The interplay between personality characteristics, gender, and
self-employment.
Master Thesis
Jan Jaap Hiemstra S3543145
24-06-2019
Supervisor: Dr. M. Wyrwich Co-assessor: Dr. S Murtinu
2
Abstract
Using a large, representative German household panel, this study examines the influence of personality traits on the likelihood to be self-employed. In addition, interaction analyses are performed to capture potential differential effects of personality traits on likelihood of self-employment across men and women. Furthermore, this study will examine whether the gender effect on self-employment changes, once the model takes into account personality variables as well. The results show that the entrepreneurial personality profile as well as the single traits openness, conscientiousness, and
3
1. Introduction
In the past decades the topics of gender equality and therewith the percentual imbalance between sexes in certain industries and positions (i.e. the gender gap) have received a lot of attention. Many studies have measured the gender gap in different areas and attempted to uncover the factors that cause it to exist. Several studies have found that entrepreneurship is more common among men as compared to women (Blanchflower, 2004; Bowen & Hisrich, 1986; Kelley et al., 2011; Wilson et al., 2007). If we do not understand the underlying mechanisms that cause this gender gap, society cannot benefit from the value that can be created by half of its populace (Kelley et al., 2011). Moreover, women can offer a different entrepreneurial perspective that might increase the social and societal benefits resulting from the entrepreneurial field.
In order to explain variations in the likelihood of becoming an entrepreneur among men and women, past research mainly focused on gender-role stereotypes (Gupta et al., 2009) and family responsibilities (Kelley et al., 2011; Wellington, 2006). Collins et al. (1964) argue that the way of the entrepreneur is a long, lonely and difficult road, and that the people who follow it are by necessity a special breed. This indicates that besides a person’s gender role and family responsibilities, his personality also plays an important role in the likelihood to become an entrepreneur. Rauch & Frese (2007) confirm the relevance of personality when investigating entrepreneurial career choice and behavior, with clear support; which has yielded a renewed interest in personality differences as important determinant of entrepreneurial behavior.
4
Some scholars utilize the entrepreneurship-prone personality profile when measuring the outcomes of certain personality characteristics. The use of the entrepreneurship-prone personality profile fits Schumpeter’s (1934) view who states that the entrepreneur is a special type characterized by a specific set of different features that together drive entrepreneurial behavior. This study will explore the predictive strength of both the individual Big 5 personality traits and the entrepreneurship-prone profile on self-employment.
Schmitt-Rodermund (2004) found that high levels of agreeableness negatively influence a person’s likelihood to become an entrepreneur. However, not all industries require the same level of competitiveness. Charboneau (1981), among others, showed that female-controlled start-ups are more likely to be found in the retail and service sectors. These sectors often seem to exhibit lower levels of competitiveness and instead require a more relationship-based approach in order to be successful. One could argue that in these industries, the trait agreeableness might have less effect on the likelihood to become an entrepreneur than in it would have in more competitive industries. Therefore, the current study will measure the differential effects of agreeableness on likelihood of self-employment across men and women.
This study will examine several issues. First, the effects of the Big 5 personality traits and the entrepreneurship-prone profile on self-employment are measured and compared. Second, the
5
2. Theoretical background
2.1 Big 5 Personality traits
Prior research on the role of personality in entrepreneurship has mainly investigated the mediating role of risk propensity and self-efficacy (Wilson et al., 2009; Koellinger et al., 2013; Verheul et al., 2012). However, this study will use the five-factor model of personality since it is the
more-established and cross-culturally validated model of personality (Benet-Martinez & John, 1998). The idea of describing an individual’s personality started with researchers writing down all words in the dictionary that described a person’s personality (Goldberg, 1990). Years later the thousands of words that
described personality were cut back to sixty adjectives most commonly used. Subsequently, factor analysis pioneer Thurstone (1934) found through multiple factor methods that all sixty words could be captured in only five factors. In the following years multiple other researchers experimented with the personality traits, ultimately leading to the Big 5 personality traits as we know them today: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism.
Rauch & Frese (2007) state that entrepreneurial behavior is, to some extent, an expression of personality. Consequently, each of the five personality traits could positively or negatively influence an individual’s likelihood to be self-employed. Existing studies that focused on the five personality traits found that entrepreneurship is related to higher levels of extraversion, conscientiousness, and openness (Zhao & Seibert, 2006; Zhao et al, 2010). However, other studies show deviations from this pattern. Envick and Langford (2000), for example, compared the personality traits of entrepreneurs with those of managers (since both groups need to direct employees and manage multiple tasks) and found that entrepreneurs were significantly less conscientious and less extraverted than managers. In order to test whether openness, conscientiousness, and extraversion truly influence self-employment positively, the following three hypotheses are proposed:
H1a: A high level of Openness increases the likelihood that an individual is self-employed.
H1b: A high level of Conscientiousness increases the likelihood that an individual is self-employed.
6
Besides the three personality traits having positive influence on self-employment, existing studies also found that entrepreneurship is related to lower levels of agreeableness and neuroticism. (Zhao & Seibert, 2006; Zhao et al, 2010). Hence, the next two hypotheses are:
H1d: A high level of Agreeableness decreases the likelihood that an individual is self-employed.
H1e: A high level of Neuroticism decreases the likelihood that an individual is self-employed.
2.2 Entrepreneurship-prone profile
Besides the variable-oriented personality approach (taking the Big 5 traits individually) discussed in the previous paragraph, the other approach to examining connections between personality and other variables is the person-oriented approach (Magnusson & Törestad, 1993). Furr (2008) argues that an individual is a complex system and thus studying single isolated variables will most likely not fully capture the complex psychological reality of the individual person. Furthermore, Joseph Schumpeter (1934) stressed that the entrepreneur is a special type characterized by a specific set of different
features that together drive entrepreneurial behavior. The possession of these different complementary features is necessary for an entrepreneur since starting a business is demanding, complex, and risky; so an entrepreneur requires the different features simultaneously to solve all the necessary tasks needed in order to be successful.
Several studies (e.g. Obschonka et al., 2014) that hypothesized different types of relationships between the Big 5 personality traits and self-employment, found that besides individual personality traits, the entrepreneurial personality profile is also a significant predictor of self-employment. An entrepreneurship-prone personality profile is characterized by high levels of openness,
conscientiousness, and extraversion, while showing low levels of agreeableness and neuroticism (Schmitt-Rodermund, 2004). According to a study of Obschonka et al. (2013) in the US, Germany, and the UK, the entrepreneurship-prone personality profile delivered remarkably consistent results, indicating its validity in the prediction of entrepreneurial behavior. The current study will test whether the entrepreneurship-prone personality profile has a positive effect on self-employment. Hereafter, the extent to which the variance in self-employment is explained (R2) by both the variable-oriented and the
person-oriented approach is explored.
7 2.3 Gender and personality
In recent years the gender gap between the income of men and women has been a popular topic of discussion. The same gender gap seems to exist in entrepreneurship, and there is a lively debate on how to explain this almost universal phenomenon (Blanchflower, 2004; Bowen & Hisrich, 1986; Kelley et al., 2011; Wilson et al., 2007). There exist different theories on exactly which factors are the cause of this gender gap. However, it is widely acknowledged that the gender gap in entrepreneurship is a complex and multi-causal phenomenon (Kelley et al., 2011; Wilson et al., 2007).
Past research has explained the gender gap in entrepreneurship by focusing on, among others, education (Cowling & Taylor, 2001; Leoni & Falk, 2010), gender-role stereotypes (Gupta et al., 2009), and family responsibilities (Kelley, 2011; Wellington, 2006). The current study focuses on the difference in personality characteristics between men and women as driver of the gender gap in entrepreneurship. The personality difference that exists between the two genders is well documented (Costa et al., 2001; Feingold, 1994; Schmitt et al., 2008). Kourilsky & Walstad (1998), for example, found that girls show less interest in an entrepreneurial career than boys. This could be explained by the fact that women in general are more risk averse than men (Byrnes et al., 1999), combined with Rauch & Frese (2007) showing that entrepreneurs are less risk averse than non-entrepreneurs. Moreover, other studies suggest that another reason less women engage in entrepreneurship is because women have lower confidence in their own capabilities (Chen et al., 1998; Wilson et al., 2009; Wilson et al., 2007). According to biological and evolutionary approaches, gender differences result from men and women’s dimorphically evolved concerns concerning reproductive issues, and parental investment in offspring (Trivers, 1972; Buss, 2008). These theories argue that women should be more concerned with raising children which requires them to be more cautious, agreeable, nurturing, and emotionally involved. Men, however, are more concerned with obtaining mating opportunities which requires them to be more assertive, risk-taking, and aggressive. Furthermore, other theories argue that gender norms are shaped by socio-cultural influences resulting in that men and women are expected to serve different roles in society and are therefore socialized to behave differently from one another (Wood and Eagly, 2002; Eagly and Wood, 2005). Even though this study does not focus on the underlying mechanisms that have resulted in personality differences between genders, the literature mentioned does show support for the stand that men and women differ in their personality characteristics. At the Big 5 personality traits level, Schmitt et al. (2008) also found that women and men differ in their personality
8
genders could contribute to the gender gap in self-employment. In order to examine the role of personality differences between genders, this paper will explore what happens to the gender effect on self-employment once the model takes into account personality variables as well. The expectation is that the gender effect on self-employment will become smaller once personality variables are introduced.
Women and men differ with respect to start-up motivations (Hisrich & Brush, 1986), which can lead women to choose different industries for a start-up than men usually would. Many studies (e.g. Charboneau, 1981) have shown that female-controlled start-ups are more likely to be found in the retail and service sectors. Wojahn (1986) describes these sectors as the ‘helping’ professions. While some female founded businesses can be found in industries like agriculture, forestry, construction and manufacturing (Hisrich and O’Brien, 1981), these areas are considered non-traditional for women to start-up a business in. The fact that women generally choose more service-orientated industries for their start-up, could have to do with women’s higher awareness of other people’s feelings and their concern with other people’s welfare (Eagly & Kite, 1987). Moreover, women’s conception of self is more relational than that of men (Cross & Madson, 1997); and so they often seem to choose industries that require them to use their social skills and willingness to help others, in order to achieve business success. Men however, have the tendency to be more assertive and controlling; leading to a greater independence of others (Eagly & Kite, 1987), which is often aimed at setting and achieving their individualistic goals (Venkatesh, Morris, & Ackerman, 2000). These differences in personality between men and women seem to influence their choice of industry in such a way that men generally prefer more competitive industries while women prefer relationship-based industries. This fits the view of Chamlee-Wright (1997), who found that women generally tend to form long-term relationships based on effective ties, while men forge short-term relationships based on mutual interest. Additionally, women aim to form relatively egalitarian coalitions, while men forge relatively hierarchical coalitions (Chamlee-Wright, 1997).
Both competitive industries and relationship-based industries require different skills and a different mentality to be successful in. The Big 5 personality dimension agreeableness consists of traits like empathy and kindness, as well as the tendency toward cooperation, social harmony, and
9
show that female entrepreneurs are more often active in relationship-based industries in which competitiveness, dominance, and hierarchy seem to be of less importance than in competitive
industries. This could mean that the trait agreeableness is of less importance for entrepreneurs that are active in the industries generally preferred by females, while being of greater importance in the more competitive industries. In order to discover whether the proposed theoretical arguments are confirmed by the panel data, the following hypothesis was developed:
H3: Agreeableness has a stronger effect on self-employment among men.
3. Methodology
3.1 Representative household panel data
For the analysis in this study, the German Socio-Economic Panel (SOEP) is used. This longitudinal panel dataset is a household based study which started in 1984 and which re-interviews adult household members annually. The SOEP dataset collects data on numerous relevant factors, including gender, race, age, social class, education, profession, and the Big 5 personality traits. The dataset contains all the data needed for testing the proposed hypotheses on more than 60,000 respondents. Data is drawn from survey years 2005, 2009, and 2013. While the data from year 2013 is more up-to-date, data from 2005 and 2009 is also needed since those year’s surveys included questions on the Big 5 traits.
10
(3.14%) self-employed without co-workers, 1,574 (2.62%) self-employed with co-workers, resulting in a total of 3,466 (5.76%) respondents being self-employed.
3.2 Measurement of single traits and personality constructs
Several waves of the SOEP data included established psychological personality inventories. This provides the opportunity to explicitly study the consequences of personality characteristics on a large sample of the population. In particular, data from 2005 and 2009 is crucial for this study, since these surveys contained personality questionnaires that measured respondent’s Big 5 personality traits. In order to establish the Big 5 personality traits, respondents were asked how much they agreed with different statements about their character on a 7-point Likert scale. Three items per personality trait led to a total of fifteen items that assessed the respondent’s Big 5 personality scores. Next, the scores of the items measuring one construct are averaged, resulting in that particular trait’s final score. The scale was inverted for the items conscientiousness, extraversion, agreeableness, and neuroticism. According to the theory, openness, conscientiousness, and extraversion are beneficial to the entrepreneurial personality profile; while agreeableness and neuroticism are detrimental to it. Accordingly, the respondents’ entrepreneurial personality profiles are calculated based on their single traits scores, resulting in entrepreneurial personality fit scores ranging from -156 (lowest fit) to -5 (highest fit).
3.3 Regression analysis and correlations
The method of analysis in this study is binary logistic regression where the dependent variable is a dummy and indicates whether a respondent is self-employed. The independent variables are
11
accounts for differences between married (1) and unmarried (0) respondents.
The empirical analysis will start with an examination of the observed correlations between sex, the Big 5 personality traits, and the entrepreneurial personality fit. As can be expected, the
entrepreneurial personality fit is heavily correlated with the Big 5 single traits. This is caused by the fact that the entrepreneurial personality fit is calculated based on respondent´s scores on the single traits. The fact that the entrepreneurial fit shows positive correlations with openness, conscientiousness, and extraversion, while showing negative correlations with agreeableness and neuroticism; suggests that the entrepreneurial fit calculation is functioning as it should, according to the literature
(Schmitt-Rodermund, 2004).
Furthermore, the Big 5 traits are all significantly correlated with one another in a weak to moderate manner (< 0,3); except for openness and extraversion, which show a correlation coefficient of 0,373. An explanation for this could be that in the Big 5 questionnaire the six questions covering
openness and extraversion show similarities in the way that they focus on the extent to which the respondent is confident socially and how he deals with new situations. Both traits require confidence and the ability to take initiative, which is what could have caused the high correlation coefficient. The correlation table (Appendix I) suggests that the personality variables used in this study are correlated with one another but still clearly distinct. This means all variables can be included in the empirical analyses. However, to be able to interpret the resulting output, the coefficients of the single traits are compared to the coefficients of the full model to check how multicollinearity influences the test
12
4. Empirical results
4.1 Big 5 traits and self-employment
We start with reporting the outcomes for the effects of the single Big 5 traits on
self-employment. The trait effects on self-employment are presented by means of the log-odds, the odds ratios, and the standard errors of the different traits (Table 1). Moreover, table 1 provides the number of observations, the Wald Chi square, and the pseudo R squared belonging to the different logistic regression models. Self-employment is represented by three groups: all self-employed, self-employed without co-workers, and self-employed with coworkers. The Beta coefficient (log-odds) represents the change in the logit of the outcome variable (i.e. self-employment) associated with a one-unit change in the predictor value. The logit of the outcome is the natural logarithm of the odds that a respondent is self-employed. For clarity of interpretation, this study will interpret the odds ratios associated with the log-odds of the independent variables. An odds ratio value greater than 1 indicates that as the predictor variable increases, the odds of the outcome occurring increase. If the odds ratio value is lower than 1, this indicates that as the predictor variable increases, the odds of the outcome occurring decrease. Statistically insignificant coefficients (p>0.1) will not be reported.
13
Table 1 Big 5 traits on self-employment (binary logistic regressions)
All SE SE without co-workers SE with co-workers
ß Odds ratio ß Odds ratio ß Odds ratio
Openness 0.243*** 1.276 0.349*** 1.418 0.095*** 1.100 (0.017) (0.023) (0.025) Conscientiousness 0.165*** 1.179 0.057** 1.058 0.288*** 1.334 (0.022) (0.029) (0.034) Extraversion 0.086*** 1.089 0.068*** 1.070 0.098*** 1.103 (0.018) (0.024) (0.026) Agreeableness -0.192*** 0.825 -0.123*** 0.884 -0.256*** 0.774 (0.019) (0.026) (0.027) Neuroticism -0.141*** 0.869 -0.097*** 0.907 -0.177*** 0.837 (0.015) (0.020) (0.022) Survey year 2005 0.048 1.049 -0.064 0.938 0.174*** 1.190 (0.038) (0.051) (0.054) Age -0.009*** 0.991 -0.007*** 0.993 -0.011*** 0.990 (0.001) (0.002) (0.002) Education 0.148*** 1.160 0.119*** 1.126 0.172*** 1.188 (0.006) (0.007) (0.008) Manufacturing industry -0.357*** 0.700 -0.405*** 0.667 -0.273*** 0.761 (0.062) (0.087) (0.087) Residence in 1989 0.171*** 1.186 0.142*** 1.152 0.201*** 1.223 (0.027) (0.034) (0.041) Married 0.261*** 1.299 0.095* 1.099 0.455*** 1.576 (0.041) (0.053) (0.062) Number of observations 3420 1866 1554 Wald X2 1794.545*** 890.415*** 996.052*** Pseudo R2 0.084 0.061 0.077 ***p<0.01, **p<0.05, *p<0.1
The individual effects of the single traits on self-employment have also been tested and are presented in Appendix II. As can be expected, the coefficients of the isolated traits on self-employment are higher than the coefficients in the complete model since the full model has more information to base its calculations on. This means that some of the variation in the independent variable that was accounted to the individual single traits in the isolated models, should actually be attributed to other dependent variables, according to the complete model. The differing traits coefficients between the individual models and the complete model do not come unexpected, since the personality traits are all correlated to one another (Appendix I). Nevertheless, this study will base its conclusions on the
14 4.2 Entrepreneurial personality profile
Table 2 presents the outcomes of the logistic regressions performed to examine the effect of the entrepreneurial personality profile on self-employment. A one-unit increase in the entrepreneurial personality profile is associated with a 2.7% increase in the odds of being self-employed. The odds of being self-employed without co-workers and being self-employed with coworkers increase with 2.5% and 2.8%, respectively. This seems quite low, however, keep in mind that this percentage is associated with a one-unit increase on a scale with a range of 152 units.
Table 2 Entrepreneurial personality profile on self-employment (binary logistic regressions)
All SE SE without co-workers SE with co-workers
ß Odds ratio ß Odds ratio ß Odds ratio
E'neurial personality fit 0.027*** 1.027 0.025*** 1.025 0.027*** 1.028
(0.001) (0.002) (0.002) Survey year 2005 0.061 1.063 -0.048 0.953 0.183*** 1.201 (0.038) (0.051) (0.054) Age -0.008*** 0.992 -0.006*** 0.994 -0.009*** 0.991 (0.001) (0.002) (0.002) Education 0.152*** 1.164 0.129*** 1.137 0.168*** 1.182 (0.006) (0.007) (0.008) Manufacturing industry -0.382*** 0.682 -0.472*** 0.624 -0.250*** 0.779 (0.062) (0.086) (0.086) Residence in 1989 0.178*** 1.194 0.150*** 1.162 0.203*** 1.225 (0.027) (0.034) (0.041) Married 0.253*** 1.288 0.063 1.065 0.476*** 1.610 (0.041) (0.053) (0.062) Number of observations 3420 1866 1554 Wald X2 1630.057*** 716.699*** 917.153*** Pseudo R2 0.076 0.049 0.071 ***p<0.01, **p<0.05, *p<0.1
4.3 Variable-oriented and person-oriented approach
15
the extent to which the model explains the variance in self-employment. Table 3 shows the Pseudo R squared values per self-employed group. For all self-employed, the entrepreneurial personality profile explains 2.7% of the variance in self-employment while the single traits approach explains 3.7% of the variance. For self-employed without co-workers, the entrepreneurial personality profile explains 1.9% of the variance and the single traits approach explains 3.5% of the variance. For self-employed with co-workers, the entrepreneurial personality profile explains 2.2% of the variance while the single traits approach explains 2.8% of the variance in self-employment.
Table 3 Predictive power of variable-oriented and person-oriented approach
All SE SE without co-workers SE with co-workers
Pseudo R2 Pseudo R2 Pseudo R2
E'neurial personality fit 0.027 0.019 0.022
Single traits 0.037 0.035 0.028
4.4 Gender and personality
At the Big 5 traits level, men and women differ in their personality characteristics (Schmitt et al., 2008). The difference in personality between men and women might contribute to the gender effect on self-employment. Table 4 presents the exploration on what happens to the gender effect when
personality variables are introduced into the model. Model 1 shows the individual effect of gender on self-employment. In model 2, the Big 5 traits were introduced as well. In model 3, the entrepreneurial personality profile was introduced.
16
Table 4 Gender effect with personality variables
Model 1 Model 2 Model 3
ß Odds ratio ß Odds ratio ß Odds ratio
Sex 0.675*** 1.965 0.697*** 2.008 0.595*** 1.812 (0.037) (0.039) (0.038) Openness 0.251*** 1.285 (0.017) Conscientiousness 0.180*** 1.198 (0.022) Extraversion 0.122*** 1.130 (0.018) Agreeableness -0.135*** 0.874 (0.020) Neuroticism -0.074*** 0.929 (0.016)
E'neurial personality fit 0.025*** 1.025
(0.001) Survey year 2005 0.035 1.036 0.025 1.025 0.054 1.056 (0.038) (0.038) (0.038) Age -0.012*** 0.988 -0.011*** 0.989 -0.010*** 0.990 (0.001) (0.001) (0.001) Education 0.159*** 1.172 0.145*** 1.157 0.149*** 1.160 (0.005) (0.006) (0.006) Manufacturing industry -0.554*** 0.575 -0.529*** 0.589 -0.549*** 0.578 (0.063) (0.063) (0.063) Residence in 1989 0.194*** 1.214 0.183*** 1.201 0.187*** 1.205 (0.026) (0.027) (0.027) Married 0.248*** 1.282 0.254*** 1.289 0.248*** 1.281 (0.041) (0.042) (0.041) Number of observations 3466 3420 3420 Wald X2 1591.501*** 2116.826*** 1885.860*** Pseudo R2 0.073 0.099 0.088
***p<0.01, **p<0.05, *p<0.1. The dependent variable in these models is all self-employed.
17
decreased the odds ratio (table 2) of the entrepreneurial personality profile with 0.2 percentage points. This can be explained by the moderate correlation between the two variables.
4.5 Interaction effects
The test results of the moderator effect of sex on the relationship between agreeableness and self-employment are presented in table 5. Agreeableness shows a negative effect on self-employment for all self-employed and a positive effect on self-employment without co-workers, though for both models the effect is not significant. If agreeableness is increased by one unit, the odds of being self-employed with co-workers decrease significantly with 10.7%. The odds for a man being self-self-employed are 2 times higher for all self-employed, 1.8 times higher for self-employed without co-workers, and 1.853 times higher for self-employed with co-workers; than the odds for a women being self-employed. Unfortunately, the interaction coefficient (agreeableness*sex) is not significant. Therefore, we can conclude that agreeableness has the same effect on self-employment for both men and women.
Table 5 Agreeableness by sex on self-employment (binary logistic regressions)
All SE SE without co-workers SE with co-workers
ß Odds ratio ß Odds ratio ß Odds ratio
18
Since the hypothesized interaction term agreeableness*sex did not yield satisfactory result, the remaining 4 traits and the entrepreneurial personality profile have also been interacted with sex to check for significant effects. When interacting sex with conscientiousness, neuroticism, and the entrepreneurial profile; either the constitutive term or the interaction coefficient (or both) were not significant. Because existing statistics literature clearly states that we may not exclude constitutive terms from the model (Brambor et al., 2006), it is no option to remove the insignificant constitutive terms in order to find significant interaction effects. Therefore, we conclude that conscientiousness, neuroticism, and the entrepreneurial personality profile have the same effect on self-employment for men and women.
When openness and extraversion were interacted with sex, both constitutive terms and the interaction coefficient were found to be significant (table 6). According to the interaction model, a one-unit change in openness increases the odds of being self-employed with 42.8%. However, the
19
Table 6 Alternative interaction effects on self-employment (binary logistic regressions)
Openness Extraversion
ß Odds ratio ß Odds ratio
Openness 0.356*** 1.428 (0.026) Extraversion 0.291*** 1.338 (0.027) Openness*Sex -0.087*** 0.917 (0.033) Extraversion*Sex -0.078** 0.925 (0.034) Sex 1.158*** 3.184 1.131*** 3.098 (0.167) (0.180) Survey year 2005 0.024 1.025 0.021 1.021 (0.038) (0.038) Age -0.011*** 0.989 -0.010*** 0.990 (0.001) (0.001) Education 0.142*** 1.152 0.158*** 1.172 (0.005) (0.005) Manufacturing industry -0.514*** 0.598 -0.531*** 0.588 (0.063) (0.063) Residence in 1989 0.183*** 1.201 0.196*** 1.216 (0.026) (0.027) Married 0.284*** 1.328 0.251*** 1.286 (0.041) (0.041) Number of observations 3445 3461 Wald X2 1950.195*** 1810.466*** Pseudo R2 0.090 0.083
20
5. Discussion and conclusion
This study addresses the question of how personality traits and gender affect the likelihood of a person to be self-employed. The dataset used for this is the German Socio-Economic Panel (SOEP). The current study aims to verify three hypotheses and explore two areas of interest.
With respect to the Big 5 personality traits, the results are in line with the hypotheses. Looking at all self-employed; openness, conscientiousness, and extraversion have a positive significant effect on self-employment. Agreeableness and neuroticism have a negative significant effect on self-employment. Openness has a positive effect on both self-employed with workers and self-employed without co-workers. This is not surprising since before one can start a new business, he has to be creative enough to develop a good idea. However, the effect is significantly greater for self-employed without co-workers. An explanation for this is that even though openness to experience is crucial in order to start any business, being capable of leading a business with co-workers requires many other traits and skills besides openness. This idea is supported by the fact that conscientiousness, extraversion,
agreeableness, and neuroticism are all stronger predictors for being self-employed with co-workers than they are for the likelihood to be self-employed without workers. An entrepreneur that manages co-workers needs to have the right social capabilities in order to lead his company to success. Part of these social capabilities is being outgoing and not afraid to take initiative in social situations, which is reflected by extraversion. Similarly, when being at the head of a company, one has to be capable of protecting his interests and not be afraid of confrontations with subordinates. This is reflected by the stronger
negative effect of agreeableness on self-employed with co-workers, compared to self-employed without co-workers. Where openness was the strongest predictor of the 5 traits for being self-employed without co-workers, conscientiousness is the strongest predictor for being self-employed with co-workers. Conscientious people are organized and take their obligations to others seriously. Hence, it is not striking that being conscientious is more important for an entrepreneur who depends (partly) on his co-workers for success, than for an entrepreneur who operates solo. Starting a business is an uncertain and difficult project which requires the founder to handle stress well and not get nervous easily. This is why high levels of neuroticism have a negative effect on the likelihood of being self-employed.
21
entrepreneurial personality profile reflects a person’s potential capacity to be an entrepreneur based on that person’s single traits. The effects of the entrepreneurial personality profile are slightly higher for self-employed with co-workers, compared to self-employed without co-workers. This again indicates that to be self-employed with co-workers, the possession of the full range of entrepreneurial traits (and corresponding skills) is more important than for self-employed without co-workers. It seems that it is quite crucial for an entrepreneur (especially with co-workers) to be a “Jack-of-all-trades” (Lazear,2004). Furthermore, education has a stronger effect on the likelihood of self-employment with co-workers than on the likelihood of self-employment without co-workers. This indicates that to be self-employed with co-workers, investing enough time in developing the different relevant skills is more important than for self-employed without co-workers.
The first exploration of this study is aimed at examining and comparing the predictive strengths of the entrepreneurial personality profile and the single traits approach on self-employment. In order to measure this, Nagelkerke’s R squared was interpreted. For both the self-employed without co-workers and the self-employed with co-workers, the single traits approach was a stronger predictor than the entrepreneurial personality profile. Furthermore, the difference between the predictive strengths of the two approaches is significantly greater for self-employed without co-workers, than for self-employed with co-workers. This is likely to be caused by the fact that out of the 5 traits, self-employment without co-workers is predicted to a large extent by openness (table 1). After having converted the single 5 traits (unweighted) into the entrepreneurial personality profile, the strong individual effect of openness reduces drastically, leading to a reduction in predictive strength. In the case of the self-employed with co-workers, the difference between predictive strengths of the two approaches is much smaller. Concerning self-employment with co-workers, the 5 traits individually contribute a more equal portion to the predictive strength. And so, after converting the 5 traits into the entrepreneurial personality profile, less predictive strength is lost as compared to the case of self-employed without co-workers.
22
the 5 traits into the model is not supported. Instead, the opposite occurred, and the gender effect increased. The fact that the gender effect increased slightly, while the effects of agreeableness and neuroticism decreased considerably, indicates that gender captures some of the effects on
self-employment that were previously attributed to agreeableness and neuroticism. This is supported by the relatively high correlations that gender has with agreeableness and neuroticism. The inclusion of the entrepreneurial personality profile into the model with gender did have the expected results. The gender effect on self-employment decreased significantly. A portion of the effect on self-employment previously attributed to gender, is apparently captured by the entrepreneurial personality profile. This does not come unexpected, since men and women on average score differently on the Big 5 traits (Schmitt et al. 2008).
The analysis results on the interaction effect of agreeableness by sex on self-employment is not in line with the hypothesis. This study found no differential gender effect on the relationship between agreeableness and self-employment. However, the trait openness has been found to be a more important predictor of self-employment for women than for men. Furthermore, extraversion is also found to be a stronger predictor of self-employment for women, as compared to men. An explanation for this could be that women, on average, have a higher awareness of other people’s feelings and well-being (Eagly & Kite, 1987), which often leads them to choose service-orientated industries for their start-ups. In these industries, social skills and high awareness of other people’s needs are crucial in order to achieve business success. In contrast, men tend to prefer more technical industries, in which they are judged mainly on their output, leading to a lower importance of social skills. Hence, the greater
importance of the trait extraversion for women to be self-employed as compared to men, might result from the difference in choice of industries among men and women.
The first implication of this study is that after having uncovered the effects of the
entrepreneurial profile and the Big 5 traits on self-employment, entrepreneurship training programs can provide a clearer picture on which traits enhance and which traits are detrimental to the likelihood of becoming an entrepreneur. The second implication is that the gender effect on self-employment did not become smaller after controlling for personality variables. This indicates that there exist other
23
References
-Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic groups: Multitrait-multimethod analyses of the Big Five in Spanish and English. Journal of personality and social psychology, 75(3), 729.
-Blanchflower, D. G. (2004). Self-employment: More may not be better (No. w10286). National Bureau of Economic Research.
-Bowen, D. D., & Hisrich, R. D. (1986). The female entrepreneur: A career development perspective. Academy of
management review, 11(2), 393-407.
-Brambor, T., Clark, W.R., & Golder, M. (2006). Understanding interaction models: Improving empirical analyses.
Political analysis, 14(1), 63-82.
-Buss, D.M. (2008). Evolutionary Psychology: The New Science of the Mind, 3rd Edn. Boston: Allyn & Bacon
-Byrnes, J. P., Miller, D. C., & Schafer, W. D. (1999). Gender differences in risk taking: a meta-analysis. Psychological
bulletin, 125(3), 367.
-Caliendo, M., Fossen, F., & Kritikos, A.S. (2014). Personality characteristics and the decisions to become and stay self-employed. Small Business Economics, 42(4), 787-814.
-Chamlee-Wright, E. (1997). The Cultural Foundations of Economic Development. Routledge: London and New York. -Charboneau, F.J. (1981). The woman entrepreneur. American Demographics, 3(6), 21-23.
-Chen, C. C., Greene, P. G., & Crick, A. (1998). Does entrepreneurial self-efficacy distinguish entrepreneurs from managers?. Journal of business venturing, 13(4), 295-316.
-Collins, O., Moore, D., & Unwalla, D. (1964). The Enterprising Man. East Lancing, Michigan State University. -Costa Jr, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality traits across cultures: robust and surprising findings. Journal of personality and social psychology, 81(2), 322.
-Cowling, M., & Taylor, M. (2001). Entrepreneurial women and men: two different species?. Small Business
Economics, 16(3), 167-175.
-Cross, S.E., & Madson, L. (1997). Models of the self: self-construals and gender. Psychological bulletin, 122(1), 5. -Eagly, A.H., & Kite, M.E. (1987). Are stereotypes of nationalities applied to both women and men?. Journal of
personality and social psychology, 53(3), 451.
-Eagly, A.H., & Wood, W. (2005). Universal sex differences across patriarchal cultures ≠ evolved psychological dispositions. Behavioral and Brain Sciences, 28, 281-28310.
-Envick, B.R., & Langford, M. (2000). The five-factor model of personality: Assessing entrepreneurs and managers.
Academy of Entrepreneurship Journal, 6(1), 6-17.
-Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological bulletin, 116(3), 429. -Furr, R.M. (2008). A framework for profile similarity: Integrating similarity, normativeness, and distinctiveness.
Journal of Personality, 76, 1267-1316.
-Goldberg, L. R. (1990). An alternative" description of personality": the big-five factor structure. Journal of
personality and social psychology, 59(6), 1216.
-Gupta, V. K., Turban, D. B., Wasti, S. A., & Sikdar, A. (2009). The role of gender stereotypes in perceptions of entrepreneurs and intentions to become an entrepreneur. Entrepreneurship theory and practice, 33(2), 397-417. -Hisrich, R.D., & Brush, C. (1986). The Women Entrepreneurs: Starting, Financing and Managing New Business. Lexington, USA: Lexington Books.
-Hisrich, R.D., & O’Brien, M. (1981). The woman entrepreneur from a business and sociological perspective.
Frontiers of entrepreneurship research, 21(11), 19-23.
-Kelley, D., Bosma, N. S., & Amorós, J. E. (2011). Global entrepreneurship monitor 2010 executive report.
-Koellinger, P., Minniti, M., & Schade, C. (2013). Gender differences in entrepreneurial propensity. Oxford bulletin
24
-Kourilsky, M. L., & Walstad, W. B. (1998). Entrepreneurship and female youth: Knowledge, attitudes, gender differences, and educational practices. Journal of Business venturing, 13(1), 77-88.
- Lazear, E.P. (2004). Balanced skills and entrepreneurship. American Economic Review, 94(2), 208-211. -Leoni, T., & Falk, M. (2010). Gender and field of study as determinants of self-employment. Small Business
Economics, 34(2), 167-185.
-Magnusson, D., & Törestad, B. (1993). A holistic view of personality: A model revisited. Annual Review of
Psychology, 44, 427-452.
-Nagelkerke, N.J. (1991). A note on a general definition of the coefficient of determination. Biometrika, 78(3), 691-692.
-Obschonka, M., Schmitt-Rodermund, E., & Terracciano, A. (2014). Personality and the gender gap in self-employment: A multi-nation study. PloS one, 9(8), e103805.
-Obschonka, M., Schmitt-Rodermund, E., Silbereisen, R.K., Gosling, S.D., & Potter, J. (2013) The regional
distribution and correlates of an entrepreneurship-prone personality profile in the U.S., Germany, and the UK: A socioecological perspective. Journal of Personality and Social Psychology, 105, 104-122.
-Portegijs, W., Boelens, A., & Keuzenkamp, S. (2002). Emancipatiemonitor 2002.
-Rauch, A., & Frese, M. (2007). Let's put the person back into entrepreneurship research: A meta-analysis on the relationship between business owners' personality traits, business creation, and success. European Journal of work
and organizational psychology, 16(4), 353-385.
-Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can't a man be more like a woman? Sex differences in Big Five personality traits across 55 cultures. Journal of personality and social psychology, 94(1), 168.
-Schmitt-Rodermund, E. (2004) Pathways to successful entrepreneurship: Parenting, personality, entrepreneurial competence, and interests. Journal of Vocational Behaviour 65, 498-518.
-Schumpeter, J.A. (1934). The theory of economic development. Cambridge, MA: Harvard University Press -Thurstone, L. L. (1934). The vectors of mind. Psychological review, 41(1), 1.
-Trivers, R.L. (1972). “Parental investment and sexual selection,” in Sexual Selection and the Descent of Man, ed. Campbell B., editor. (Chicago: Aldine; ), 136-179.
-Venkatesh, V., Morris, M.G., & Ackerman, P.L. (2000). A longitudinal field investigation of gender differences in individual technology adoption decision-making processes. Organizational behavior and human decision processes,
83(1), 33-60.
-Verheul, I., Thurik, R., Grilo, I., & Van Der Zwan, P. (2012). Explaining preferences and actual involvement in self-employment: Gender and the entrepreneurial personality. Journal of Economic Psychology, 33(2), 325-341. -Weinkopf, C. (2014) Women's Employment in Germany, Robust in Crisis but Vulnerable in Job Quality. Revue de
l'OFCE, 133, 189-214.
-Weisberg, Y.J., DeYoung, C.G., & Hirsh, J.B. (2011). Gender differences in personality across the ten aspects of the Big Five. Frontiers in psychology, 2, 178.
-Wellington, A. J. (2006). Self-employment: the new solution for balancing family and career?. Labour
Economics, 13(3), 357-386.
-Williams, J.E., & Best, D.L. (1990). Sex and psyche: Gender and self-viewed cross-culturally. Sage Publications, Inc. -Wilson, F., Kickul, J., & Marlino, D. (2007). Gender, Entrepreneurial Self‐Efficacy, and Entrepreneurial Career Intentions: Implications for Entrepreneurship Education 1. Entrepreneurship theory and practice, 31(3), 387-406. -Wilson, F., Kickul, J., Marlino, D., Barbosa, S.D., & Griffiths, M.D. (2009). An analysis of the role of gender and self-efficacy in developing female entrepreneurial interest and behavior. Journal of developmental
entrepreneurship, 14(02), 105-119.
25
-Wyrwich, M. (2013). Can socioeconomic heritage produce a lost generation with regard to entrepreneurship?
Journal of Business Venturing, 28(5), 667-682.
-Zhao, H., & Seibert, S. E. (2006). The Big Five personality dimensions and entrepreneurial status: A meta-analytical review. Journal of applied psychology, 91(2), 259.
26
Appendix
I Correlation matrix [1] [2] [3] [4] [5] [6] [7] [1] Openness 1 [2] Conscientiousness 0.151*** 1 [3] Extraversion 0.373*** 0.175*** 1 [4] Agreeableness 0.122*** 0.299*** 0.082*** 1 [5] Neuroticism -0.063*** -0.089*** -0.161*** -0.104*** 1[6] E'neurial personality fit 0.535*** 0.172*** 0.559*** -0.386*** -0.517*** 1
[7] Sex -0.063*** -0.072*** -0.094*** -0.171*** -0.208*** 0.127*** 1
***p<0.01, **p<0.05, *p<0.1
II Effect of separate traits on self-employment (binary logistic regressions)
All SE SE without co-workers SE with coworkers
ß Odds ratio ß Odds ratio ß Odds ratio
Openness 0.277*** 1.319 0.367*** 1.444 0.146*** 1.157 0.016 0.021 0.023 Conscientiousness 0.208*** 1.232 0.132*** 1.141 0.291*** 1.337 0.021 0.027 0.032 Extraversion 0.207*** 1.230 0.211*** 1.235 0.185*** 1.203 0.016 0.022 0.024 Agreeableness -0.080*** 0.923 -0.028 0.972 -0.136*** 0.873 0.018 0.025 0.027 Neuroticism -0.155*** 0.857 -0.112*** 0.894 -0.193*** 0.825 0.015 0.020 0.022
