Education and training as non-psychological characteristics that influence university students' entrepreneurial behaviour.

Barahona, Juan Hernangomez ; Cruz, Natalia Martin ; Escudero, Ana Isabel Rodriguez 等

ABSTRACT

It is well known that entrepreneurship plays a key role in boosting innovation and well-being in an economy, but researchers have shed little light on how universities can motivate this propensity. In this current paper, our aim is to examine which complementary education and work experience measures prior to graduation universities can encourage in order to foster entrepreneurship. For this purpose, we analyse a sample of 2012 Spanish university students in the final years of their degrees in the academic year 2003-2004. Using a binary logistic regression analysis, we obtain results that generally confirm our expectations. That is, complementary educational activities and work experience prior to graduation can help to explain the propensity to create business start-ups.

THEORETICAL APPROACH AND RESEARCH OBJECTIVES

In recent years, public institutions such as education centres and other European institutions have become aware of the importance of developing an efficient business structure, which is capable of discovering and exploiting opportunities in an increasingly dynamic, complex and uncertain environment (Ronstandt, 1985; Scott et al., 1998; Meyer, 2003; Snijders & Vander Horst, 2002).

The European Union (EU) is developing an active strategy of promoting entrepreneurship. Thus, the current strategy is reflected in policies such as the Multi-annual programme for Enterprise and Entrepreneurship for 2001-2005, Euro Info Centres and Innovation Relay Centres.

The emphasis on young people is particularly noteworthy, although hardly surprising, since this group contains the largest proportion of entrepreneurs in the EU (CEEDR, Middlesex University, 2000). By analysing the characteristics of this group of entrepreneurs in detail, we find that a large proportion of these individuals have taken courses up to degree level at university. In Spain, a higher proportion of entrepreneurs are young university graduates than the European average: 40.5% compared to 35% (Reynolds et al., 2002).

Given this positive outlook for entrepreneurship among young graduates, the responsible authorities should design useful educational policies (Henderson & Robertson, 1999). It is also clearly advisable to further our understanding of the elements that promote entrepreneurial inclinations. Thus, before teaching people how to start entrepreneurial activities, they must be induced with the desire to do so. Little is known about this aspect (Henderson & Robertson, 1999).

Following Shane's (2003) reasoning, the attributes or features of individuals that influence their ability to evaluate and subsequently exploit the opportunities that emerge can be classified into two groups: psychological and non-psychological factors. This research focuses on the non-psychological features of individuals that favour the exploitation of business opportunities. Among the main non-psychological factors we are interested in the demographic characteristics considered by the upper-echelon literature, such as age and gender, and the sociological variables, specifically those regarding education and work experience.

We consider that age has an inverted U-shaped relationship with entrepreneurship. On the one hand, it includes the positive effect derived from the individual's maturity, but on the other hand, there is the negative effect of the opportunity cost and uncertainty. The opportunity cost increases with age insofar as this latter variable is positively associated with income. However, uncertainty increases because life expectancy declines (Reynolds et al., 1994; Bates, 1995). Consequently, age has a positive effect on entrepreneurship for younger individuals, but after a certain age the relationship between age and entrepreneurship becomes negative.

Another feature, which is no less important but rarely studied, is the gender variable. Studied in the context of modern society, which is characterised by equal opportunities for men and women, it is of interest to researchers on entrepreneurship (Hisnichet et al., 1996; Duchenant & Orhan, 2000; Orhan and Scott, 2001).

Researchers have shown that women's motivation to engage in business activities is very varied and can originate from diverse factors, such as the influence of the environment and the necessity or desire for self-realisation (Orhan & Scott, 2001). According to Peters (2004), the role women play in the economy is one of the most significant phenomena of the early 21st century. In the US, the number of women running a company increased from 9 million in 1997 to around 10.6 million in 2004 (Center for Women's Business Research, National Numbers, 2004). In spite of these figures, it seems that nowadays men are still more prone to set up their own company and take on the risk involved in any new venture.

From the sociological perspective, another set of variables affecting the decision to set up a firm comprises the individual's education and previous work experience (Bates, 1990; Cressy, 1996). Education and experience in other business activities provide individuals with information (about the market, the workforce, etc.) and skills (in sales, planning, decision-making, etc.) that improve their ability to combine the resources, develop a strategy, organise business activities and, in short, exploit opportunities more successfully (Corman et al, 1988; Lee, 1999; Shane & Khurana, 2001).

Specifically, researchers have found that one of the main characteristics of individuals running small enterprises is a higher educational level (Smith, 1967; Robinson & Sexton, 1994; Lorrain & Dussault, 1998). In particular, among the determinant factors of entrepreneurship, university education has been found to be very influential. Henderson and Robertson (1999) observe that students who have participated in postgraduate education programmes are more likely to set up their own firm. Researchers have also found a positive relationship between education and the firm's growth aspirations (Davidson et al., 2003; Matos, 2003).

Education at all levels plays a key role in the development of an entrepreneurial society. However, it is necessary to develop education in the most fundamental dimensions associated with the definition of personality and the learning of leadership skills (Fillion, 2004). In this vein, universities play an important role in training entrepreneurs as individuals of multiple characteristics (Lazear, 2003; Pleitner, 2003; Fayolle et al, 2004). Entrepreneurs, unlike specialists, possess a more harmonious education in all the areas of knowledge required to successfully exploit opportunities and manage the attitudes and tasks of individuals (including the specialists) who will go on to help them successfully develop their new business (Lazear, 2003). Greater knowledge, together with a higher level of information and skills, provides the individual with a greater capacity to undertake entrepreneurial activities and to assume or demonstrate entrepreneurial attitudes.

New entrepreneurs, apart from their higher educational level, also stand out for their increased work experience (Lorrain & Dussault, 1998; Smith, 1967). The individual's experience is one of the variables that researchers have most frequently found to be significant in distinguishing between successful and unsuccessful entrepreneurs (Cooper et al., 1994; Gimeno et al., 1997; Burke et al., 2000 and 2002; Capelleras et al., 2004; Fayolle et al, 2004). Moreover, if the first experience has something to do with starting a business activity during the individual's youth, this makes the individual more likely to start another entrepreneurial activity later on (Ikei, 1997; Dobrev & Barnett, 1999; Wandosell & Garcia, 2004). To this we should add that entrepreneurs with some managerial experience acquired previously tend to create firms that grow faster compared to inexperienced individuals (Henderson & Robertson, 1999; Capelleras et al., 2004). Also, people with a more varied previous experience are more likely to start their own business (Littunen, 2002) because they have been able to develop networks of influence and contacts. In this respect, an important finding is that the majority of European entrepreneurs have related past positions as specialist workers or managers in other firms. Thus, all of them had considerable knowledge and some previous business experience (Reynolds et al., 2002).

The current research (see Figure 1) has the aim of studying the influence of the abovementioned variables (age, gender, education and work experience) as fundamental determinants of the propensity to set up firms by university students close to graduation. In other words, we apply the general approach of the literature on entrepreneurship to an analysis of entrepreneurial orientation among university students.

[FIGURE 1 OMITTED]

METHODOLOGY

The information that we used to test the empirical model proposed here was provided by the General Foundation (www.funge.uva.es) of the University of Valladolid (Spain). The Foundation has initiated a study called the "Professional Observatory of the University of Valladolid", whose aim is to conduct an exhaustive analysis of the current situation and problems of the academic degrees/studies and educational areas taught by the University of Valladolid with regards to their professional development, as well as to identify complementary education that would help graduates to adapt more closely to current labour-market needs.

In order to achieve this objective, various data collection tools have been developed that are directed at different students. One of these tools consists of a questionnaire directed at students from cycles 3-5 of the various degrees/studies (The Spanish university system offers students the possibility of choosing either 3-year or 5-year studies. Students successfully completing a 3-year degree ("first cycle") are awarded diplomas. These can then optionally continue for two more years ("second cycle") to complete a 5-year degree, whereupon they graduate). This is the source of information used in the current study. 2012 completed questionnaires were collected for the year 2004. The sample consists of 59.6% engineering students, 25.7% social sciences students and 14.6% humanities students.

The means of the age and gender variables show that more than half of the students surveyed are women, and that their average age is around 23 years. The group of education variables includes the acquisition of knowledge complementary to the university studies themselves, such as the level of foreign language ability, the number of stays abroad, the level of computer skills, and receiving other types of complementary non-university education (for instance, courses for developing job search skills, courses for developing teaching skills, sports training, musical training, among others). To evaluate the students' previous work experience, we considered firms' internships, voluntary/social work, work with a contractual work (in a position either related or unrelated to the student's speciality), work without a non-contractual work, and previous experience as a freelancer. The dependent variable of the study is a dichotomous variable measuring the students' orientation to create their own firm after graduating. 63% of students in the sample do not want to become entrepreneurs and 37% desire to set up a new company.

Table 1 provides more information about the variable measurements and their average values for the sample individuals. Before we tested the hypotheses, we examined the correlation matrix (see Table 2). The signs of the bivariate correlation appear to be consistent with the hypothesized relationships.

The analytical methodology used was binary logistic regression, in which the dependent variable was the students' propensity to set up their own firm. Logistic regression analysis is well suited when the dependent variable is non-metric and consists of just two groups. Compared to discriminant analysis, choosing logistic regression is justified by the fact that the multivariate normality assumptions do not need to be met. Logistic regression is much more robust when these assumptions are not met. But even if they are met, many researchers prefer this methodology to discriminant analysis, because the interpretation of the results is similar to that of regression analysis results.

Logistic regression also tests the hypothesis that a coefficient is different from zero as is done in multiple regression, where the t value is used to assess the significance of each coefficient. Although logistic regression uses a different statistic, the Wald statistic, it also provides the statistical significance for each estimated coefficient so that hypothesis testing can occur just as it did in multiple regression.

Specifically, we used a hierarchical logistic regression. This methodology allows us to sequentially introduce different blocks of variables and to check their respective explanatory capacities. Firstly, we included the block corresponding to the main effects of all the independent variables (Model 1). Finally, we add the variable age squared to these variables, to test for the existence of an inverted U-shaped relationship (Model 2). The relevance of the inverted U-shaped effect cannot be rejected if the corresponding Wald-statistic is significant.

We used three global goodness-of-fit indices. Firstly, we use the log-likelihood (-2LL), for which low values indicate a better model fit. This index is similar to the residual or error sums of square values for multiple regression. Secondly, we use the Hosmer and Lemeshow test, which measures the correspondence between the observed and expected results of the dependent variable. A non-significant value indicates a good fit. Finally, we apply the Nagelkerke R2, which is interpreted similarly to the R2 of any multiple regression model. In logistic regression, there is no true R2 value as there is in OLS regression. However, because deviance can be thought of as a measure of how poorly the model fits (i.e., lack of fit between observed and predicted values), an analogy can be made to the sum of squares residual in ordinary least squares. The proportion of unaccounted for variance that is reduced by adding variables to the model is the same as the proportion of accounted for variance, or R2. An index that reflects this basic idea has been developed by Nagelkerke.

RESULTS

Table 3 presents the results from the regression. The logistic regression conducted on the entire sample has suitable adjustment indexes. An important number of variables are significant in explaining the students' entrepreneurial orientation.

With regards to gender, our expectations are met (-0.36, p<0.01). Women are indeed less likely to start businesses than men. Age, however, does not have an effect on entrepreneurship, neither when we consider the positive and linear effect, nor when we consider the quadratic effect. When the 'age' variable is included the -2LL index barely changes and the Nagelkerke [R.sup.2] does not change. This may be because the sample individuals show little variability in this variable. 98% of the sample are under 30 years old.

With regards to complementary education, computer skills (0.39, p<0.01) and education (0.10, p<0.05) in other areas are both important. We do not find foreign language ability to be relevant, but stays abroad are (0.30, p<0.05). However, both variables--foreign language ability and number of stays abroad--are highly correlated (0.28, p<0.01). This result can indicate that the first condition is necessary but not enough to motivate entrepreneurial behaviour. Foreign language ability during stays abroad has a much more positive effect on entrepreneurship than when it is put into practice without travelling from home.

Previous work experience also has an important role in explaining the propensity to set up firms. Specifically, the uncertainty that is related to non-contractual work (0.50, p<0.01) seems to stimulate students to create their own professional environment. We might interpret internships (0.28, p<0.01) and volunteering (p<0.37, p<0.01) that way. However, contractual work experience is not associated with the desire to create firms. This may be because once students have experienced the absence of risk in contractual work they will adapt to this or look for a similarly comfortable situation, rather than contemplate starting their own firm.

Finally, among all these results perhaps the most noteworthy is the strong propensity to create firms among individuals that have already had experience as entrepreneurs/freelancers (0.82, p<0.01). The extent of the effect is much larger than that observed for the rest of the variables.

DISCUSSION AND CONCLUSIONS

In general, the factors recognised in the entrepreneurship literature as determinants of a stronger entrepreneurial orientation are highly significant in the present study. In other words, university students' propensity to create firms is explained by demographic, educational and work experience variables.

Women continue to show lower entrepreneurial behaviour compared to men, however nowadays the situation has improved. This conclusion is consistent with the results of previous international literature. In particular, in Spain this reality can be explained by our historical and cultural heritage. Our results have to be assessed by universities while discussing new and specific policies to promote entrepreneurial behaviour among members of this institution.

Education is one of the strongest engines of entrepreneurial orientation, especially, if it is associated with the opening of minds brought about by travelling and confronting the individual's knowledge with that of other countries. Initiatives such as exchange programmes are fundamental for producing enterprising students. At the same time, complementary education that is not part of their formal degree gives students the possibility of interrelating their specialised knowledge with that of other areas, which will be the seed of entrepreneurial ideas. Thus, these extra areas should complement formal education in order to foster entrepreneurship.

This is particularly the case with computer ability, since fast technological development means that this skill is essential in almost all business projects. Our findings show that it catalyses students' entrepreneurial initiatives.

On the other hand, work experience in firms helps to turn students into entrepreneurs. This is particularly true when the job is not too formal. Thus, rather than the flexibility of the labour market achieved through temporary work agencies, what helps the most is to design internships during which students face varied experiences in an organisation, in either a for-profit or non-profit context. Thus, universities should encourage university-firm collaboration, or the creation of groups of entrepreneurs that support and advise students about how to exploit their ideas and set up their own firms. We can suggest the following recommendations for university authorities wishing to foster entrepreneurship among their students:

1. Since students with computer skills are more likely to start firms than others, this type of education should be essential in all degrees. University teaching should incorporate this technology as part of normal teaching routines, so that all students are assumed to have such knowledge when they graduate, just as correct use of language and the ability to express oneself well are assumed nowadays.

2. Students who have previously created their own business or participated in a start-up are more likely to repeat the experience. This result helps us to claim that Universities should promote courses in which students are faced with the creation of a new venture. This can be done both by trying to create real new businesses or by participating in simulation exercises.

3. Another possible way of generating entrepreneurs would be to make it easier and quicker for students to set up their own firms in practice, by establishing collaborations between the university and the public institutions responsible for fostering entrepreneurship. These collaborative agreements can lead to the simplification of the bureaucratic procedures prior to the start of business activities.

4. Although the situation has improved, women still show fewer propensities to create firms than men. This situation should be evaluated carefully by universities, which should then establish policies for promoting entrepreneurship specifically for potential female entrepreneurs.

LIMITATIONS AND FUTURE RESEARCH

This paper has some limitations. However, the identification and correction of these limitations could give rise to new streams of future research. Firstly, the low value for the Nagelkerke R2 indicates a limited explanation of the variance by variables included in this research. We think that other individual-related variables, like the psychological profile of students--risk-aversion, autonomy or self-confidence- could improve the explanation of entrepreneurial behaviour. Risk-aversion is negatively related to the exploitation of new business opportunities. Taking risks is an essential part of entrepreneurial activity (Van Praag & Cramer, 2001; Stewart & Roth, 2001). Students' independent behaviour could have a positive relationship with the need to create their own business. We claim that independent individuals have difficulties in accepting rigid organisations and procedures and are averse to hierarchical structures. Thus, they will be willing to make their own decisions and to initiate their own ventures, independently of other peoples' beliefs (Koh, 1996; Douglas, 1999). Finally, self-confidence is positively related to entrepreneurial behaviour. This psychological feature means that the individual will trust in his own skills and abilities to control the environment in which his business will grow (Davidsson, 2001; Matos, 2003).

These psychological variables could be used in an enlarged model to explain the entrepreneurial propensity to create new ventures. Other than that, it could be interesting to evaluate the potential interactions and synergies among these individual-related characteristics and the non-psychological ones. By identifying these effects, we could focus the institutional efforts more efficiently to increase the number of new businesses created by university graduates.

Another limitation of our research is related to the cross-sectional data. We cannot observe with our data if students with a high entrepreneurial propensity will indeed become new entrepreneurs. Thus, we suggest longitudinal research in which we follow the students once they finish university. This information will allow us to find out if students with an entrepreneurial inclination finally became entrepreneurs and then, we could identify those individual-related variables that are more persistent and determinant of entrepreneurial propensity.

REFERENCES

Bates, T. (1990). Self-employment trends among Mexican Americans. Economic Studies, CES, 90-9.

Bates, T. (1995). Self-employment entry across industry groups. Journal of Business Venturing, 10(2), 143-56.

Burke, A.E., F.R. Fitzroy & M.A. Nolan (2000). When less is more: distinguishing between entrepreneurial choice and performance. Oxford Bulletin of Economics and Statistics, 62(5), 565-587.

Burke, A.E., F.R. Fitzroy & M.A. Nolan (2002). Self-employment wealth and job creation: the roles of gender non-pecuniary motivation and entrepreneurial ability. Small Business Economics, 19, 255-270.

Capelleras, J.L., E. Genesca & J.M. Veciana (2004). Determinants of Start-up and subsequent growth: A multivariate analysis. Proceedings of the XIV Congreso ACEDE, Murcia.

Casson, M., (2003). The Entrepreneurship: An Economic Theory (2nd Edition). Ed. Edward Elgar, USA.

CEEDR (2000). Young Entrepreneurs, Women Entrepreneurs, Co-Entrepreneurs and Ethnic Minority Entrepreneurs in the European Union and Central and Eastern Europe. Final report to the European Commission, DG Enterprise. Middlesex University Business School, UK.

Center for Women's Business Research (2004). National Numbers.

Cooper, A. C., J. Gimeno & C. Y. Woo (1994). Initial human and financial capital as predictors of new venture performance. Journal of Business Venturing, 9, 371-395.

Corman, J., B. Perles & P. Yancini (1988). Motivational factors influencing high-technology entrepreneurship. Journal of Small Business Management, 10(1), 7-23.

Cressy, R. (1996). Are Business Startups Debt-Rationed?. The Economic Journal, 106, September, 1253-1270.

Davidsson P. & J. Wiklund (2001). Levels of analysis in entrepreneurship research: Current research practices and suggestions for the future. Entrepreneurship Theory and Practice, 25(Summer), 81-99.

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

Dobrev, S. & W. Barnett (1999). Organizational roles and transitions to entrepreneurship. Research paper n. 1587, Graduate School of Business, Stanford University.

Douglas, E. (1999). Entrepreneurship as a career choice: Attitudes, entrepreneurial intentions, and utility maximization. In P. Reynolds, W. Bygrave, S. Manigart, C. Mason, G. Meyer, H. Sapienza y K. Shaver (Editors), Frontiers of Entrepreneurship Research (PP. 152-66), Babson Park, US: Babson College.

Duchenant, B. & M. Orhan (2000). Les femmes entrepreneurs en France : Percee des femmes dans un monde construit au masculin, Seli Arslan, Paris.

Fayolle, A., A. Vernier & B. Djiane (2004). Les jeunes diplomes de l'enseignement superieur sont-ils des createurs d'entreprise comme les autres?. Gestion 2000, 22(5), 39-56.

Fillion, L. J, (2004). Um Roteiro para Desenvolver o Empreendedorismo. Congresso Internacional de Empreendedorismo, Recife, Brasil.

Henderson, R. & M. Robertson (1999). Who wants to be an entrepreneur? Young adult attitudes to entrepreneurship as a career. Education & Training, 41(4/5),236-245.

Hisrich, R.D., C. Brush, D. Good & G. De Souza (1996). Some preliminary findings on performance in entrepreneurial ventures: Does gender matter?. Frontiers of Entrepreneurship Research, Babson College, Wellesley, MA.

IKEI (1997). Dinamica empresarial en Navarra: 1990-1995, Gobierno de Navarra, Espana.

Koh, H. C. (1996). Testing hypotheses of entrepreneur Characteristics. Journal of Managerial Psychology, 11, 12-25.

Lazear, E. P. (2003). Entrepreneurship. Discussion Paper Series, IZA DP No 760.

Lee, A. (1999). Empirical studies of self-employment. Journal of Economic Surveys, 13(4), 381-416.

Littunen, H. (2002). "Entrepreneurship and the characteristics of the entrepreneurial personality". International Journal of Entrepreneurial Behaviour & Research, 6(6), 295-302.

Lorrain, J. & L. Dussault (1988). Les entrepreneurs artisans et opportunistes: une comparaison de leurs comportements de gestion. Revue Internationale P.M.E., 1(2), 157-169.

Matos, J.J., (2003). Estudo do Crescimento e Desempenho das Pequenas Empresas: A Influencia da Orientacao Estrategica Empreendedora. PHD Dissertation, University of Beira Interior, Portugal.

Meyer, M. (2003). Academia entrepreneurs or entrepreneurial academics? Research-based ventures and public support mechanisms. R & D Management, 33(2), 107-115.

Minniti, M. & W. Bygrave (1999). The microfoundations of entrepreneurship. Entrepreneurship Theory and Practice, 23(4), 41-52.

Orhan, M. & D. Scott (2001). Why women enter into entrepreneurship: An explanatory model. Women in Management Review, 16(5/6), 232-243.

Peters, T. (2004). As mulheres e o novo mercado. HSM Management, 3, 6-14.

Pleitner, H. J. (2003). "Entrepreneurship-Fashion or driving force?". In Creacion de Empresas--Entrepreneurship (PP. 45-48), Servei de Publicacions, University Autonoma de Barcelona.

Reynolds, P., D. Storey & P. Westhead (1994). Cross-national variations in new firm formation rates. Regional Studies, 28, 443-456.

Reynolds, P., W. Bygrave, E. Autio, L. Cox & M. Hay (2002). Global entrepreneurship monitor. 2002 Executive report, Babson Collage.

Robinson, P. & E. Sexton (1994). The effect of education and experience on self-employment success. Journal of Business Venturing, 9(2), 141-156.

Ronstadt, R. (1985). The educated entrepreneurs: A new era of entrepreneurial education is beginning. American Journal of Small Business, 10(1), 7-23.

Scott, M, P. Rosa & H. Klandt (1998). Introduction: Educating entrepreneurs for wealth creation, Aldershot: Ashgate.

Shane, S., (2003). A general theory of entrepreneurship-The individual-opportunity nexus, Edward Elgar Editores.

Shane, S. & R. Khurana (2001). Career experiences and firm foundings. Paper presented at the Academy of Management Meetings.

Shane, S. & S. Venkataraman (2000). The promise of entrepreneurship as a field of research. Academy of Management Review, 26(1), 13-17.

Smith, N.R. (1967). The entrepreneur and his firm. E. Lansing, MI: Michigan State University Bureau of Business and Economic Research.

Snijders, J. & R. Van der Horst (2002). Las PYMEs europeas en estudio. Principales resultados del observatorio de las PYMEs europeas 2000, Observatorio de las PYMEs europeas 2002, Belgium.

Stewart, W. & P. Roth (2001). Risk taking propensity differences between entrepreneurs and managers: A meta-analytic review. Journal of Applied Psychology, 86(1), 145-53.

Van Praag, C. & J. Cramer (2001), "The roots of entrepreneurship an labour demand: Individual ability and low risk aversion. Economica, 68(269), 45-62.

Venkataraman, S. (1997). The distinctive domain of entrepreneurship research: An editor's perspective. In J. Katz & R. Brockhaus (editors), Advances in Entrepreneurship, Firm Emergence, and Growth, 3, Greenwich, CT, US: JAI Press, 119-38.

Wandosell, G. & A. Garcia (2004). Motivaciones y obstaculos en la creacion de empresas, efectos de la experiencia empresarial. Boletin Economico del ICE, 2819, 11-17.

Juan Hernangomez Barahona, University of Valladolid Natalia Martin Cruz, University of Valladolid Ana Isabel Rodriguez Escudero, University of Valladolid Table 1: Variable measurement and descriptive analysis Variable Measure Mean (std. dev.) Propensity to set up firms 0/1(No/Yes) 0.37 (0.48) Age Years 23.41 (2.93) Gender 0/1 (Male/Female) 0.56 (0.49) EDUCATION Foreign language skills Numeric 2.74 (1.72) Number of stays abroad Numeric 1.15 (3.57) Computer skills level Numeric 1.24 (0.48) Complementary non- Numeric 1.12 (1.04) university education EXPERIENCE Internship experience 0/1(No/Yes) 0.21 (0.41) Experience as a volunteer 0/1(No/Yes) 0.14 (0.35) Contractual experience 0/1(No/Yes) 0.30 (0.46) (job not related to studies) Contractual experience 0/1(No/Yes) 0.10 (0.30) (job related to studies) Non-contractual experience 0/1(No/Yes) 0.32 (0.47) Experience as an entrepre- 0/1(No/Yes) 0.025 (0.16) neur/freelancer Table 2: Pearson correlation matrix -1 -2 (3) (4) (1) Propensity to set up firms (2) Age .06 *** (3) Gender -.10 *** -.07 *** (4) Foreign language skills .06 ** -.04 * .05 ** (5) Number of stays abroad .08 *** .05 ** 0 .28 *** (6) Computer skills level .11 *** 0.03 -.29 *** .09 *** (7) Comple- mentary non- university education .10 *** .16 *** .07 ** .23 *** (8) Internship experience .07 *** .11 *** .04 * .07 *** (9) Experience as a volunteer .07 *** -.03 .11 *** .05 ** (10) Contract- ual experience (job not related to studies) .07 *** .14 *** -.09 *** .04 * (11) Contract- ual experience (job related to studies) .07 *** .19 *** .00 .05 * (12) Non- contractual experience .13 *** -.00 -.00 .04 * (13) Exper- ience as an entre-preneur/ freelancer .09 *** .11 *** -.08 *** 0.03 (5) (6) (7) (8) (1) Propensity to set up firms (2) Age (3) Gender (4) Foreign language skills (5) Number of stays abroad (6) Computer skills level -.01 (7) Comple- mentary non- university education .11 *** .09 *** (8) Internship experience .02 -.03 .17 *** (9) Experience as a volunteer .04 * -.02 .17 *** .01 (10) Contract- ual experience (job not related to studies) .06 *** -.03 .13 *** .07 *** (11) Contract- ual experience (job related to studies) .10 *** -.04 .16 *** .12 *** (12) Non- contractual experience .04 * -.00 .11 *** .00 (13) Exper- ience as an entre-preneur/ freelancer 0 .05 *** 0.03 0 (9) (10) (11) (12) (1) Propensity to set up firms (2) Age (3) Gender (4) Foreign language skills (5) Number of stays abroad (6) Computer skills level (7) Comple- mentary non- university education (8) Internship experience (9) Experience as a volunteer (10) Contract- ual experience (job not related to studies) .08 *** (11) Contract- ual experience (job related to studies) .09 *** .09 *** (12) Non- contractual experience .10 *** .09 *** .02 (13) Exper- ience as an entre-preneur/ freelancer 0 .06 ** 0 0 Significance levels: *** p<.001; ** p <.05; p <.10. Table 3: Results of logistic regression on entrepreneurial orientation Variable Beta Wald Constant -1.83 ** 16.8 Age 0.02 1.12 Gender -0.36 ** 10.92 EDUCATION Foreign language skill 0.02 0.39 Number of stays abroad 0.30 ** 4.62 Computer skills level 0.39 *** 12.29 Complementary 0.10 ** 4.11 non-university education EXPERIENCE Internship experience 0.28 ** 5.34 Experience as a volunteer 0.37 ** 6.63 Contractual experience 0.12 1.09 (job not related to studies) Contractual experience 0.17 1.11 (job related to studies) Non-contractual experience 0.50 *** 22.92 Experience as an 0.82 *** 7.37 entrepreneur/freelancer -2LL value 342 MODEL 1 Hosmer and Lemeshow 4.38 (0.82) Nagelkerke [R.sup.2] 0.08 No. observations 2012 [Age.sup.2] -0.002 1.43 -2LL value 2341 MODEL 2 Hosmer and Lemeshow 6.54 (0.59) Nagelkerke [R.sup.2] 0.08 No. observations 2012 Significance levels: *** p<0.01, ** p<0.05