The influence of entrepreneurial education on entrepreneurial intention

(1)

The influence of

entrepreneurial education

on entrepreneurial

intention

Entrepreneurship and education are two such extraordinary opportunities that need to be leveraged and interconnected if we are to develop the human capital required for building the societies of the future.

(World Economic Forum, 2009)

J. Pijnacker S1611739 Date 05-09-2012

(2)

First Supervisor: Dr. C.H.M. Lutz Second Supervisor: MSc. F. Noseleit

Abstract

Much research is conducted in entrepreneurial intention and the effect of entrepreneurial education. This study has a random assigned treatment group and three control groups with different characteristics. A more extensive theory of planned behavior is used; the entrepreneurial education program to test the entrepreneurial intention. The results shows entrepreneurial education does not have the desirable effect. However, the perceived creativity is increased and the over-optimism of students is tempered.

Keywords: Entrepreneurial intention, entrepreneurial education, sorting function, the

(3)

Acknowledge

I would like to thank some people who were involved in the process of writing my thesis: - Dr. C.H.M. Lutz and Msc. F. Noseleit for their flexibility and more important for the,

constantly provided, valuable feedback.

- Dr. Nassaucollege and my colleagues for the opportunity to do research at the school and the provided support.

- W. Moossdorff, K. Visser and T. van Rijs for their critics and the feedback/ review of the English writing style.

- My friends and girlfriend for their feedback and support during the summer vacation. - And at last but not lease my family for the support and encouragement during my study

period. Could not enjoy as much as I did without them.

I hope you will enjoy reading my thesis,

(4)

List of figures

FIGURE 1THEORY OF PLANNED BEHAVIOR (AJZEN,1991) ... 2-14 FIGURE 2ENTREPRENEURSHIP EDUCATION PROGRAM ASSESSMENT MODEL,FAYOLLE ET AL.(2006) ... 2-17 FIGURE 3RESEARCH MODEL ... 3-32 FIGURE 4TIME FRAME DATA PROCEDURE ... 3-36

List of Tables

(7)

1 Introduction

There is a growing interest among policymakers in educating entrepreneurship as a way to increase the entrepreneurial activity in a country. In the past decade policymakers and economists see entrepreneurship as an important factor for economic growth (Fayolle, Gailly and Lassas-Clerc, 2006). The first who came up with empirical evidence for the importance of entrepreneurship for economic growth was Schumpeter (1934). In general, recent thinking stresses the importance of innovation and creative destruction in a Schumpeterian spirit (Audretsch, 2007; Baumol, 2002). In addition, Audretsch and Thurik (1998) found an increasing demand for entrepreneurship, because of the globalization and the ICT-revolution.

The Dutch government has invested millions in EE programs since 2000. A Dutch newspaper published an article (L. Benschop, 2012) titled: ‘The Dutch government reserved two million for

entrepreneurial students in 2012’. This is remarkable, since the Netherlands is currently in a crisis

and the government has to cut back on many areas; like education, health care etc. It is likely to think the government seeing entrepreneurial education as catalyzer for the future and a solution for the ongoing crisis. In the year 2000, the Dutch government formulated two goals (Department of Education, 2000):

 An increasing number of educational institutions in the Netherlands have integrated entrepreneurship in their policy, their organization and their curriculum.

 A growing number of pupils and students show more entrepreneurial behavior, for example by starting their own business within a period of five years after completing their education.

(8)

positively according to a report of the EU (2006); ‘when only suitable persons choose the career

direction it will lower the cost of unsuccessful start-ups’.

In contradiction to the result of Oosterbeek et al. (2010), Peterman and Kennedy (2003) found a positive increase in perceived feasibility and desirability as a result of EE. Students desire a career choice as entrepreneur and they think they have the right characteristics to become a successful entrepreneur. In addition, Souitaris (2007) sees EE as a trigger event and this explains the increase in aspiration due to the event. At last, Martin et al., (2012) performed a meta-analysis and according to the six leading studies in the field, EE can influence intention and especially self-efficacy.

In the meta-analysis it appeared most studies which reported a positive impact had significant methodological deficiencies. Implying the validity of the results is limited. Some studies did ex-post measurements (Souitaris, 2007), others do not have a control group (Lee et al., 2005; Graevenitz, 2010), have small samples (Fayolle, 2006) or did not have random assignment selecting the treatment and control groups. When EE is not obligatory, participating students already have a higher intention to become an entrepreneur when they participate in the course (self-selection bias). In addition, most studies measured the impact on universities, while the impact is bigger for secondary students (Zhao et al., 2005). As a consequence, the existing literature leaves a research gap for a randomized chosen treatment group at a secondary school. Therefore, most studies recommend future studies with a more robust research design. Peterman and Kennedy (2003) stated: ‘there has been little rigorous research on its effects’. In the meta-analyses of Martin et al. (2012) they propose studies follow a pre- and post-test design and have a treatment and a control group. Moreover, they suggest the groups are randomly chosen. Considering the current state of research, in this research there will be a treatment and three control groups with different characteristics. The treatment and one control group is randomly assigned. A pre-post-test design is used to measure the impact of EE. The research question studied is: ‘Is entrepreneurial education effective in influencing entrepreneurial intention?’.

(9)

(10)

2 Literature review

In order to get an understanding of entrepreneurial education and associated concepts, a review of the literature is performed. First, literature of the concept entrepreneurial education (EE) is reviewed, followed by a description of the different intention models and the different entrepreneurial education programs. Finally, this review will give an overview of the existing studies concerning effects on intention due to entrepreneurial education.

2.1 Entrepreneurial education

Entrepreneurial education is a hot topic in the literature. However, the question asked by Fiet (2000): ‘Can entrepreneurship be taught..’, cannot be answered without giving an answer on what must be taught?

Entrepreneurship has not an unambiguous definition in the literature and consequently does not have a common language (Shane and Venkataram, 2000). In order to develop a clear definition Wiklund and Davidsson (2011), suggest to approach the research field of entrepreneurship not as a theory-based research field, but as a phenomena-based research field:

‘We propose that the phenomenon of “emergence of new economic activity” lies at the heart of entrepreneurship...’. Therefore, the 'emergence of new economic activity' will be used as the

definition for entrepreneurship in this research. This definition of entrepreneurship suits the description of EE in the literature review of Garavan and O’Cinnéide (1994).They concluded EE is widespread and has different ways of teaching. The goal is to catalyze new economic activity. Hence, this definition covers a wide variety of contents, methods, teaching approaches and aims. In the next paragraphs this will be explored in greater depth.

EE is frequently studied and many definitions were used. Jamieson (1984) derived three categories:

i. Education about enterprise; this category deals with awareness creation and the aim is to educate students on various aspects concerning entrepreneurship. It will have a more theoretical perspective and which seek ‘to foster skills, attitudes and values appropriate to

(11)

ii. Education for enterprise; this category focuses, compared to the previous category, more on training and to aspire people for self-employment. It is more practical and it encourages entrepreneurs to start their own business. An example of a subject in the program is to learn how to make up a business startup scheme.

iii. Education in enterprise; the third category focuses on management training and is for established entrepreneurs. Although, this definition refers to owners of small businesses, it is irrespective to the type of organization.

These categories are refined in the article of Garavan and O’Cinnéide (1994). They took a broader view and distinguish on the one hand entrepreneurship education and on the other hand, education and training for small business ‘owners’. The former is focused on knowing what entrepreneurship is. Education and training is focused on the skills needed to act as an entrepreneur. Garavan and O’Cinnéide distinguish three types of education and training for small business ‘owners’: (i) entrepreneurial education; (ii) continuing small business education; and (iii) small business awareness education. They focused on the life stage of EE participants. Since, this research is focusing on a program in the first class of secondary school this program falls into the category ‘small business awareness education’ or as Jameson (1984) formulated ‘education about enterprise’. This category is described as education process focusing on secondary schools and introduces. It makes students sensitive for and aware of entrepreneurship.

(12)

entrepreneurial knowledge, but also on the ‘know why’ and ‘know who’ (Haase and Lautenschläger, 2010).

A second objective mentioned by Hytti and O’Gorman (2004) is that entrepreneurial education can be seen as a way to inspire and prepare students for the emergence of activity. The whole set of entrepreneurial activities, educational and training, has the aim to influence intentions. Students need a growing perceived desirability and feasibility to become an entrepreneur (Linan, 2004). This is in line with the advocate of Sanchez (2010). Sanchez concluded the aim of entrepreneurial education must be to positively catalyze entrepreneurial intention.

Based on the previous, this research sees EE as an activity with the aim to educate entrepreneurial knowledge or/and to make people aware of the possibility to become entrepreneur. However, EE does not have always the desirability effect. In the literature there are different insights on how to measure the effect of EE. This will be discussed in the next paragraph.

2.2 Entrepreneurial education (EE) and intention

EE can have different impacts. In the literature different models and research designs are used to measure the effects. One option to measure the effectiveness is to capture the actual start-up activities among attendants. The problem is the time delay between the course and the actually foundation (Graevenitz, 2010). Due to the time delay the participant can be influenced by other factors. This makes it difficult to measure how EE influenced the decision. On the other hand EE directly effects the awareness of entrepreneurship, especially when it is a subject on the secondary school (Garavan and O’Cinnéide, 1994). In general, studies examine EE make use of intention models (Martin et al., 2012).

(13)

feasibility to predict the entrepreneurial process (Bird, 1988; Ajzen, 1987, 1991). Attitudinal antecedence can be seen as an independent factor in relation to intention. Kreuger et al. (2000) specified this and came up with a definition for intention: ‘intention is ‘a ‘representation’ of the

willingness or readiness to be an entrepreneur’. This definition for intention is used in this research.

This cognitive representation is a good predictor for actual behavior for two reasons. At first, Kim and Hunter (1993) found, based on meta-analyses, intention can predict behavior for 67 percent. This is in line with the psychological literature; starting a business is seen as planned behavior and in addition planned behavior can be predicted by intention (Kreuger et al., 2000). Secondly, in a longitudinal study of Lee et al. (2011), they examined perceived feasibility and perceived desirability (Shapero and Sokol, 1982) as the most important predictors of entrepreneurial behavior. In conclusion, intention seems a good measurement to analyze the influence of entrepreneurial education. The next paragraphs will give an overview of the concepts determining intention and in what way it can be measured.

2.2.1 Concepts of entrepreneurial intentions

In general, there are two intention models frequently used in the literature; the theories of Ajzens (1991) and Shapero and Sokol (1982). These models are grounded by Vrooms (1964) process-oriented entrepreneurial motivation. He states the intention to become an entrepreneur is predicted by two questions:

- Is entrepreneurship desirable to me (i.e. does it lead to desired outcomes)?; and

- Is entrepreneurship feasible for me (i.e. do I have what it takes to succeed as an entrepreneur)?.

These questions can be applied in the intentions models. First, the model of Ajzen (1991) is discussed, followed by the model of Shapero and Sokol (1982). This intention models are most frequently used (Martin et al., 2012).

2.2.2 Ajzens intention model

(14)

Figure 1 Theory of planned behavior (Ajzen, 1991)

The attitude towards the act (ATB): this means the positive or negative valuation of the consequence by a specific behavior. In this case a positive or negative value of becoming an entrepreneur.

The social norms (SN): this means the perception of important peoples in the individuals’ life, like family, friends and/or role models. And the social pressure an individual feels negatively of positively to perform a specific behavior. This can also be a positive or negative evaluation of behavior by a person who is important to the individual.

The perceived behavourial control (PBC) means the percieved difficulty in performing a specific behavior. Self-efficacy gives an appropriate measurement whether or not a person thinks he is suitable to become an entrepreneur (Krueger and Brazael, 1994).

The three concepts can be summarized in three questions: How desirable is it to perform this behavior? How desirable is this behavior to people who are important to the individual? Do I believe in my own ability to perform this behavior? The theory of planned behavior (hereafter TPB) can be simplified in a mathematical formula:

BI = SN + ATB + PBC

(15)

antecedent of the planned behavior theory influences the perceived feasibility. Perceived feasibility influences career choices. This antecedent perceived behavioral control (Ajzen, 1988), is a perceived control over a specific behavior. This is based on abilities, sources and situational factors and has an internal en external component. The internal component focuses on self-efficacy and the external component focuses on the influence of external factors in contrast to the desirable behavior.

The strongest component affects intention was planned behavioral control. Moreover, a positive relation was found between desirability and feasibility (Kreuger et al., 2000). In contradiction with other studies Fitzsimmons and Douglas (2011) also found a negative interaction effect in determining the strength of their entrepreneurial intention. In line with McMullen and Shepperd (2006), they found when a person has a low desirability but a high feasibility, forms the intention to act as an entrepreneur. And the other way around, meaning when a person really wants to become an entrepreneur, while he or she thinks he or she is not capable to become one, someone will be operating in entrepreneurship. This is in contradiction with the other researchers, because those models argue as well as the perceived desirability, as the feasibility needs to be high to have a high entrepreneurial intention.

There exists controversy in the literature concerning the second antecedent ‘social norms’, especially the direct effect on entrepreneurial intention. Kolvereid and Isaksen (2006) found social norms had a direct significant effect on EI. However, Krueger et al. (2000) found this relationship is not significant. Moreover, Paco et al. (2010) found subjective norms did not have a direct effect for secondary students.

Many researchers neglected the internal component self-efficacy as a factor for intention. However according to Kreuger et al. (2000) this is the most important variable influencing intention: ‘Entrepreneurship researchers largely ignore the concept of self-efficacy despite its importance

and proven robustness at predicting both general and specific behaviors’. Self-efficacy influences the

(16)

to make the choice to become an entrepreneur. In addition, empirical evidence for a positive relationship between self-efficacy and entrepreneurial intentions was found in previous work of Wilson et al. (2007). Wilson et al. (2007) suggest self-efficacy plays a key role in the level of interest as it influences the self-confidence to acquire the skills needed to succeed. Self-efficacy not only influences perceived behavioral control, but also to moderates the relationship between perceived desirability and intention (Lee et al., 2011). For example, when somebody is has low job satisfaction, the propensity to act as an entrepreneur is higher when somebody has a high self-efficacy.

Consequently, self-efficacy has an important place in Ajzens’ perceived behavioral control model and in Shapero's model of perceived feasibility.

2.2.3 Shapero and Sokols model

The model of Shapero and Sokol (1982) is quite similar to Ajzens’ model. The Entrepreneurial Event Theory states the option of becoming an entrepreneur arises as a consequence of an external change, called an event. For instance, a potential opportunity or idea for a new product. The perception of the person of the alternatives and becoming an entrepreneur will influences whether or not someone become an entrepreneur (Linan et al., 2010). The perception has two factors:

- (1) Perceived desirability, the attractiveness of becoming an entrepreneur; and

- (2) Perceived feasibility, the degree someone thinks he can succeed in being an entrepreneur and has the required skills. This is highly influenced by someone's self-efficacy.

Shapero and Sokol (1982) assume individuals can be motivated by a negative or positive experience. When they have a negative prior experience with starting a business, an individual will be less motivated and the other way around. The behavior depends on the propensity to act and the credibility of other behavior.

2.3 Specific model to answer the research question

(17)

Shook et al. (2003) noted: ‘Future work on entrepreneurial intentions should attempt to integrate and

reduce the number of alternative intention models’. This supposes to use a model already validated

and empirical tested.

The models of Ajzen (1991) and Shapero and Sokol (1982) are quite similar due to the focus on willingness and capabilities (Kreuger et al., 2000). Kreuger found strong statistical support for both models.

Fayolle et al. (2006) advocate to use TPB to test the influence of EE on intention. This is in line with van Gelderen et al. (2008), they advise to use TPB to measure intentions. There is broad empirical evidence in the literature and the theoretical specification of TPB is better. The model of Shapero and Sokol (1982) focuses on the event of business creation and not the evolution and the process to becoming an entrepreneur. Though, it could be used as an application to the model of TPB.

(18)

Considering the aforementioned arguments in this research, the TPB will be used to test the intention of students. The attitude towards the behavior is tested through perceived desirability. Perceived behavioral control can be tested through self-efficacy and perceived feasibility. Social norms can influence EI, however their consists some controversy. Attitudes towards behavior are positively related. However, Fitzsimmons and Douglas (2011) found if perceived desirability or feasibility is low, it can be ‘overruled’ by the positive effect of the other and someone show EI.

Fayolle et al. (2006) advocates an assessment model to test intention by using the TPB, the methodology is presented in figure 2. They examined it is necessary to first describe the entrepreneurship education program (EEP) in detail, before applying TPB. The EEP has according to Fayolle et al. (2006) an effect on the constructs of TPB: ‘The latter implies this

framework can be used not only to assess but also to improve the design and execution of EEP, by linking specific characteristics of the EEP with particular outcomes in terms of attitudes and intentions’. It is

conspicuous Fayolle et al. (2006) compile a model without testing the effect of EEP on the different constructs.

2.4 Describing entrepreneurship education programs (EEP)

In the assessment model of Fayolle et al., (2006) EEP can have different audience, objectives, contents, teaching methods and approaches. In this paragraph the literature is reviewed concerning EEP. The different components are described by Fayolle et al. (2006) in brief. They explain what kind of contents, audience, etcetera exist. However, they do not pay attention on how the different components influence ATB, SN and PBC.

Institutional settings

(19)

Audience

The type of participants can influence the effect on intention. Literature does strongly suggest career choices and expectations are formed in the teen years, especially when students are college-bound (Low, Yoon et al., 2005). Wilson et al. (2007) stated middle- and high school students have higher entrepreneurial self-efficacy and therefore intend to become an entrepreneur. Moreover, Fillion (1994) identified adolescents have the ideal age to develop a positive attitude towards entrepreneurship. This is in line with Peterman and Kennedy's (2003) advocate to provide EE at earliest possible age. Argued students should be prepared for their future career.

Pfeier and Reuss (2008) suggest investments in cognitive skills are relative more important during pre-school years. Though, investment in non-cognitive skills, like self-efficacy, is best during primary school and decreases over time. However, in contradiction to the previous most EEP are offered during tertiary school years and years following. Most important of all, there is an innate entrepreneurial attitude in younger students what must be preserved and developed. Students at this age tend to display an entrepreneurial attitude in everything they do. They are usually very creative, straightforward and unconcerned with the potential risks inherent to their actions. In conclusion can be said the effect on TPB should be higher when the audience is young. Younger people are more vulnerable and the career choice is shaped in teen years. Another difference in audience can be gender. Zhao et al. (2005) found a direct link between gender and intention. Females seem to have lower entrepreneurial intention. Similar to this research, Wilson et al. (2007) found the effect on entrepreneurial self-efficacy was higher for men.

Type of EEP and objectives

The different types and objectives are already reviewed in paragraph 2.1 (Entrepreneurial Education). The EEP is categorized (Garavan and O’Cinnéide, 1994) as follows:

(i) entrepreneurial education.

(20)

This can influence intention in different ways. The third category has the aim to influence entrepreneurial intention and is therefore focused on the three constructs of the TPB. This category wants to influence the attitude towards behavior in a positive way. Secondly, the perceived behavioral control should be higher, because participants had more self-confidence in their chances of succeeding as entrepreneur.

In the first two categories people already have the intention to become entrepreneur. They can set up a bigger network of entrepreneurs and therefore increase the social norm (Souitaris et al., 2007). Due to the knowledge gathered in the programs, people should report a higher perceived behavioral control. However, literature never tested this specific relation.

Teaching type and content

Entrepreneurial education is interconnected with business education and the concepts are not always used in a proper way. Therefore Gibb (1993) summed up the main differences between Table 1 University/Business School versus Entrepreneurial Education/Training Focus. (Gibb, 1993)

University/business school learning focus Entrepreneurial education/training

learning focus

Critical judgment after analysis of large amounts of information.

“Good feel” decision making with limited information.

Understanding and recalling the information itself.

Understanding the values of those who transmit and filter information.

Assuming goals away. Recognize the widely varied goals of others.

Seeking the correct answer with time to do it. Developing the most appropriate solution under pressure.

Understanding basic principles of society in the metaphysical sense.

Seeking to apply and adjust in practice to basic principles of society.

Seeking (impersonally) to verify absolute truth by study of information.

Making decisions on the basis of judgment of trust and competence of people.

(21)

business education and entrepreneurial education (see table 1). Some characteristics can be deleted or swapped for a specific program. Martin et al. (2012) advocates when training is focused on learning by doing, it is more likely to affect EI. They stated: ‘demonstrate the core

entrepreneurship knowledge and skills required to start a particular business in a particular setting, because the learning and application context and content are more “near” than with academic-focused EET’.

Teaching methods and approaches

Hytti and O’Gorman (2004) distinguish action-based and traditional learning methods. Action-based learning is described as: students learn by doing and teachers are coachers and facilitators and the goal is to learn ‘how to act’ as an entrepreneur. Action-based learning is the most common used method for entrepreneurial education programs. Still, there are some difficulties, like the balance between teacher and coach when students are independent. They may feel the process is slow, the experience frustrating and there is an excessive workload. This can be overcome by a set of interventions, no authorized teaching, but to provide students with questions so they can address their own critical issues. Second, programs can give an overload of work for students and teachers, especially when it is extra, next to the standard curriculum (Mueller, 2011).

‘Traditional teaching’ is the other teaching method studied by Hytti and O’Gorman (2004); learning through lectures and assignment. A better understanding by students of the benefits from entrepreneurial activity is the main goal. Consequently, the greatest influence on secondary school students is not found through traditional teaching, but action-based teaching. Rasmussen and Sorheim (2006) found a great effect of the program on intention, because Gleaning information from experts and

authoritative sources.

Gleaning information personally from any and everywhere, and weighing it.

Evaluation through written assessment. Evaluation by judgment of people and events through direct feedback.

Success in learning measured by examination knowledge-based.

(22)

students were experiencing the phenomena entrepreneurship. In addition, Mueller (2011) found this way of teaching substantially increases students’ self-efficacy and therefore the ‘perceived behavioral control’. More specific, Mueller advocates, the course needs to conduct practical experience, business planning activities, interactive elements or integrated feedback processes. Secondly, the classes need to be oriented and focused on reflective elements. The greatest influence was found while students were learning by doing and direct feedback.

Conclusion

In conclusion, audience, the type of EEP and teaching approaches are valuable within this research. Firstly, the audiences in this research are secondary students. The existing literature advocate EE has a great influence on this group, because this age group is most vulnerable for change. Secondly, the type of EEP is ‘small business awareness education’ and thirdly the teaching approach is action-based teaching. According to the literature, this method is the best way to change secondary students’ attitude towards EI.

2.5 Overview of studies on entrepreneurial intention

Controversy to Fiet (2000), O’Gorman and Cunningham (1997) state in their meta-analysis entrepreneurship can be taught. On the other hand there is some contradiction according to the effects of entrepreneurial education (EE). The effect of EE on entrepreneurial intention (EI) is studied multiple times. The most significant studies are illustrated in table 2; what theory, sample size, method and the main conclusion. A few studies found a positive effect of EE on entrepreneurial intention (Tkachev and Kolvereid, 1999; Peterman and Kennedy, 2003; Fayolle et al., 2006; Souitaris et al., 2007; Sanchez, 2010; Paco et al., 2010) on intention or venture creation. Others (Oosterbeek et al, 2010; Von Graevenitz et al., 2010) found a negative effect. The most interesting studies are reviewed in depth.

(23)

a post test. Martin et al. (2012) advocate in their meta-analyses future studies should use pre-post-test methods, because this is the only way to measure real changes within and between groups. Allowing for the time lapse, the study must consists of two or more questionnaires. Furthermore, Peterman and Kennedy (2003) found a positive effect on perceived feasibility and desirability after participating in the EE. However, the effect was significant higher by students who had less prior experience. When students were already aware of the opportunity, the effect of EE is not high. Although they used a pre-post-test method, they missed a trivial point in their research. The control and treatment groups do not involve randomized assignment. When groups are not random assigned there exists a selection bias; students who choose to follow a particular EE are already interested and show more interest towards a career choice as entrepreneur.

The study of Oosterbeek et al. (2010) has random assignment. A total of 250 students on higher education were questioned during the research, 146 in the treatment group and 104 in the control group. They used the E-scan to measure intention and attributes. Different from other studies, they did not use a verified intention model. The E-scan, developed by Driessen (2005), is used as a tool in the Netherlands to test if a person is suitable, through testing skills, to start an own business. The intention was measured by one question: ‘do you want to start a business’. The students who followed the program were business students on different schools. As Fayolle (2006) noted, institutional settings could be an influence on the constructs of TPB. Nonetheless, Oosterbeek et al. (2010) found a negative effect. They think the negative effect occurred due to a loss of over-optimism. Although this is the only study with a methodology advocated by Martin et al. (2012) they do not use a common intention model. Therefore, they could not quantify why there is a loss of EI occurred in the program. None of the constructs of TPB (Ajzen, 1991) or EET (Shapero and Sokol, 1982) is examined.

(24)

business. He found a great trigger when the required competences to begin an 'entrepreneurial adventure', like self-efficacy, pro activeness and risk-taking, are trained. This study illustrated EE is a trigger-event and influence inspiration (emotional mindset) and therefore intention. Souitaris et al. (2007) dedicate the positive effect on subjective norm and attitude towards behavior to the enthusiasm of their project. However, they did not find a positive effect on perceived behavioral control, because elite-students already have a high self-efficacy. Similar to Souitaris et al.'s research, Sanchez (2010) explores 400 university students who chose an elective module. Sanchez found a significant positive effect on self-efficacy, risk taking and intentions in contrast to the control group.

Fourthly, Graevenitz et al.'s (2010) study has some method limitations; they did not use a verified model, had no control group and the students were not randomly assigned. However, they had some interesting remarks according to the sorting function of EE. The students did not choose the specific course as it was obligatory in the curriculum of the business school in Munich. It differed from other studies by focusing on entrepreneurial aptitude as a learning outcome: ‘The most important effect of entrepreneurship education may therefore lie in students

adjusting and refining their assessment of their own entrepreneurial aptitude’. This view implies some

students learn entrepreneurial careers are not well-suited for them while the educational measures lead other participants to come to a more positive assessment of their entrepreneurial aptitude. Nevertheless, they do not test the sorting function in a proper way. They only concluded the students were significantly more aware whether or not an entrepreneurial career suits for them, without hard facts. Moreover, this is found by Scherer et al. (1991). He concluded the perceived feasibility and desirability will amplify as a consequence of the greater awareness of an extra career option.

(25)

researchers (Graevenitz, 2010; Noel, 2002; Kennedy and Peterman, 2003). Besides a trigger function for ‘attitudes towards behavior’ and ‘perceived behavioral control’, it can have also a sorting function, because it is a first acquaintance with entrepreneurship.

(26)

Theory Sample Method Result Reference

Entrepreneurial even theory

- Secondary students

- 110 student in treatment group and - 110 students control group

- No random assignment

Pre-test-post-test

Positive effect on perceived feasibility and desirability.

Peterman and Kennedy, 2003

TPB - University students

- 144 students in treatment group - 126 students in control group - No random assignment

Pre-test-post-test

Subjective norms increased. Perceived Behavioral Control did not change.

Inspiration and intention is the program’s biggest benefit.

Souitaris et al., 2007

Escan (Driessen, 2005)

- Higher education

- 146 students in treatment group - 104 students in control group - Random assignment

Pre-test-post-test

Negative effect, loss of over-optimism. Oosterbeek et al., 2010 A model of learning about own entrepreneurial ability

- German business university - 196 in treatment group - No control group - No random assignment

Pre-test-post-test

Intentions decline because of sorting function.

Graevenitz, 2010

(27)

- 265 students - No control group - No random assignment

post-test entrepreneurial intentions.

TPB - Graduated

- 84 treatment group - 84 control group

- No random assingnment

Post-test Higher intention to start a business.

Noel, 2002

EEP - University

- 20 students treatment group - No control group

- Random assignment

Pre-test-post-test

Strong impact on entrepreneurial intention.

Not very significant, but a positive impact on perceived behavioral control.

Fayolle et al., 2006

TPB - University

- 403 students treatment group - 460 students control group - No random assingnment

Pre-test-post-test

Increase in competencies and intention towards

self-employment.

(28)

This chapter will describe the method used to answer the research question. A quantitative approach is used to identify whether the entrepreneurial education changes the intentions of students in the Netherlands or not. The empirical study consists of two questionnaires in April and July with matched pairs. Both questionnaires measure the attitude towards behavior and perceived behavioral control, prior experience and intention towards entrepreneurship. To measure the change in entrepreneurial intentions the pretest-posttest design (Cohen and Manion, 1989) is applied. This method is chosen to measure the development of entrepreneurial intentions between the beginning of the program (Tstart) and the end of the program (Tend). First,

this chapter will give an overview of the sample groups, followed by the variable operationalization and questionnaire development.

3.1 Sample description

The dataset consists of four groups: three control groups and one experimental group. All students followed education on Dr. Nassaucollege in Assen, in the North of the Netherlands. This is a secondary school and the students follow the curriculum of VWO, the highest level of education in the Netherlands. The students are in the first class, which means they are about 12 to 13 years old. Table 3 gives an overview of the sample composition. It is important to notice all

Table 3 Sample composition

Group 1 2 3 4 Total

Classification Experimental Control Control Control -

Experiment Entrepreneurial None Engineering Cultural -

Action-based learning Yes No Yes Yes -

(29)

follow the same curriculum. Besides the normal curriculum of VWO group 1, 3 and 4 follow a specialization course; they have four extra hours during the week. In these four hours students follow a project in the field they have chosen (group 3 and 4), or a project randomly assigned to them (1). This opportunity of four different groups occurs because Dr. Nassaucollege is ‘in the race’ for a certificate called ‘Business School’. To make a chance in being a Business School a school needs an entrepreneurial project. Therefore one class is chosen to participate in this project. The students in the engineering and cultural group voluntarily choose to participate in the specialization at the beginning of the year.

The reason for choosing three control groups is twofold:

Beforehand the treatment and control group followed the same (normal) curriculum. In April the groups are randomly assigned. Group 1 participates in the EE and group 2 do not and therefore do not have extra hours. The difference-in-difference analyzes between these groups show the changes due to the EE, because there do not exists pre-treatment differences. In addition, group 1 can see EE as a punishment, the result will show if this has an effect.

The control groups with the specialization courses engineering and cultural have chosen the specialization by themselves at the beginning of the year. Before the first questionnaire, they have followed a project over a period of eight weeks. It is interesting to measure and see if these children score higher or lower on different constructs at the pre-test.

All groups follow a specific program to develop certain competences and to make them aware of the career choices:

Entrepreneurial specialization, group 1

This is the group randomly assigned to EE. Group 1 followed entrepreneurial education between Tstart and Tend. The way of teaching used in the treatment group is action-based. The

(30)

elevator pitch to sell their product. Students follow the specialization four hours a week during eight weeks.

Normal curriculum, none specialization, group 2

This is the group who followed the standard curriculum in the first year of secondary school (e.i. without specialization). This curriculum is followed by all the students in the sample. It is obligatory for students at VWO. Students follow courses, such as Dutch, English, Mathematics, Music etc. None of these subjects have special attention for entrepreneurial subjects and most of the lessons are structured based on the principle of traditional teaching (Hytti and O’Gorman, 2004). Students in this control group did not choose to do a specialization. Until April, the treatment group followed the same program as the students in this group.

Engineering specialization, group 3

Engineering education is the most developed and popular program inside the school. The school possesses a network of businesses around the school whom deliver ‘real world assignments’. Although, every individual develops a working method weekly, the students work in groups. The program has a large connection with EE, considering the teachers encourage students to think ‘out of the box’ by developing a product. Moreover, there is a reflection part (Mueller, 2011) in the program. At the end of a period (2.5 months) students need to describe their own strengths and weakness and the strengths and weakness of their group members.

In the specific period, between Tstart and Tend, the assignment was to develop a bridge for the city

of Assen. The assignment was attached to the profession of civil engineer. Cultural specialization, group 4

(31)

3.2 Variable operationalization

In this research, the model of Fayolle et al. (2006) is used: Entrepreneurship Education Programs (EEP). The basis for EEP is Ajzens TPB (1991). The TBP has three constructs: ‘attitude towards behavior’ (ATB), ‘social norms’ (SN) and ‘perceived behavioral control’ (PBC). In the model used, EEP is added since TPB only examined the direct effects of the constructs on entrepreneurial intention, while audience, teaching approaches and so on can affect intention directly and indirectly.

However, in this research some subjects in EEP are neglected for the in-between differences. The sample is too small and therefore it is not possible to significantly test the differences between the control groups and the treatment group. The audience and (other) institutional settings are the same for all groups. Therefore it is challenging to make a link between the content and the influence on the TPB. In this research only the direct link between EEP and the ATB and PBC is studied. Since the teaching method in group 1, 3 and 4 is action-based and the method in group 2 is traditional differences can be found due to the teaching methods (Meuller, 2011). Secondly, the content of the specializations are different; entrepreneurship, engineer and cultural. And therefore differences can be reported due to the content. The model used in this research is presented in figure 3.

What remains is the TPB. There exists controversy in the literature about the effect of ‘subjective norms’ in TPB. Paco et al. (2010) already tested SN by secondary students and found SN does not have significant influence on EI. They advocate to delete the social norms part in the questionnaire. Moreover, it is difficult to measure social pressure in a questionnaire. Therefore SN are neglected in this research. This is offset through the use of role model (parents) as a control variable in the analyses.

(32)

approaches of Wilson et al. and Peterman and Kennedy. Secondly, this is legitimatized, since a reliable and valid questionnaire for this age-group do not exists in the literature. Therefore, the use of two different approaches gives a broad result and there is a possibility to delete variables if they are not consistent within the construct. Since the questionnaires are put together, the dis-advantage could be the questionnaire is to extensive. However, the dis-advantages compensate for this dis-adavantage.

3.2.1

Attitude towards the

act

The attitude towards the act means whether the valuation of the consequence by a specific behavior is considered positive or negative. To test these construct Wilson et al. (2007) and Peterman and Kennedy (2003) used different questions. Peterman and Kennedy used the construct ‘perceived desirability’ in the model of Shapero and Sokol (1982). According to Kreuger et al. (2000), the constructs ‘perceived desirability’ and ‘attitude towards the act’ are interchangeable. They both test valuation to display a specific behavior.

Wilson et al. (2007) used the same method as Autio et al. (2001) did and asked how desirable professions like nurse, soldier, writer, entrepreneur etcetera are. Peterman and Kennedy derived their questions from Krueger (2003) and transformed them to fit for secondary students. Combining these methods result in four questions to test the construct ATB (see table 4).

(33)

Table 4 Questionnaire attitude towards behavior

3.2.2 Perceived behavioral control

The perceived behavourial control is best described as the difficulty or ease someone thinks to have in performing a specific behaviour. PBC has an external and internal component. The external component accounts for the control someone has over external factors, for example funding. Students aged 12/13 do not have the capabilities and responsibility to control external factors and are unable to evaluate the control over external factors in the future. Therefore the external component is neglected and the questionnaire measure the internal component of PBC.

Again the questionnaires of Peterman and Kennedy (2003) and Wilson et al. (2007) are used. The researchers used different method to measure the internal component of PBC. Peterman and Kennedy developed five questions to classify the perceived behavioral control (see table 5).

Career: Definitely

not interested

Probably not Possible interested Somewhat interested Extremely interested 1. Entrepreneur ☐ ☐ ☐ ☐ ☐ 2. Starting/owning my own business ☐ ☐ ☐ ☐ ☐

Do you agree with these statements? Strongly

disagree

Disagree Neutral Agree Absolutely agree

3. I will enjoy starting an own business _☐ _☐ _☐ _☐ _☐

4. I think I will be enthusiastic owning an own business

☐ ☐ ☐ ☐ ☐

Table 5 Questionnaire perceived behavioral control

Do you agree with these statements? Strongly

disagree

Disagree Neutral Agree Absolutely agree

1. I think it is easy to start a own business _☐ _☐ _☐ _☐ _☐

2. I think I will be successful owning an company

☐ ☐ ☐ ☐ ☐

3. I think I will be overworked owning an company

☐ ☐ ☐ ☐ ☐

4. I am smart enough to start an own business

☐ ☐ ☐ ☐ ☐

5. I believe in my own capabilities _☐ _☐ _☐ _☐ _☐

6. I think I will be stressed when I own my own business

(34)

They tested how well somebody thinks he is in owning a business and the self-confidence of the students. Wilson et al. (2007) illustrate self-efficacy through six questions where students need to qualify themselves against peers (see table 6). Question three and six (see table 5) are recoded to give consistent constructs. The questions are inversed versions of the other four questions and therefore recoded as follows: 1=5, 2=4, 3=3, 4=2 and 5=1.

The questions of Peterman and Kennedy (2003) and Wilson et al. (2007) are not straightforward. The questions are to broad and not specifically focused on entrepreneurial intention. Question 2 and 4 (see table 5) are specifically focusses on PBC. Question 1, 3 and 6 (see table 6) do not focus on entrepreneurial attributes. The reason for a negative answer can be he or she knows it is not easy to start a business, while he or she feels is able to succeed as entrepreneur. However, these questionnaires are chosen because of the lack of questionnaires specifically for secondary students.

3.2.3 Entrepreneurial intention

Entrepreneurial intention can be measured from a behavioral perspective (I intend to perform behavior X). Peterman and Kennedy (2003) used this perspective in their survey; they asked whether or not a student wants to start a business.

Table 6 Questionnaire self-efficacy

Ik ben slechter of beter dan andere leerlingen in de 1ste_{klas in:}

Much worse A little worse About the same A little better Much better

7. Being able to solve problems _☐ _☐ _☐ _☐ _☐

8. Managing money _☐ _☐ _☐ _☐ _☐

9. Be creative _☐ _☐ _☐ _☐ _☐

10. Getting people to agree with you _☐ _☐ _☐ _☐ _☐

11. Be a leader _☐ _☐ _☐ _☐ _☐

12. Making decisions _☐ _☐ _☐ _☐ _☐

Table 7 Question Entrepreneurial Intention

Do you think you will ever start a business?

(35)

3.3 Questionnaire development

After designing a first draft, the questionnaire was tested in a pilot (N=5). In a session children in the same age group as the target group were asked to make the survey. This session gave the opportunity to find flaws in the current questionnaire. After the pilot the final questionnaire was developed (see appendix C and D). The questionnaire consists of five section: (i) the first section starts with asking names to match pairs, gender, and which stream they follow to conduct the groups. (ii) The second section tests desirability as described by Wilson et al (2007) and identifies how interested students were in specific jobs, including entrepreneur and small business owner. (iii) In the third section, the questionnaire of Peterman and Kennedy (2003) was used and identifies the ‘attitudes towards behavior’ and the ‘perceived behavioral control’. The perceived behavioral control was tested by questions like ‘Would you be successful as entrepreneur?’. (iv)The fourth section measured prior experience for the analyses. Do parents or acquaintances have experience with entrepreneurship and did they even start a company by themselves. (v) The fifth and last section tested self-efficacy. The questions in this section are also derived from Wilson et al. (2007).

3.4 Data collection procedure

(36)

September July

3.5 Data Analysis Procedures

A pre-test-post-test (Cohen, L., and Manion, L. (1994) control group design was adopted to measure the change in intentions of the treatment group and control group. The questionnaires were not anonymous, so it is possible to link the questionnaires individual and per group. The data analyses are executed using the statistical program Statistical Package for the Social Sciences (SPSS).

To see if there exist significant pre-treatment differences between the groups, a t-test was conducted. The differences between the treatment group and the three groups separately were compared. To test the changes in and between groups, a difference-in-difference framework was utilized (Oosterbeek et al., 2010). First, the difference scores between Tend and Tstart were

calculated for all groups. The second step was to subtract the difference scores of the treatment group of the control groups. Finally, a t-test was conducted at difference-in-difference scores.

Start all groups with curriculum VWO. Control

group engineering and cultural start with

specialization. April 15th: Tstart. April: Start treatment group July 2nd: Tend

Mid July: End of program treatment and control groups

(37)

4-37| The influence of entrepreneurial education on entrepreneurial intention

4 Analysis and discussion

Firstly, the descriptive statistics are analyzed and discussed in this chapter. Secondly a correlation table is analyzed and the validity and the reliability are checked. At last, the three constructs ‘attitude towards behavior’ (ATB), ‘perceived behavioral control’ (PBC) and ‘self-efficacy’ (SE) are discussed by analyzing the pre-treatment difference and the change in and between the groups due to the program.

4.1 Descriptive statistics

The sample consists of 51.8% male respondents and 48.2% female respondents. According to the data of CBS (2011), 54.4% of the students following VWO is male and 45.6% is female. Therefore, the sample is a good representation of the population following VWO in the Netherlands.

Nevertheless, the sample composition in the groups is not in proportion with the numbers of CBS. With a 100% score for female, the cultural group is most out of proportion. The project is focused on creative exploration and obviously it attracts only females. This is also found in the correlation (see table 9) between gender and creativity, females scored higher on this component of self-efficacy. The other way around, nearly three quarters of the engineering group comprise of male students.

In addition, group one and two are randomly selected and due to a small sample effect the distribution is inverse. With 40% male and 60% female in group 1 and 43.75% female and 56.25% male in group.

Table 8 Pre-treatment descriptive statistics Yes =1 – No =2 Experimental group

(1) None specialization control group (2) Engineering specialization control group (4) Cultural specialization group (3)

Start Start Start Start

Parents own business? 1,70 1,75 1,70 1,92

(38)

According to Peterman and Kennedy (2003), prior experience (parents and acquaintances with an own business) can influence the choice to participate in a program. Moreover, Martin et al. (2012) see prior experience as the most important human capital variable. They state: ‘prior

entrepreneurial experience is the most important human capital variable in the determination of entrepreneurial intentions, and, along with self-efﬁcacy and attitude to ownership, provides a sufﬁcient characterization of the individual when the interaction between perceived desirability and perceived feasibility is included in the model of entrepreneurial intentions’.

(39)

Table 9 Correlation table

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1. Gender

1 -,069 -,117 -,251* ,096 -,070 -,003 ,002 -,054 -,044 ,037 ,259* -,280** ,221* ,181 -,070 -,106 -,175

ATB - Attitude towards behavior 2. Entrepreneur -,069 1 ,478** ,546** ,413** ,025 ,280** ,029 ,086 -,044 -,260* ,063 ,209 -,117 ,194 ,235* ,236* ,386** 3. Own business -,117 ,478** 1 ,575** ,478** ,138 ,391** -,047 ,143 ,083 ,182 -,004 ,147 -,168 ,105 ,168 ,078 ,380** 4. Enjoy -,251* ,546** ,575** 1 ,528** ,071 ,343** -,149 ,030 ,134 -,249* -,022 ,113 -,044 ,062 ,360** ,021 ,467** 5. Enthusiastic ,096 ,413** ,478** ,528** 1 ,093 ,453** -,242* ,047 ,090 -,347** ,141 -,125 -,016 ,081 ,196 ,096 ,211

PBC – Perceived behavioral control 6. Easy -,070 ,025 ,138 ,071 ,093 1 ,320** -,018 ,047 -,024 -,251* -,060 ,041 -,063 -,059 -,053 -,174 ,064 7. Successful -,003 ,280** ,391** ,343** ,453** ,320** 1 ,017 ,249* ,235* -,260* ,209 -,029 -,245* ,215* ,214* ,176 ,367** 8. Working hard ,002 ,029 -,047 -,149 -,242* -,018 ,017 1 -,090 ,042 -,264* -,188 ,096 ,000 ,087 ,060 ,001 -,039 9. Smart enough -,054 ,086 ,143 ,030 ,047 ,047 ,249* -,090 1 ,176 -,080 ,364** ,207 -,251* -,181 -,090 ,153 ,163 10. Faith -,044 -,044 ,083 ,134 ,090 -,024 ,235* ,042 ,176 1 -,238* ,184 ,060 -,102 -,003 ,162 ,319** -,092 11. Stressed ,037 -,260* -182 -,249* -,347** -,251* -,260* -,264* -,080 -,238* 1 -,064 -,080 ,282** -,008 -,021 -,128 ,073 SE – Self-efficacy 12. Problem solving ,259* ,063 -,004 -,022 ,141 -,060 ,209 -,188 ,364** ,184 -,064 1 ,064 -,079 ,144 ,076 ,207 -,083 13. Handling money -,280** _,209 _,147 _,113 _-,125 _,041 _-,029 _,096 _,207 _,060 _-,080 _,064 ₁ _-,056 _-,029 _-,046 _,048 _,164 14. Creative ,221* -,117 -,168 -,044 -,016 -,063 -,245* ,000 -,251* -,102 -282** -,079 -,056 1 ,000 ,056 -,078 ,095 15. Getting to agree ,181 ,194 ,105 ,062 ,081 -,059 ,215* ,087 -,181 -,003 -,008 ,144 -,029 ,000 1 ,426** ,161 ,193 16. Be a leader -,070 ,235* _,168 _,360** _,196 _-,053 _,214* _,060 _-,090 _,162 _-,021 _,076 _-,046 _,056 _,426** ₁ _,269* _,181 17. Decision making -,106 ,236* ,078 ,021 ,096 -,174 ,176 ,001 ,153 ,319** -,128 ,207 ,048 -,078 ,161 ,269* 1 ,073 EI - Entrepreneurial intention 18. Start an own business?

(40)

should be at least 30 and measured on an interval or ratio scale (Chen and Popovich, 2002). This research meets these requirements, because the sample size is 85 and the variables are measured on an interval scale. The correlation table is displayed in table 9. According to Cohen and Manion (1989) a correlation of 0.10 is weak or small, a correlation of 0.30 moderate, and a correlation of 0.50 or higher is strong.

In the construct ‘attitude towards behavior’ (ATB) all variables are positively correlated. This argues for a valid and reliable construct, tested in the next paragraph. Three out of four variables correlate with successfulness, stress and EI. When the attitude towards behavior is positive, they also score high on perceived successfulness in owning a company. When somebody is thinking he is successful and the stress for choosing a career is low, a logical consequence is the attitude towards such behavior is positive. The high correlation between TPB and EI shows ‘attitude towards behavior’ is positively interconnected with EI. This is similar to studies of Peterman and Kennedy (2003), Souitaris et al. (2006) and Martin et al. (2012).

In the second construct, ‘perceived behavioral control’ (PBC), the correlation between most of the variables is not significant. This makes a case for a negative valuation for the use, by students aged 12/13, of the questionnaire of Peterman and Kennedy (2003). In the correlation table ‘successfulness’ and ‘smart enough’ has a positive correlation with each other and EI. Except these questions, as mentioned before, the questions are to broad and do not give a good measurement for PBC. The correlation table found this statement due to the insignificance of the construct.

(41)

is interesting. In general, artists are not the most capable managers/ entrepreneurs, obviously students at the age of 12 are aware of this.

4.2.1 Reliability

Litwin (1995) recommends to measure internal consistency by the Cronbach's coefficient alpha. Table 10 provides an overview of the Cronbach’s alpha’s for the constructs. Coefficients of the construct ‘attitude towards behaviour’ are high (>.775) for both questionnaires.. According to a general rule in the literature, a construct is consistent when the Cronbach alpha is >.6. Therefore we can assume ‘attitude towards behavior’ is consistent in this research.

In contradiction, the Cronbach’s Alpha’s for the perceived behavioral control and self-efficacy do not meet the general rule. In table 11 the Cronbach Alpha’s are shown for PBC when items are deleted. Deleting items in this section of the questionnaire should not result in a Cronbach’s Alpha higher than .6. The variables ‘successful’ and ‘smart enough’ show lower Cronbach’s Alpha then the current Cronbach Alpha (.503 and .217). This means the variables have a positive influence and therefore are consistent within the construct. In contradiction, ‘working Table 10 Cronbach Alpha's Constructs

Cronbach Alpha Tstart Cronbach Alpha Tend Attitude towards the behavior .794 .775 Perceived behavioral control ,503 ,217 Self-efficacy _,305 _,504

Table 12 Cronbach’s Alpha Perceived Behavioral Control

PBC

_T_start _T_end Cronbach's Alpha if Item Deleted Cronbach's Alpha if Item Deleted Easy ,477 ,143 Successful ,361 ,194 Working hard _,535 _,432 Smart enough ,498 ,095 Faith ,465 -,014a Stressed _,366 _,157

Table 11 Cronbach’s Alpha Self-efficacy

(42)

hard’ and ‘faith’ show the opposite and are not consistent within the construct. This result is in line with the correlations found.

It seems this construct, used by Peterman and Kennedy (2003), is not suitable for testing at an age of 12/13. The reason for this can be students cannot imagine to start and run an own company. Therefore they do not give consistent answers. I suggest to compile a new questionnaire for this young group of students. A brief description or practical example of an entrepreneur should be given. This way should make answers more consistent.

Self-efficacy is the other construct that did not work out. The Cronbach’s Alpha if items were deleted are presented in table 12. The variables ‘creativity’ and ‘handling money’ show inconsistency within the construct (higher Cronbach’s Alpha if item deleted). Students face difficulties to qualify themselves against peers when it comes to ‘handling money’. Students at the age of 12 do not have a consistent self-perception and it evolves overtime due to puberty. To compile reliable and validated constructs, some of the items within the construct self-efficacy are deleted in the constructs. Three variables were deleted; how they deal with money, creativity and if they can get what they want. A reason for the low Cronbach Alpha’s can be students misinterpret the questions. Especially questions on the subjects 'handling money' and ‘getting what you want’ are sensitive for

misinterpretation for children of the age of twelve. The first one is difficult because they only receive pocket-money and the second is not seen as a positive personal trait.

After confirming the reliability, the final and last step before the result section is to test the validity with a factor analysis. Factor analysis is an appropriate way for examining construct validity

(Nunnally and Bernstein, 1994). Construct

validity indicates if there is a close fit between the

constructs. The factor analyses are only conducted for the constructs which has a Cronbach’s Alpha higher than 0.6. These factor loadings, the correlation between the variables and the

(43)

factor, are higher than 0.6 in all cases (see table 11). In conclusion, the analysis provides evidence to support the factors.

4.2.2 Preferred variables

After discussing the correlation, validity and reliability, some variables are preferred. While others are deleted before discussing and analyzing the changes due to the treatment. The constructs for PBC and SE did not work out for this specific age-group. Therefore, in the construct PBC, the variables ‘successfulness’ and ‘smart enough’ are analyzed for PBC. These variables display a positive correlation with EI, consistency in the construct and are the most straightforward questions for the students. The four variables deleted do not.

The influence of entrepreneurial education on entrepreneurial intention

The influence of

entrepreneurial education

on entrepreneurial

intention

Abstract

Acknowledge

Table of Contents

List of figures

List of Tables

1 Introduction

2 Literature review

2.1 Entrepreneurial education

2.2 Entrepreneurial education (EE) and intention

2.3 Specific model to answer the research question

2.4 Describing entrepreneurship education programs (EEP)

2.5 Overview of studies on entrepreneurial intention

3.1 Sample description

3.2 Variable operationalization

Attitude towards the

act

3.3 Questionnaire development

3.4 Data collection procedure

3.5 Data Analysis Procedures

4 Analysis and discussion

4.1 Descriptive statistics

PBC

(Nunnally and Bernstein, 1994). Construct