphically presented in Figure 1.

Figure 1: Conceptual Model 1.1

19 The described model and tests are examined by using logistic regression. The three tests can be expressed in terms of three equations, which represent the three above described paths:

Path c: EntrProp' = cGender + E1+ Xdemo.; social; macro-environm.

Path a: Percep' = aGender + E2 + Xdemo.; social; macro-environm.

Path b + c’: EntrProp" = bPercep + c'Gender + E3 + Xdemo.; social; macro-environm.

Xdemo.; social; macro-environm. is a set of control variables, more precisely, the above described

demographic, social capital and environmental factors (incl. country dummies and institutional variable). The country dummies and the institutional variable (GuvSupPol) are both at country-level. More precisely, the data is not only collected on the same level of analysis, each country is one-to-one matched with its institutional score. If both were included in the same regression, perfect collinearity between these two variables would occur. Therefore, two separate tests for each of them are conducted.

As additional robustness check, the same mediation test is performed with entrepreneurial risk attitude included among the control variables, as well as with entrepreneurial risk attitude as supposed proxy for gender.

Moderation Analysis

In order to test the possible moderation effect of gender equality, I use as well the framework of Baron & Kenny (1986). Firstly, I center the independent variables and the moderation variable to ensure that collinearity is not affecting the moderation analysis. Secondly, I compute the interaction effect of gender and equality (gender*GEI). Lastly, based on the assumption that the moderator effect is linear, ordinary least squares regression (OLS) is identified as the appropriate method to test the two relationships (Baron & Kenny, 1986). In other words, increasing gender equality is assumed to affect the relation by linearly decreasing the negative effect of gender on entrepreneurial propensity as well as on entrepreneurial perception. The same control variables are included in the regression. However, because the moderator variable

GEI is associated with each country, it is not feasible to include simultaneously the control

20 The described moderation effect with the prior derived hypotheses are graphically presented in Figure 2.

Figure 2: Conceptual Model 1.2

4. Results

From all respondents only 2,9% (in sum 2028 individuals) could be coded as nascent entrepreneurs due to opportunities, and therefore are indicating an entrepreneurial propensity. On average, nearly twice as many men as women indicated entrepreneurial propensity. More precise, 1.8 men for every women. The gender difference for perception towards entrepreneurship is slightly less, although men are still about 50% more likely to have a fully positive perception. The Gender Equality Index ranges from 50.0 points to 82.6 points with Greece having the lowest score while Sweden having the highest. The EU-28 average is 66.2 points, while the average score of the 20 countries used in this study is 66.4.

Further descriptive statistics of the variables of interest are shown in Table 2, while Table 3 reports the correlation matrix. The first column shows that entrepreneurial propensity is negatively related to being female. Furthermore, respondents’ age (as the average age of the sample is relatively high), and being unemployed, shows - regardless of gender - a negative relation to entrepreneurial propensity. All other variables are positively correlated, while perception shows the strongest positive correlation. By testing the Variance Inflation Factor, no collinearity among the independent variables is ensured.

21 Table 2 Descriptive Statistics

Number of Observations Mean S.D. Minimum Maximum

EntrProp 69138 0.03 0.169 0 1 Gender 69138 0.51 0.500 0 1 Age 69138 4.30 1.371 1 7 WorkStatus 65035 2.40 1.822 1 6 Income 53121 2.08 0.822 1 3 EducLevel 68603 2.23 0.960 0 4 KnowEnt 68120 0.34 0.472 0 1 Percep 54779 1.31 0.937 0 3 GEI 69138 0.66 0.071 0.500 0.826

Table 3 Correlation Table

1 2 3 4 5 6 7 8 1 EntrProp 2 Gender -0.052** 3 Age -0.048** 0.029** 4 WorkStatus -0.060** 0.111** -0.108** 5 Income 0.038** -0.125** -0.014** -0.278** 6 EducLevel 0.067** 0.006 -0.088** -0.188** 0.291** 7 KnowEnt 0.133** -0.068** -0.116** -0.091** 0.115** 0.117** 8 Percep 0.174** -0.163** 0.016** -0.135** 0.153** 0.156** 0.237** 9 GEI -0.024** 0.000 0.087** 0.008* 0.109** 0.037** 0.002 0.074** 4.2 Mediation Results

Table 4 reports the results obtained for hypothesis 1 based on the three regression equations. It was investigated if individuals’ perception mediates the effect of gender on entrepreneurial propensity.

Firstly, the direct effect are examined, presented under path c (see Figure 1). Gender and

EntrProp show statistically significant coefficients. The result establishes the significant

relationship between gender and entrepreneurial propensity. The negative beta of Gender constitutes a negative effect on the likelihood of being involved in starting a business when the gender is female.

22 gender on the perception towards entrepreneurship. This implies that the women in the sample compared to men have a more negative perception towards entrepreneurship.

Thirdly, the indirect effect is examined, presented under path b + c’. The results indicate a significantly positive effect of Percep on EntrProp, which shows that individuals’ positive perception increases their entrepreneurial propensity. When controlling for path a and path b, the effect of Gender on EntrProp remains significant. It appears that there is no full mediation, and that perhaps multiple mediating factors are involved (Baron & Kenny, 1986). However, when comparing the effect of Gender before and after including perception into the model, it shows a decrease of the effect of Gender from -56 bps to -37 bps3: a change of 19 bps. This indicates partial mediation. Overall, the results of each path combined with the partial mediation effect, provides support for hypothesis 1, suggesting that women are more likely to have a negative perception towards entrepreneurship than men, and are therefore less likely to be inclined to start a business.

One major critique of Baron & Kenny’s (1986) method to test mediation is that the indirect effect (test of path a and path b) is logically inferred most likely to be nonzero but it is not yet jointly tested. Reacting to this limitation, bootstrapping is identified as valid and powerful method for testing the significance of the indirect effect (Hayes, 2009). Using 1000 bootstrap samples (with replacement), the bias-corrected confidence interval of the indirect effect shows significance at the 0.01 level. This gives substantial evidence for the mediation effect of hypothesis 1.

Looking at the demographic control variables in all three paths, I find all significant except for the income of the respondents. Interestingly, respondents’ income is not significantly related to entrepreneurial propensity but is significantly related to perception. Concluding, individuals’ financial situation slightly affects their perception but it does not determine the final decision on being inclined to entrepreneurship or not. This is rather contrary to the evidence in existing research.

The variable KnowEnt shows positive significant coefficients in all models. These estimates confirm that individuals who know an entrepreneur are more likely to have a positive perception towards entrepreneurship, and are more likely to have a propensity to entrepreneurship. The findings are in line with Langowitz & Minniti (2007) reporting that knowledge of other entrepreneurs has a significant impact on engaging in entrepreneurship.

23 After the variable Percep shows KnowEnt the strongest effect on entrepreneurial propensity. This result indicates that, next to individuals’ perception, individuals’ social capital may play an important role in assessing their propensity to entrepreneurship.

Eleven countries display significant effects across all three equations with mostly positive although very small betas. This finding suggests that the countries’ characteristics slightly increase the propensity towards entrepreneurship despite the effects of gender and/or individuals’ perception. The fact that more than half of the countries included in this study show a significant effect is in line with existing literature about the importance of country effects for entrepreneurial propensity as well as for perception towards entrepreneurship.

Surprisingly, the institutional variable (GuvSopPol) is statistically significant with a positive effect on individuals’ perception. However, there is no effect on entrepreneurial propensity. This leads to the assumption that country-specific policies positively influence the perception and consequently the likelihood of being involved in starting a business may increase. However, a country’s support for entrepreneurship appears not to belong to the direct factors determining entrepreneurial propensity.

Controlling for institutional effects also allows testing the robustness of the results. In fact, adding GuvSopPol did not change the results of the independent variables Gender and

Percep. Even though, GuvSopPol shows significant coefficients in path a, it does not influence

the main results: being a woman shows the same negative effect on perception towards entrepreneurship. I conclude that institutional effects do not drive the results.

Table 4 Regression Results for Mediation Effect

Path c Path a Path b + c’

B SE Sig. B SE Sig. B SE Sig.

24 Environmental Controls Greece 0.071 0.028 0.011 -0.011 0.001 0.000 0.109 0.029 0.000 Netherlands 0.115 0.023 0.000 0.018 0.001 0.000 0.092 0.025 0.000 Belgium 0.017 0.031 0.583 -0.001 0.001 0.421 0.022 0.034 0.515 France 0.022 0.029 0.443 -0.002 0.001 0.141 0.043 0.031 0.160 Hungary 0.110 0.026 0.000 0.000 0.001 0.805 0.144 0.027 0.000 Italy 0.106 0.028 0.000 -0.006 0.001 0.000 0.136 0.030 0.000 Austria 0.194 0.026 0.000 0.025 0.002 0.000 0.132 0.030 0.000 UnitedKingdom 0.152 0.025 0.000 0.017 0.001 0.000 0.141 0.027 0.000 Denmark 0.031 0.029 0.286 0.015 0.001 0.000 0.015 0.031 0.625 Sweden 0.113 0.025 0.000 0.027 0.001 0.000 0.077 0.029 0.009 Poland 0.052 0.023 0.027 0.002 0.001 0.209 0.073 0.026 0.005 Germany 0.124 0.028 0.000 0.017 0.002 0.000 0.098 0.030 0.001 Portugal 0.145 0.025 0.000 0.005 0.001 0.001 0.150 0.026 0.000 Luxemburg 0.113 0.024 0.000 0.009 0.001 0.000 0.148 0.028 0.000 Ireland 0.059 0.026 0.025 0.008 0.001 0.000 0.043 0.028 0.120 Finland 0.016 0.027 0.565 0.005 0.001 0.000 0.012 0.031 0.702 Estonia 0.168 0.025 0.000 0.009 0.001 0.000 0.171 0.028 0.000 Slovenia 0.035 0.030 0.253 0.002 0.001 0.172 0.034 0.035 0.329 Slovakia 0.097 0.023 0.000 0.004 0.001 0.005 0.103 0.025 0.000 GuvSupPol -0.030 0.079 0.699 0.056 0.005 0.000 -0.068 0.089 0.443 Constant -3.895 0.151 0.000 0.285 0.008 0.000 -4.904 0.174 0.000

Risk Attitude as Additional Control Variable and Alternative Proxy for Gender

The fact that women are found to have a substantially lower risk attitude than men (Charness & Gneezy, 2012), may drive their more negative perception and lower propensity towards entrepreneurship. Thus, entrepreneurial risk attitude is included among the control variables. Most importantly, the main results reported above are robust to control for entrepreneurial risk attitude. In fact, RiskAtt appears to be insignificant in all three models, while the main effects remain highly significant. I conclude that even when the pure gender effect is disentangled from the risk attitude component, the investigated effects on entrepreneurial propensity and perception are not driven by women’s greater risk aversion.

25 this comprehensiveness, gender seem to be the more suitable measurement. Please find the detailed result tables for this section in Appendix II, Table 7 and 8.

4.3 Moderation Results

The moderation analyses investigate if gender equality affects the relationship of gender and perception, as well as the relationship of gender and entrepreneurial propensity. Moderation effects indicate if the included interaction variable proved to be significantly related to the outcome while the predictor variable and the moderator variable are controlled (Baron & Kenny, 1986).

The results show no significant interaction effect of Gender and GEI on both outcomes. In other words, increasing gender equality does affect neither the negative relation of gender and entrepreneurial propensity, nor the negative relation of gender and perception towards entrepreneurship. Thus, providing no support to the moderation effect of gender equality, and rejecting the proposed hypotheses 2 and 3.

Table 5 Regression Results for Moderation Effect

Gender – Perception Gender – Entr. Propensity

B B B B SE Sig. Predictor Gender -0.088 0.003 0.000 -0.015 0.002 0.000 GEI 0.309 0.021 0.000 -0.047 0.011 0.000 Demographic Controls Age 0.010 0.001 0.000 -0.004 0.001 0.000 WorkStatus -0.009 0.001 0.000 -0.003 0.000 0.000 Income 0.023 0.002 0.000 0.000 0.001 0.798 EducLevel 0.036 0.002 0.000 0.008 0.001 0.000 Social Controls KnowEnt 0.139 0.003 0.000 0.044 0.02 0.000 Interaction Gender*GEI -0.022 0.039 0.532 -0.005 0.021 0.813 Constant 0.080 0.015 0.000 0.060 0.008 0.000 4.4 Additional Results

Direct Effect of Gender Equality

26 outcomes. Consequently, the effect of gender equality is examined as additional control variable of the mediation model. The coefficients of GEI are highly significant in all three models. While

GEI is negatively related to entrepreneurial propensity (path c: B=-1.325, path b+c’: B=-2.163),

it is positively related to perception towards entrepreneurship (path a: B=0.270).

Comparing the prior models to the models including GEI, the negative effect of Gender on both outcomes (Percep, EntrProp) changes slightly, being less negative when GEI is inserted. Reversely, the effect of Percep on EntrProp becomes more positive in the model including

GEI. The results table can be found under Appendix III, Table 9. Analysis on Subsamples

The effect of gender equality is further investigated by creating subsamples, which split the countries by their high or low Gender Equality Index. The median is chosen as appropriate threshold.

First of all, all main relationships held in the subsamples (p<0.001). Furthermore, considering the sample with low equality countries the result indicates a lower gender effect on entrepreneurial propensity (B=-0.430, p<0.001) compared to the high sample and the full sample. However, the gender effect on perception shows reverse results: being a woman seem to have a more negative effect on perception towards entrepreneurship in countries with lower gender equality (B=-0.101, p<0.000). The country effects indicated further proof as for all countries with low equality the negative effect on perception is significant. Moreover, the mediation effect of perception is most pronounced in the low equality sample (gender effect: B=-0.233, p<0.01, change of 20 bps).

Indeed, the results for the high equality sample show the reverse effects. While gender indicates the highest effect on entrepreneurial propensity (B=-0.671, p<0.001), the effect on perception appeared to be the lowest (B=-0.078, p<0.001) – both compared to the low equality sample and the full sample. Hence, also the mediation effect of perception is less pronounced. Within the high equality countries, no consistent country effects are recognized.

27

5. Discussion

5.1 Theoretical Implications

The implications of this study are divided into two emphases: the relation of women’s perception towards entrepreneurship and their propensity to entrepreneurship, and subsequently the role of gender equality on both.

Overall, positive perception towards entrepreneurship was found to be an important factor for individuals’ propensity to entrepreneurship. Indeed, a negative perception hinders both gender to pursue an entrepreneurial career but for women this is much more likely. The evidence became even more substantial by eliminating that the lower likelihood is driven by women’s greater risk aversion. Women perceive themselves and their environment less optimistic, resulting in lower propensity to entrepreneurship – independent from their lower entrepreneurial risk attitude. Due to only few significant country effects, I further conclude that general country conditions may have less influence than perceptional and social capital factors. This is in line with the findings by Langowitz & Minniti (2007). However, as this is only evident in the present sample, this implication is limited to developed countries and opportunity entrepreneurs.

Building on the evidence on perception, the role of gender equality for female entrepreneurship was taken into account. I aimed to investigate if the reasons for women’s lower perception are rooted in the missing equality of women and men. However, independent from the level of gender equality, women showed to be more likely than men to have a negative perception, as well as a lower propensity to entrepreneurship. It is difficult to say whether these results demonstrate the absence of any relationship altogether, or that the data or analysis are insufficient to establish its presence. One possible reason may be that gender and gender equality are two different constructs, one being individual-specific while the other is at the environmental level. As such, it is recommendable that scholars keep these two constructs separate.

28 Firstly, low equality leads to the expectance of higher barriers and challenges for women engaging in entrepreneurship. Thus, women’s perception towards entrepreneurship is likely to be negative.

Secondly, inequality seem to be accompanied by more pronounced gender stereotypes resulting in discouragement of women’s confidence in their entrepreneurial identity. Underlying societal expectations for gender and stereotypes associating entrepreneurship with masculine characteristics lead to several impacts on the perceptional factors, such as less positive opportunity evaluation and less entrepreneurial self-efficacy (Chen et al. 1998; Gupta et al. 2009; Gupta et al., 2013). Especially the consistent effect for all low equality countries proof the presence of gender-role stereotypes across different countries, which is in line with findings in countries with particularly different cultures (Gupta et al., 2005). Therefore, especially in countries with low equality, women have a lower perception towards succeeding in an entrepreneurial career.

Nevertheless, the actual entrepreneurial propensity was the least affected by gender in the low equality sample compared to the high equality and the full sample. This implies that other factors next to perception seem to have more influence on women’s propensity to entrepreneurship in a low equality environment. Inequality seems to trigger women to be more assertive and positively affect their motivation to pursuit equality by themselves through starting a business (Lerner, Brush, & Hisrich, 1997). Additionally, due to higher barriers in employment and the “glass-ceiling” as career inhibitor, women may expect to achieve career success not in wage jobs but through the independence of starting a business.

29 5.2 Policy Implications

Within the study, I tested whether governmental support and policies for new firms have any impact on women’s entrepreneurial propensity. The insignificant results suggest that a general support of entrepreneurship does not affect women. Instead, policies and governmental programs should explicitly aim at the encouragement of female entrepreneurship. However, taking into account the gender equality of a country may enable interventions that are more effective.

In countries with rather low equality, specifically improving women’s perception could further encourage them in entrepreneurial activities. As perception is individual-specific and shaped by cultural and societal factors, localized and personal programs are more favorable than, for instance, monetary incentives (Bird & Brush, 2002; Langowitz & Minniti, 2007).

Countries with high equality may anticipate a positive perception, but should focus on policies that increase the attractiveness of entrepreneurship for women compared to wage employment. Women could be further liberated from their stronger family responsibility, for example, through the improvement of childcare services, or through tax incentives for self-employed women with children. Building on the aforementioned evidence that women’s motivation behind running a business lies especially in achieving social or environmental value goals, policies that support specifically social entrepreneurship or green start-ups may foster higher rates of business start-ups founded by women.

5.3 Limitations & Future Research

30 family responsibilities of women change with the level of gender equality, and subsequently, vary in influence on women’s entrepreneurial propensity.

Lastly, as the study is based on cross-sectional data, only correlation could be assessed. For future research, causality needs to be tested (for instance, by assessing how increasing gender equality within certain countries changes women’s entrepreneurial behavior).

6. Conclusion

In this thesis, I aimed to investigate the factors influencing women’s entrepreneurial propensity, and consequently explain women’s lower participation in entrepreneurship.

In line with recent literature, a partial mediation effect of perception towards entrepreneurship on the relationship of women and their entrepreneurial propensity was confirmed. Rejecting the hypothesis of a moderation effect of gender equality on women’s perception as well as propensity initially implied that women are independent from the level of gender equality less likely than men to have a positive perception and propensity to entrepreneurship. However, more in-depth analysis in subsamples, divided by countries with high and low gender equality, shed light on the role of equality for female entrepreneurship.

These results allow several specific theoretical implications. In a low equality environment, women expect higher barriers to engage in entrepreneurship, and as inequality strengthens gender stereotypes, women evaluate business opportunities less positive as well as their own entrepreneurial self-efficacy. The result is a negative perception towards entrepreneurship. Despite this negative perception, women in low equality countries are compared to women in high equality countries more likely to be involved in entrepreneurship. Yet, women in countries with high equality show almost the same perception as men.

32

References

Acs, Z. J., & Szerb, L. (2007). Entrepreneurship, economic growth and public policy. Small

Business Economics, 28(2-3), 109-122.

Ahl, H. (2006). Why research on women entrepreneurs needs new directions. Entrepreneurship

Theory and Practice, 30, 595–621.

Arenius, P., & De Clercq, D. (2005). A network-based approach on opportunity recognition. Small Business Economics, 24(3), 249-265.

Arenius, P., & Minniti, M. (2005). Perceptual variables and nascent entrepreneurship. Small

Business Economics, 24(3), 233-247.

Astebro, T., Herz, H., Nanda, R., & Weber, R. A. (2014). Seeking the roots of entrepreneurship: Insights from behavioral economics. Journal of Economic Perspectives, 28(3), 49-70.

Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (2001). Self‐efficacy beliefs as shapers of children's aspirations and career trajectories. Child Development, 72(1), 187-206. Bandura, A., Freeman, W. H., & Lightsey, R. (1999). Self-efficacy: The exercise of control Barber, B. M., & Odean, T. (2001). Boys will be boys: Gender, overconfidence, and common stock investment. The Quarterly Journal of Economics, 116(1), 261-292.

Barbieri, D., Franklin, P., Janeckova, H., Karu, M., Lelleri, R., Lestón, I., R., Luminari, D., Madarova, Z., Maxwell, K., Mollard, B., Osila, L., Paats, M., Reingardė, J., Salanauskaitė (2017). Gender Equality Index 2017: Measuring gender equality in the European Union 2005-2015, https://eige.europa.eu. Retrieved at Dec 2018

BarNir, A., Watson, W. E., & Hutchins, H. M. (2011). Mediation and moderated mediation in the relationship among role models, self‐efficacy, entrepreneurial career intention, and gender.

Journal of Applied Social Psychology, 41(2), 270-297.

Baron, R. A. (2000). Psychological perspectives on entrepreneurship: Cognitive and social factors in entrepreneurs' success. Current Directions in Psychological Science, 9(1), 15-18. Baron, R. A., & Ensley, M. D. (2006). Opportunity recognition as the detection of meaningful patterns: Evidence from comparisons of novice and experienced entrepreneurs. Management

Science, 52(9), 1331-1344.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of

Personality and Social Psychology, 51(6), 1173.

Baron, R. A., Markman, G. D., & Hirsa, A. (2001). Perceptions of women and men as entrepreneurs: Evidence for differential effects of attributional augmenting. Journal of Applied

33 Baughn, C. C., Chua, B. L., & Neupert, K. E. (2006). The normative context for women's participation in entrepreneruship: A multicountry study. Entrepreneurship Theory and

Practice, 30(5), 687-708.

Baum, J. R., & Locke, E. A. (2004). The relationship of entrepreneurial traits, skill, and motivation to subsequent venture growth. Journal of Applied Psychology, 89(4), 587

Beck, T., & Demirguc-Kunt, A. (2006). Small and medium-size enterprises: Access to finance as a growth constraint. Journal of Banking & Finance, 30(11), 2931-2943.

Beck, T., Demirgüç‐Kunt, A. S. L. I., & Maksimovic, V. (2005). Financial and legal constraints to growth: does firm size matter?. The Journal of Finance, 60(1), 137-177.

Blanchflower, D. G. (2004). Self-employment: More may not be better (No. w10286). National Bureau of Economic Research.

Blanchflower, D. G., & Oswald, A. J. (1998). What makes an entrepreneur?. Journal of Labor

Economics, 16(1), 26-60.

Boden Jr, R. J., & Nucci, A. R. (2000). On the survival prospects of men's and women's new business ventures. Journal of Business Venturing, 15(4), 347-362.

Bosma, N., Hessels, J., Schutjens, V., Van Praag, M., & Verheul, I. (2012). Entrepreneurship and role models. Journal of Economic Psychology, 33(2), 410-424.

Brush, C. G. (1992). Research on women business owners: Past trends, a new perspective and future directions. Entrepreneurship Theory and Practice, 16(4), 5-30.

Burke, A. E., Fitzroy, F. R., & Nolan, M. A. (2002). Self-employment wealth and job creation: The roles of gender, non-pecuniary motivation and entrepreneurial ability. Small Business

Economics, 19(3), 255-270.

Carter, N. M., Williams, M., & Reynolds, P. D. (1997). Discontinuance among new firms in retail: The influence of initial resources, strategy, and gender. Journal of Business Venturing, 12(2), 125-145.

Carter, S. & Marlow, S. (2003). Accounting for change: Professionalism as a challenge to gender disadvantage in entrepreneurship. In J. Butler (Ed.), New Perspectives on Women

Entrepreneurs (pp. 181–202). Greenwich, CT: Information Age Publishing

Carter, S. & Rosa, Peter, R. (1998). The financing of male- and female-owned businesses.

Entrepreneurship & Regional Development, 10(3), 225-242.

Charness, G., & Gneezy, U. (2012). Strong evidence for gender differences in risk taking.

Journal of Economic Behavior & Organization, 83(1), 50-58.

34 Chiles, T. H., Bluedorn, A. C., & Gupta, V. K. (2007). Beyond creative destruction and entrepreneurial discovery: A radical Austrian approach to entrepreneurship. Organization

Studies, 28(4), 467-493.

Chlosta, S., Patzelt, H., Klein, S. B., & Dormann, C. (2012). Parental role models and the decision to become self-employed: The moderating effect of personality. Small Business

Economics, 38(1), 121-138.

Chowdhury, S., & Endres, M. (2005). Gender difference and the formation of entrepreneurial self-efficacy. In United States Association of Small Business (USASBE) Annual Conference,

Indian Wells, CA.

Cliff, J. E. (1998). Does one size fit all? Exploring the relationship between attitudes towards growth, gender, and business size. Journal of Business Venturing, 13(6), 523-542.

Colombo, M. G., & Grilli, L. (2005). Founders’ human capital and the growth of new technology-based firms: A competence-based view. Research Policy, 34(6), 795-816.

Cowling, M., & Taylor, M. (2001). Entrepreneurial women and men: two different species?. Small Business Economics, 16(3), 167-175.

Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic

Literature, 47(2), 448-74.

Davidsson, P., & Honig, B. (2003). The role of social and human capital among nascent entrepreneurs. Journal of Business Venturing, 18(3), 301-331.

Del Junco, J. G., & Brás-dos-Santos, J. M. (2009). How different are the entrepreneurs in the European Union internal market? An exploratory cross-cultural analysis of German, Italian and Spanish entrepreneurs. Journal of International Entrepreneurship, 7(2), 135-162.

Delmar, F., & Davidsson, P. (2000). Where do they come from? Prevalence and characteristics of nascent entrepreneurs. Entrepreneurship and Regional Development, 12(1), 1–23.

DeTienne, D. R., & Chandler, G. N. (2007). The role of gender in opportunity identification. Entrepreneurship Theory and Practice, 31(3), 365-386.

Dyer, W. G., Jr. (1994). Toward a theory of entrepreneurial careers. Entrepreneurship Theory

and Practice, 18(3), 7–21.

Eagly, A. H., & Carli, L. L. (2003). The female leadership advantage: An evaluation of the evidence. The Leadership Quarterly, 14(6), 807-834.

Eckhardt, J. T., & Shane, S. A. (2003). Opportunities and entrepreneurship. Journal of

Management, 29(3), 333-349.