Life after university, outcomes of
entrepreneurial education.
A study on the effect of different factors on the self perceived entrepreneurial self efficacy in entrepreneurial education programs in the Netherlands
Anton Stienen (1766864)
Msc Entrepreneurship - Vrije Universiteit & Universiteit van Amsterdam Supervisor: F. Meddens
Table of contents
1. Abstract ... 3 2. Introduction ... 4 3. Literary overview ... 6 2.1. Self efficacy ... 6 2.2. Entrepreneurial Self-Efficacy ... 82.3. Entrepreneurial education and entrepreneurial self-efficacy ... 9
2.4. Entrepreneurial minor programs in the Netherlands ... 12
2.4.1 Minor entrepreneurship of the University of Amsterdam (UvA) ... 14
2.4.1 Minor entrepreneurship of the Erasmus University in Rotterdam (EUR) ... 15
3. Data & Method ... 17
3.1. Sample description ... 17 3.2. Study material ... 18 3.3 Statistical Analysis ... 24 4. Results ... 28 4.1. Sample characteristics ... 28 4.2. Dependent variable ... 30
4.3. Correlation and multi collinearity ... 32
4.4. Regression ... 34
5. Discussion ... 36
5.1. Summary of results ... 36
5.2. Interpretation ... 37
5.3. Contribution, limitations and further research ... 39
6. References ... 42
7. Appendix ... 45
3
1. Abstract
The purpose of this paper is to contribute to a better understanding of the effects of a entrepreneurial education
program (EEP) on entrepreneurial self-efficacy (ESE) by exploring the relationship between perceived ESE and
different university EEPs and examining which factors enhance this (possible) relationship. A survey was sent to
alumni of different EEP programs in the Netherlands. Participants (n=73) reported a perceived increase in ESE
as a result from following the minor. Results suggest that the quality grade given to the minor by the alumni
positively influence the perceived change in ESE. Gender is found to be a marginally significant factor,
predicting higher perceived change in ESE for males than females. Faculty has been found to be marginally
significant as well, suggesting a higher perceived change in ESE for students who followed a bachelor in other
fields than business or economics. Also the requirement or stimulation to start a business during the EEP shows a
marginally significant effect size, indicating higher perceived ESE as a result of the EEP for students who were
required or stimulated to do so, compared to those who were not. Furthermore, the variables Nationality,
Finished the first 60 ECT of the bachelor in one (study) year, Year minor started and Entrepreneurial experience
before the minor have not found to be a significant factor suggesting that, according to this study, they do not
influence self perceived ESE as a result of following EEPs.
This study contributes to scientific literature by supporting previous claims that EEP could have a positive
impact on individual levels of ESE. Additionally we identified factors that could influence the relationship
between ESE and EEPs. However, more testing needs to be done as the field is relatively understudied. This
study therefore opts for more studies investigating ESE as an outcome measure for EEPs. In order to do so, there
needs to be consensus about how to measure ESE and the use of control groups and longitudinal studies to
achieve better understanding in the field. The value of this research lies in the fact that it responds to calls in
scientific literature for more testing on ESE as an outcome of EEPs and the identification of factors that could
enhance this relationship.
4
2. Introduction
Entrepreneurship as a topic continues to attract interest from both academicians and policy makers. Policymakers generally believe that, in order to reach higher levels in economic growth and innovation, a higher level of entrepreneurial activity must be achieved and academics found a positive relation between entrepreneurial activity and economic growth (Van Praag and Versloot, 2007). As interest from students about entrepreneurial careers increased (Fleming, 1994)and policymakers’ awareness with respect to the importance of entrepreneurship as a contributor to economic development grew (Hytti and Kuopusjarvi, 2004), more and more specialized entrepreneurship education programs (EEPs) became available (Jones, 2012).
Although a positive relation between levels of general education on entrepreneurial activity (Van der Sluis, van Praag & van Witteloostuijn, 2006) and entrepreneurial performance have been found (Nabi and Linan 2011), little is actually known about the impact of EEP on entrepreneurial intentions (Sanchez, 2013). Not surprisingly, the relation between EEP and new business development remains unproved (Dickson et al., 2008).
Many scholars (Corbett 2011; Maritz and Brown 2013) conclude that research that aims to identify the best practice for EEP has been unable to keep up with the rapid growth that characterizes EEP, leading to a gap in the literature in how to shape EEPs to effectively educate entrepreneurs and equiponderate different aspects of education. As a results,
researchers are groping in the dark about what works and what does not (Maritz and Brown 2013). Due to the fact that little agreement exist among researchers on how to stimulate entrepreneurship, a wide range of different frameworks and constructs are used in studies in order to get a better understanding of entrepreneurial learning (Fayolle, Gailly, and Lassas-Clerc, 2006). It is therefore not without reason that Jones, Matlay and Maritz (2012)
5
emphasize the gap in literature and express the need for further investigation of the relationship between entrepreneurial learning and EEPs.
A promising concept within entrepreneurial learning, and a key tool in EEP, is entrepreneurial self-efficacy (ESE), which, according to Bandura (1977) refers to “a person’s belief in his or her capability to perform a given entrepreneurial task” , as it could be elevated by training and education (Florin, Karri, and Rossiter, 2007; Zhao, Seibert, and Hills, 2005). ESE has been demonstrated as particularly important in the development of entrepreneurial intentions (Barbosa, Gerhardt and Kickul, 2007) due to the fact that it incorporates personality as well as environmental factors, therefore, extending on Bird’s model (1988) of intentions, making it a strong predictor of entrepreneurial intentions and ultimately action. Pihie and Bagheri (2010), amongst others, argue that entrepreneurial education programs should be designed in order to maximize the entrepreneurial self-efficacy.
Surprisingly, ESE has not been properly investigated as an outcome measure for EEPs (Wilson, Kickul, and Marlino, 2007). Although some studies (Maritz and Brown, 2013; Wilson et al., 2007) looked at the effect of following an EEP on ESE, only the latter based their research on an EEP at university level.
This paper attempts to contribute to a better understanding of the effects of EEP on ESE by exploring the relationship between perceived ESE and different university EEPs and examining which factors enhance this (possible) relationship. By doing so, more will be known about factors such as age and gender, and other factors, which have not yet been studied in EEP and ESE context, will be introduced and tested. Ultimately, a better understanding of the effects of EEP on ESE could lead to better designed EEP’s, possibly resulting in an increase of entrepreneurial activity enhancing economic growth.
6
“What factors influence the self-perceived change in entrepreneurial self-efficacy in entrepreneurial education programs?”
3. Literary overview
This chapter introduces the research area and important constructs will be defined by
providing an extensive overview on what is known on the relevant topics. Firstly, self-efficacy will be defined and will be elaborated upon. Secondly, the concept of entrepreneurial self-efficacy will be discussed. Afterwards, the importance of entrepreneurial self-self-efficacy as an outcome measure for entrepreneurial education programs will be discussed and important findings in the field will be discussed. Lastly, the most important Dutch entrepreneurial ‘minor’ programs, which are the research object, will be discussed in detail.
2.1. Self efficacy
The concept of self-efficacy is derived from the social learning theory proposed by Bandura (1977) and is seen as helpful construct when it comes to explaining the dynamic process of evaluation and choice that influences the development of intentions and the decision to engage in certain behavior (Boyd and Vozikis, 1994).
Self efficacy is defined as a person’s belief in his or her capability to perform a given task (Bandura, 1977). According to Bandura (1997), efficacy believesaffect the courses of action people choose to pursue, how much effort they put into tasks, how long they will persevere when confronted with obstacles and failures, their resilience to adversity, whether their thought patterns are self hindering or self-aiding, how much stress and depression they experience in coping with external demands, and the level of accomplishments they realize. As a consequence, if specific behaviour is perceived to be beyond the ability of an individual,
7
even if the person is perceiving a social demand or pressure for that kind of behaviour, he or she will not act (Boyd and Vozikis, 1994).
According to Gist (1987), self-efficacy can be gradually increased by the development of complex cognitive, social, linguistic, and physical skills that can be obtained through experience. The acquisition of experience therefore can reinforce self efficacy and, in turn, can lead to higher aspirations and future performance (Herron & Sapienza, 1992)
Beliefs about efficacy can be developed and strengthened, according to Bandura, (1982) in the following four ways :
1) Mastery experience 2) Observational learning 3) Social persuasion
4) Judgements of their own physiological states
Mastery experience refers to repeated performance accomplishments and is seen as the most influential source of efficacy beliefs (Bandura, 1997; Zimmerman, 2000). In other words, mastery experience provides a positive form of experience that can lead to more positive beliefs about future performance. But, as Bandura (1997) notes, in order to gain a stable belief about self efficacy, it is essential that one must not only experience ‘easy wins’ but acquire experience in overcoming obstacles ‘the harder way’, through effort and perseverance.
Observational learning is seen as less of an influential source on efficacy beliefs (Gist, 1987) and refers to the process when people compare themselves to others when forming
judgements about their own capabilities (Zimmerman, 2000). By observational learning, a person makes an estimation of the skills and behaviour that are required by the role model when he or she performs a task, makes an estimation to what extend those skills are similar to
8
his or her own, and decides on the amount of effort versus skill that would be required to accomplish the same task in the same way (Gist & Mitchell, 1992).
Social persuasion is, compared to mastery experience and observational learning, usually less effective in the development and strengthening of self efficacy beliefs (Bandura, 1982) and refers to discussions and feedback about one’s ability to perform a task (Gist & Mitchell, 1992). It refers to the process in which people are led, through suggestion, that they can cope with certain challenges and achieve certain results that in the past seemed to be difficult or not possible (Bandura, 1977). Mastery experiences are, as mentioned previously, usually stronger with respect to inducing self efficacy due to the fact that actual accomplishments contribute more than external persuasion or encouragement.
Lastly, judgements of their own physiological status refers to perceived fatigue, stress and other emotions that are often seen as negative and indicative of possible future physical incapability (Zimmerman, 2000). The perception of physiological status might have
informative value regarding competency of an individual (Bandura, 1977) and therefore can effect self efficacy.
The concept of self efficacy has been well studied. For example, a study of Markham, Balkin and Baron (2002) conclude that self-efficacy can accurately predict the scope of career
options considered, occupational interests and personal effectiveness. Additionally, Zimmerman (2000) found that students with high self-efficacy undertake difficult and challenging tasks more often than students reporting low levels of self-efficacy.
2.2. Entrepreneurial Self-Efficacy
Extending on self efficacy, entrepreneurial self efficacy refers to the individual’s belief about his or her competence or abilities to fulfil an entrepreneurial task. Maritz and Brown (2013)
9
define it as attaining success and controlling cognitions in order to deal with challenging goals that arise during the start up phase of a business.
Furthermore, ESE has been proposed by Krueger and Brazeal (1994) as a strong predictor of entrepreneurial intentionality. Avis (2012) adds that it can be used to predict entrepreneurial behavior, perseverance and effectiveness and according to Chen Greene and Crick (1998), ESE is a reliable measure to differentiate between business founders and non-founders. An individual with high ESE is said to be an entrepreneur who will put more effort for a greater length of time, develops better strategies and plan and shows more persistence for an
entrepreneurial task (Shane, Locke, and Collins 2003). Individuals with low ESE are more likely to avoid these challenges and are more risk averse regarding entrepreneurial
opportunities.
The process of starting a business is described numerous times as a purposive and intentional career choice (Bird, 1988; Katz and Gartner, 1988) and despite the various contextual and individual factors that are associated with entrepreneurial choice and behavior, ESE has shown to be particularly useful and important (Boyd and Vozikis 1994).
2.3. Entrepreneurial education and entrepreneurial self-efficacy
From an entrepreneurial education perspective, the concept of ESE is pivotal for a number of reasons. First of all, a highly efficacious student is confident in his or her own abilities to thrive in the process of creating a new business. Additionally, ESE has been demonstrated as particularly important in the development in entrepreneurial intentions (Barbosa, Gerhardt and Kickul, 2007). Thirdly, targeted education is argued to successfully enhance ESE and
therefore could be considered to be one method of developing ESE levels amongst individuals (Wilson et al., 2007; Florin, Karri, and Rossiter 2007; Zhao, Seibert and Hills, 2005). This is
10
supported by the results of a study of Maritz and Brown (2013) in the UK, suggesting ESE may be elevated through training and education, thus potentially improving the rate of entrepreneurial activities.
Pihie and Bagheri (2010) argue that the curriculum and learning techniques of EEPs should be designed to effectively increase levels of ESE amongst students. As mastery experiences are proven to achieve the most result in enhancing higher levels of self-efficacy (Zimmerman, 2000), the curriculum should incorporate aspects that facilitate learning by doing (Wilson et al., 2007) such as experiencing real-life business situations as to increase risk taking and innovative behaviour (Chen et al., 1998). On top of that, it is emphasized by Maritz and Brown (2013) that appropriate role models should be introduced in the EEPs to encourage observational learning and social persuasion could be enhanced by EEP mentors reinforcing students and providing them with positive feedback. Additionally, students working in teams could enhance observational learning for individuals. By doing so, Maritz and Brown (2013) argue that EEPs could increases entrepreneurial motivation and entrepreneurial performance by changing attitudes and confidence associated with entrepreneurial skills and competencies.
Although the importance of a focus on ESE when developing and assessing EEPs has been expressed by numerous scholars (Maritz and Brown, 2013; Pihie and Bagheri, 2010) , ESE as an outcome measure remains relatively untested (Wilson et al., 2007; Sanchez, 2013). In the past years however, some studies attempted to provide more insight in this possible
relationship.
Wilson et al (2007), for instance, found that among more than 5000 high school students in the US, males report higher levels of ESE than females. Furthermore they found that the same results among 933 MBA students from different universities in the US, whereas males score higher on ESE than females, suggesting that gender is a relevant factor determining levels of
11
ESE. The study of Wilson et al. (2007) argues that targeted education, an MBA, in this case, could lead to a significant increase in ESE for females, compared to more general forms of education. ESE was measured based upon a six item self assessment scale where respondents were asked to compare themselves in the following area’s in respect to ‘others in the business world’: ‘being able to solve problems’, ‘making decisions’, ‘managing money’, ‘being
creative’, ‘getting people to agree with you’ and ‘being a leader’.
The scope of the study is limited, however. Not only did it incorporate only gender and level of current education as a factor with respect to ESE, the sample did not include EEP students and levels of ESE were measured at one point in time – neglecting the effect of following an (entrepreneurial) educational program and possible biases due to the characteristics or intentions people who have chosen to follow a MBA.
Maritz and Brown (2013) took a longitudinal approach when testing 50 participants in a single UK-based EEP for ESE. ESE was measured by means of a questionnaire consisting of twenty items (part selected from previous research and partly self composed) at which the respondents could respond by means of a eleven point Likert scale ranging from 0 to 10. The point were eventually summated to calculate a sum score and topics included, for example, opportunity recognition, transforming an idea into a product, marketing, finance, management skills.
They found that at the beginning of program, males reported higher levels of ESE than females. After the program however, Maritz and Brown (2013) found that ESE levels of females were increased more significantly than ESE levels of males, suggesting the impact of an EEP on females ESE is larger than for males. They also found that participants under forty years reported higher levels of ESE before the program than people older than forty.
12
participants younger than forty. Furthermore their study found the presence of business owning relatives to be significant – participants without business owning relatives reported a larger increase in ESE after the EEP compared to participants who have business owning relatives. Ethnicity proved to be insignificant. Although the mediating effect of gender, age and business owning relatives has been suggested, the results of this study must be carefully interpreted due to the low sample size. Additionally, Maritz and Brown (2013) did not provide information about the length and content of the entrepreneurial education program, therefore limiting comparability of the study conducted by these authors.
Maritz and Brown (2013) and Wilson et al. (2007) both identified gender as a notable factor to be included when investigating ESE. This is in line with preliminary results suggesting that females have lower entrepreneurial self-efficacy (Chen et al., 1998; Chowdhury & Endres, 2005).
Age has also been found as a factor that could explain different levels of ESE after following an EEP (Wilson et al., 2007). Although both have contributed to a better understanding of ESE as an outcome measure of EEPs, more research is need in order to test the influence of gender and age. Furthermore a gap still exists with respect to other factors which could influence the ESE outcome. For instance, the incorporation of mastery experience has been proposed as an important factor, but never been tested in an EEP setting.
2.4. Entrepreneurial minor programs in the Netherlands
As part of following a bachelor program, students in the Netherlands can apply for and attend a specialized education program, called a minor. Generally speaking, a minor in
entrepreneurship in the Netherlands last for six months (e.g. one semester) during which the students can achieve 30 ECTS (European credit transfer system) credits. Entrepreneurial
13
minor programs consist of both lectures and tutorials and functions as a replacement for other, bachelor specific, courses.
Compared to other entrepreneurial education programs, a minor entrepreneurship can be seen as a more extensive program, due to its length and intensity. A common form of
entrepreneurial education are entrepreneurial workshops, which usually are seen as less extensive due to their limited time horizon (e.g. a day or a week) and therefore arguable less effective in changing perceptions about changes in entrepreneurial attitudes and confidence towards skills and competencies behavior are possible. One important characteristic of entrepreneurial minor programs in the Netherlands is their focus on academic as well as practical oriented learning goals. Wilson et al. (2007) suggest that the implementation of practical oriented learning in the curriculum could effectively enhance level of ESE amongst students, where the incorporation of learning by doing, such as experiencing real life business situations, could elevate one of the key drivers of self-efficacy, mastery experiences. Most entrepreneurial minor programs in the Netherlands offer the possibility to experience real life business situations by stimulating or requiring students to systematically identify
opportunities, write business plans and ultimately start a new venture. However, as Bandura (1997) notes, in order to gain a stable belief about self efficacy, it is essential that one must not only experience ‘easy wins’ but acquire experience in overcoming obstacles ‘the harder way’, through effort and perseverance. So in order to firmly enhance students’ levels of ESE the practical oriented learning goals should be challenging.
Another important characteristic of (most) entrepreneurial minor programs in the Netherlands, is the incorporation of mentoring. Participants in a entrepreneurial minor program are, during the practical aspects of the program, usually guided by experts in the field of
entrepreneurship. The guidance of these experts could be viewed as a possible source for enhancing students’ ESE since it could be influence observational learning and social
14
persuasion, which are proposed drivers of ESE by Bandura (1982). For instance, with respect to observational learning, students could observe and make an estimation about the skills and behavior that are required by the (entrepreneurial) expert or mentor when he or she performs a entrepreneurial task and could estimate to what extend those skills are similar to his or her own and lastly, deciding on the amount of effort versus skill that would be required to
accomplish the same task in the same way. The guidance of the experts or mentors could also be seen as a factor influencing another driver of self efficacy, social persuasion. These experts or mentors could assist in the practical oriented learning process by influencing beliefs about coping with certain challenges and achieving results, through suggestion, which in the past seemed to be difficult or impossible. According to Bandura (1982) and Gist (1987), mastery experiences are expected to be the most influential source on efficacy beliefs, followed by observational learning and social persuasion, suggesting actual accomplishments like successfully starting a new venture during the minor, could contribute most effectively to changes in beliefs about ESE. Below will be elaborated upon the two main minor programs central in this research and the curriculum will be discussed in detail.
2.4.1 Minor entrepreneurship of the University of Amsterdam (UvA)
This minor in entrepreneurship is offered by the faculty of Economics and Business (FEB) in corporation with the Amsterdam Center for Entrepreneurship (ACE). The minor lasts one semester (e.g. six months) and students can achieve a total of 30 ECTS during this period. During the minor program, a broad range of students who wish to pursue entrepreneurial ambitions, will be introduced to entrepreneurship education from both an academic, theoretical, and a practical perspective. The minor consists of three courses: ‘Cases in entrepreneurship’, ‘Entrepreneurship in practice’ and one elective course.
15
In the first course, cases in entrepreneurship, students will discuss articles, case studies and practitioner readings with a focus on topics like entrepreneurial opportunity discovery, risk and uncertainty, entrepreneurial marketing, the capture of value and motivation in
entrepreneurial organizations or small teams. This first course can be seen as mainly focused on (academic) theory with a total workload of 6 ECTS. Within the second course,
entrepreneurship in practice (18 ECTS), students will focus more on the practical experience and, consequently, helps students to find out whether a career as an entrepreneur could appeal to them. Students are expected to start a new venture in a multi disciplinary team which allows students to gain practical experience with marketing, finance, coordination, negotiating and presentation skills in a legally protected environment. Secondly, students will write business plans that are reviewed by experts. Lastly, individual students are expected to describe their personal lessons learned from the course, focusing on linking theory to practical lessons learned. The third course is elective and students can choose a variety of 12 different courses to complement their theoretical and practical entrepreneurship skills.
Although part of the program aims at building theoretical knowledge on entrepreneurship, the main focus of the program is to increase practical knowledge and experience by participating in the process of starting a new venture. Specifically, students will be trained in actively identify entrepreneurial opportunities. Teams of students are coached by experienced business coaches from consultancy firms who will guide them throughout the process.
2.4.1 Minor entrepreneurship of the Erasmus University in Rotterdam (EUR)
The minor Entrepreneurship & New Business Venturing is offered by the Erasmus University Rotterdam and the Rotterdam School of Management. The emphasis of the minor is on identifying entrepreneurial opportunities and acting upon those opportunities by transforming
16
them into a new venture. The program consists of a core module and one out of the two elective courses, entrepreneurial marketing and entrepreneurial finance.
The core module (12 ECTS) focuses on ‘getting started’, e.g. the (pre) launch stage of new ventures. Within this module, students will be familiarized with the (theoretical) phenomenon of entrepreneurship and will experience what it is like to be an entrepreneur by developing a business idea in a team of students, guided by expert entrepreneurs. Key to practical oriented learning focus is the effectual (action-based) approach to creation of new ventures.
The first elective course, entrepreneurial marketing, is about marketing challenges that are faced by new ventures like the initial position of the company and the product. Within this course, as new ventures face different marketing challenges than larger (e.g. wealthier) companies, students are taught how to focus on strategies and tactics with limited budget. Also included are topics like, online marketing, how to convince others and free advertising. The second elective course, entrepreneurial finance, is about how new ventures deal with financial obstacles and strategies. Within this course, students are familiarized with
investment decisions entrepreneurs are confronted with during different phases of the venture. Topics include sources of financing, analyzing its specific advantages and disadvantages, and basic financial analysis.
Compared to the minor in entrepreneurship offered by the University of Amsterdam, this minor is shorter and less extensive due to the fact that students can achieve a total of 15 ECTS within two months (compared to 30 ECTS and two semesters). Since there are, based on the curriculum, little differences between the minor programs offered by the UvA and the EUR, it could be argued that the latter would lead to less significant changes in the perception about the effect of the minor on ESE of the students.
17
3. Data & Method
In this chapter the target population will be described and how the data was collected. Secondly will be elaborated upon the measures and properties used for both the dependent variable and the independent variables. Lastly, the statistical tests that are used in this study will be described in detail.
3.1. Sample description
Data used in the analysis is collected by means of an online survey sent in May and June 2015 to alumni of minor programs in entrepreneurship from different universities across the
Netherlands. This survey, hosted on ‘Qualtrics’ software, was created and self-composed.
The alumni, of different universities, were approached via mail listings provided by
universities and / or social media groups, and invited to fill in this questionnaire. Specifically, an email was sent to 269 alumni of the University of Amsterdam (UvA). Furthermore, the link to the questionnaire was shared on two social media groups of the UvA consisting of 88 alumni in total and a social media group of the Amsterdam university of applied sciences (HvA) consisting of 365 alumni. On top of that, an attempt was made to contact other alumni via more indirect ways such as posting on pages specifically designed for alumni of
entrepreneurial minor programs. As it is impossible to find out how many alumni on social media actually have seen the invitation, the number of approached alumni is estimated, rather than precisely calculated, at 600. All emails sent to alumni and messages on social media were accompanied by a short description of the topic, its purpose and a link to the
18
A total of 86 alumni started the questionnaire of which 13 prematurely ended. As a result, the dropout rate is 15 percent and the total amount of alumni who finished the questionnaire is 73. The response rate is estimated on 12 percent.
Participation was voluntary and the data acquired in the survey was, as stated in the beginning of the questionnaire, handled confidentially and anonymously.
3.2. Study material
This research investigates multiple variables where a distinction should be made between dependent and independent variables. Below can be found how the different variables were measured and their properties will be discussed. The full questionnaire can be found in the first section of the appendix.
3.2.1. Dependent variable
The dependant variable, or output variable, is the “perceived change in self efficacy” that resulted from following the minor in Entrepreneurship. To measure this, participants were asked about their opinion on a series of four statements regarding how the minor
retrospectively influenced their self-efficacy. The following self-composed statements were incorporated in the survey to measure their perceived change in self-efficacy:
“The minor enhanced my chances of starting a business successfully” (Q22.2)
“The minor made me more confident about my entrepreneurial qualities” (Q22.3)
“Because of the minor, I feel more confident when it comes to starting a business” (Q22.4)
19
As suggested by Bandura (2006), the efficacy belief system is not a global trait but a differentiated set of self-beliefs linked to distinct realms of functioning. Therefore, if one wants to measure perceived self-efficacy, the items should be phrased in terms of can do rather than will do and linked to a specific and distinctive task,to accurately reflect this construct (Bandura, 2006). The statements used in the survey reflect this by phrasing it in terms of a perceived capability in a specific realm of functioning, namely entrepreneurial behaviour.
The response scale was based on a 5-item Likert scale, ranging from completely disagree, disagree, neutral, agree to completely agree and was presented in a straight (vertical) line because this is the clearest and efficient way for respondents to process their data (Saunders et al., 2009).
3.2.2. Independent variables
The independent variables being researched consist of the following variables as shown in the table below.
Variable Response scale Categories
Age Ordinal 18-20
20-22 23-25 >25
Gender Nominal Male
Female
Nationality Nominal Dutch
Non-Dutch Finished the first 60 ects of the bachelor in one
(study) year
Nominal Yes
No
The quality grade given to the minor (by the alumni)
20
The requirement or stimulation to start a business during the minor
Nominal Yes
No
Entrepreneurial experience before the minor
Year started with the minor
Faculty of bachelor University Ordinal Ordinal Nominal Nominal
Yes, started one business (or more) before the minor No < 2014 ≥2014 FEB Non-FEB University of Rotterdam (EUR). Amsterdam university of applied sciences (HvA). Free university Amsterdam (VU).
Leiden University. University of Amsterdam. (UvA).
Hotelschool the Hague. Table 1: Categories for the independent variables
Age
Both Maritz and Brown (2013) and Wilson et al. (2007) identified age as a factor that could explain different levels of ESE after following an EEP. For instance, Maritz and Brown (2013) found that participants under forty years reported higher levels of ESE before the program than people older than forty, but after the EEP, participants older than forty reported higher increases in ESE than participants younger than forty. Although the differences in age are smaller in this study, age was included as a variable since it could be a factor that
influences self perceived change in ESE as a result of following an EEP.
Gender
Maritz and Brown (2013) and Wilson et al. (2007) both identified gender as a notable factor to be included when investigating ESE. Maritz and Brown (2013), for instance, found that at
21
the beginning of an EEP in the UK, males reported higher levels of ESE than females. After the program however, Maritz and Brown (2013) found that ESE levels of females were increased more significantly than ESE levels of males, suggesting the impact of an EEP on females ESE is larger than for males. As such, gender was included as a variable in this study as past research has identified it as a factor explaining differences in ESE after following an EEP.
Nationality
Although most entrepreneurial minor programs investigated in this study are taught in English, the variable nationality was included as a independent variable in this research. It could be the case that the effect on the self perceived change in ESE as a result of the minor, for example, for non-Dutch students is lower due to cultural and or language differences. Obtaining and accessing knowledge or finding partners, for instance, could be realized by Dutch students with less effort compared to non-Dutch students. On the other hand it could be that non-Dutch students are more motivated than Dutch students. This study controlled for either a Dutch or non-Dutch nationality and by doing so, could measure whether there is a (significant) difference in terms of the dependent variable between these groups of students.
Finished the first 60 ECTS of the bachelor in one (study) year
The variable ‘finished the first 60 ECTS of the bachelor in one (study) year’, is included in this study as it could be an indication of how motivated or capable a student is since it measures whether or not the student did had any (initial) delays in the process of completing their bachelor. This study opted for choosing the first 60 ECTS (e.g. propedeuse) because the questionnaire invited students who did not participated in the minor program recently and it is
22
believed that most respondents will accurately memorize whether or not they achieved the first 60 ECTS of their bachelor without any delay.
The quality grade given to the minor (by the alumni)
This variable captures the perceived quality of the minor by the students and therefore could give an indication to whether differences in self perceived change in ESE could be explained by the (perceived) quality of a minor.
The requirement or stimulation to start a business during the minor
As previously mentioned, the minor programs offered by the UvA and EUR require students to start a new venture during the minor. However, since other minor programs could not have required or stimulated students to do so, this variable could offer insights in whether or not this has a significant effect on the dependent variable. Extending on Wilson et al. (2007) who suggest implementation of practical oriented learning in the curriculum could effectively enhance the level of ESE amongst students, it can be argued that the requirement or stimulation to start a new venture is ESE enhancing since it could affect students’ mastery experiences as it provides students with real life business situations.
Entrepreneurial experience before the minor
This variable captures whether or not the respondent had any entrepreneurial experience before the minor by measuring if the respondent started a (or more than one) business before starting the minor. Since previous business experience could lead to higher levels of perceived ESE due to mastery experiences, the self perceived change in ESE as a result of the minor could be less compared to those whom had no business experience before the minor.
23
Year started with the minor
To see whether there is a relation between the year in which the minor was started and the self perceived change in ESE due to the minor, this variable was introduced dividing respondents into two groups – started before and after 2014. It is included since respondents could be biased if the minor was not recently started. Furthermore the content, e.g. curriculum, of different minor programs could be altered over the course of years and therefore have influenced self perceived change in ESE.
Faculty of bachelor
This variable, measuring whether a respondent followed their bachelor at a financial or business faculty (FEB) or another faculty, was included to see whether there is a difference between FEB and non-FEB students with respect to the dependent variable. It could be argued that respondents who were following their bachelor at a FEB faculty have more knowledge about different aspects associated with starting a business, leading to less change in self perceived ESE as a result of the minor due to the fact that some material was covered in previous (non EEP) courses.
University
This variable was introduced to find out if there are any differences between the minor programs as offered by the universities with respect to self perceived changes in ESE. Since all programs are expected to be (slightly) different the incorporation of this variable could be useful to see if there are any differences between different programs. For example, since the program of the EUR is arguably less extensive than the program offered at the UvA, a
difference between these two universities could be expected regarding self perceived changes in ESE.
24
3.3 Statistical Analysis
To find out more about possible relationships, a multiple regression analysis is used. A multiple regression analysis is appropriate when research investigates the relationship between multiple independent variables and a single dependent variable (Hair, Anderson, Tatham & Black, 1998).
The perceived change of entrepreneurial self-efficacy as a result of following the minor functioned as the dependent variable. Because this construct was measured by four different statements in the same way, a mean score was computed in order to create one score, or value, for the dependent variable. By assigning numeric values to the different responses to a given statement, e.g. ranging from completely disagree = 1 to completely agree = 5, the output can be analysed in more detail.
To check whether the previous assumption about internal validity in these four statements is reasonable, an reliability analysis is conducted measuring the Cronbach’s alpha, which is a inter-correlation statistic frequently used in psychometrics. By doing so, one can get a grasp on the internal coherency of the different statements and see whether they are measuring the same construct.
Furthermore, the outcome on the dependent variable is tested for significance to see whether the result differs significantly from the null hypothesis (e.g. no effect). This is done by using the ‘one sample Wilcoxon signed rank test’ in SPSS, where the observed mean is subjected to a test against a theorized null hypothesis mean (of 3.00). The ‘one sample Wilcoxon signed rank test’ is a non parametric statistical hypothesis test used if normality assumptions are not
25
met when comparing two related samples. In this case, the observed mean and the theorized population mean.
The independent variables were subjected to a multiple regression analysis, exploring possible relationships with the dependent variable. Dummy variables were manually created in SPSS for the categorical variables that could take on more than two possible values. As a result, the variables age and university were subjected to this technical transformation - resulting in the creation of respectively three and five dummy variables. Age was measured on a ordinal scale consisting of four categories; 18 to 20, 20 to 22, 23 to 25 and older than 25, since, due to the characteristics of the minor, most students are expected to be in their early twenties. For confidentiality reasons, e.g. ensuring participants’ anonymity, a ordinal scale was preferred over an exact measurement approach. The other variables were included in the regression model without further processing since the variables could either take on two values or are measured on a discrete scale. The variable University was measured on a nominal scale and consists of six categories: University of Rotterdam (EUR), Amsterdam university of applied sciences (HvA), Free university Amsterdam (VU), Leiden University, University of
Amsterdam. (UvA) and the Hotelschool the Hague. Since for both age and university only, respectively three and five dummy variables are needed, the category ‘older than 25’ and “Leiden University’ were left out of the regression model.
With respect to the variable gender, males were coded as 0, whereas females were coded as 1. Furthermore, respondents without a Dutch nationality were coded as 0, and respondents with Dutch nationalities were coded as 1. Additionally, the finished the first 60 ECTS of the bachelor in one study year variable was coded in a way to assign a 0 to respondents who did not finish the first 60 ects of the bachelor in one study year, whereas respondents who did were coded as 1. The quality grade of the minor (given by the alumni) was measured on a discrete scale and therefore did not need coding. The requirement or stimulation to start a
26
business during the minor variable was treated as follows: respondents who were not required or stimulated to start a business were coded as 0, whereas respondents who were required or stimulated to do so were coded as 1. With respect to entrepreneurial experience before the minor, 0 indicated that a respondent did not start a business before the start of the minor whereas respondents who started one or more businesses before the minor were coded as 1. Year started with the minor could take on two forms, respondents who started the minor before 2014 were coded as 0 and respondents who started the minor in 2014 or afterwards were coded as 1. The faculty of bachelor variable also could take on two forms were respondents who were not following their bachelor at a business or economics faculty were coded as 0, and respondents who were following their bachelor at a business or economics faculty as 1.
3.3.1. Missing values
To extract as much reliable data as possible from the survey, one must deal in an appropriate way with missing values. Respondents could start a survey and decide to prematurely end it for many reasons. Because the survey’s design doesn’t allow progressing unless the previous question is answered, either the respondent could fill in the whole survey or decide to stop at a given moment. For the sake of reliability, this research only incorporates data provided by respondents who reached a point in the survey where all questions regarding the variables important in this research are answered.
3.3.2. Regression model
As mentioned before, a multiple regression model is used to find out whether there is a relation between the dependent variable and the independent variables. The multiple
27
regression model was analyzed by using SPSS17 software and all independent variables mentioned before will be added to see whether there is a significant relationship between the independent variables and the dependent variable. In case of signs of multi collinearity, the variables found to be problematic due to possible collinearity are excluded from the reported multiple regression results. Statistical significance is tested and assessed based on an alpha of 0.05 and will be two-sided.
28
4. Results
This chapter will provide a clear summary on the collected data. Firstly, the most important characteristics of the sample population will be discussed. Thereafter the data on the (other) independent and dependent variables will be discussed including a short note on the internal validity of the latter. Thirdly will be elaborated on the correlation and collinearity. Lastly, the outcomes of the regression model are being presented.
4.1. Sample characteristics
A total of 86 respondents participated in this survey. After excluding the respondents that did not include all answers referencing to the variables included in the regression model, a total of 73 respondents (85 percent of participants) remained which provided the data which will be of use in this research.
The table below shows the general results that give an indication about the sample population.
Category Frequency Valid percentage
Gender Male 47 64.4% Age 18-20 4 5.5% 21-22 28 38.4% 23-25 22 30.1% >25 19 26.0% Nationality Dutch 62 71.3%
University Erasmus University 11 15.1%
Hogeschool van Amsterdam 4 5.5%
Free university Amsterdam 3 4.1%
Leiden University 6 8.2%
University of Amsterdam 48 65.8%
29
Faculty FEB 23 32.9%
Year started with the minor
< 2014 33 46.5%
Finished the first 60 ECT’s in one study year Yes 27 37.0% Required or stimulated to start a business during the minor
Yes 66 90.4%
Previous entrepreneurial experience
Yes, started one business (or more) before the minor
19 26.0%
Table 4.1: General results of the sample population
Mean Median Std. Deviation N
Quality grade minor (assessed by alumni) from 1-10
7.48 8.00 1.57 73
Table 4.2: General results from the sample population
As can be deducted from the tables above, most respondents are male (64.4 percent), between 21 and 25 years old ( total of 68.5 percent), have a Dutch nationality (85.0 percent), followed the minor at the University of Amsterdam (65.8 percent) and started the minor in 2014 or later (53.5 percent). Furthermore less than half of the respondents (37 percent) finished the first 60 ECT’s of their bachelor in one study year and 26 percent had previous entrepreneurial
experience. Additionally, most respondents (91,4 percent) report that they were either stimulated or required to start a business during the minor. Lastly, we can see that the
30
different minor programs were considered to be of high quality by the alumni - a mean of 7.48 and median of 8.00 was reported.
4.2. Dependent variable
Four statements were included in the survey to measure the perceived change in
entrepreneurial self-efficacy as a result of the minor. As discussed in the previous chapter, a value of 1 indicated that the respondent strongly disagrees, and a value of 5 indicated the respondent strongly agreeing with the statement provided.
To find out whether the internal consistency between the four statements regarding the dependent variable, is sufficient, a reliability test was conducted. A Cronbach’s alpha of 0.92 was found, indicating the most common recommendation for a sufficient Cronbach’s alpha of higher than 0.70 is met. The Cronbach’s alpha found regarding these four statements indicate that 92 percent of the variability, when combining the four statements, can be considered as true score variance. The highest value for internal consistency is reached by combining the four statements - excluding one statement would result in a lower Cronbach’s alpha.
To transform the values of these four statements into one variable, a new variable, measuring the mean scores of these statements, was created. The following table and histogram show the result of this computation, which will be treated as the dependent variable in this research.
Mean Median Std. Deviation N
Perceived change in entrepreneurial self efficacy
3.70 4.00 0.94 73
Table 6: Results new variable measuring the mean score reporting a self-perceived increase in ESE as result of following the minor
31
Figure 1: Histogram of the dependent variable.
To see whether or not the mean is significantly different from the null hypothesis (e.g. no effect) a ‘one sample Wilcoxon signed rank test’ was performed. This is a appropriate test due to the fact that both the kurtosis (1.064) and the skewness (- 1.068) are more than one away from zero, which violates normality assumptions. The null hypothesis (mean equals 3) is rejected at a significance level of lower than 0.001, implying that the reported mean score of the dependent variable differs significantly of the null hypothesis, suggesting following a minor is associated with a increase in the self perceived ESE.
32
4.3. Correlation and multi collinearity
The table showing the correlations between all relevant variables will be shown on the next page. After conducting a collinearity test in SPSS, the dummy variable ‘University of Amsterdam’ and ‘University of Rotterdam’ are found to likely suffer from multi collinearity as indicated by Variance Inflation Factor (VIF) scores of respectively 5,916 and 3.585. Because multi collinearity can result in incorrect conclusions about the relationship between the dependent and independent variables, all the ‘university’ dummy variables will be excluded from the multiple regression model. After removing the ‘university’ dummy variables, preparatory testing revealed no indication of multi collinearity.
33
Table 7: Correlation matrix of independent variables * indicates p < 0.05 ** indicates p < 0.01
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 1. Gender 1 .072 .119 -.114 .153 -.096 .007 -.153 -.053 -.154 .115 -.088 -.054 -.045 .145 .246* -.150 2. Age 18 to 20 1 -.190 -.158 -.067 -.060 .002 -.101 -.058 -.050 -.080 -.028 .228 -.267 * .078 .132 -.148 3. Age 21 to 22 1 -.518** -.140 -.196 -.075 .376** .058 -.163 -.262* -.093 .521** -.080 .066 -.211 -.047 4. Age 23 to 25 1 .110 .177 -.015 -.193 -.027 .014 .222 -.077 -.262* .123 .011 .087 .076 5. Nationality 1 .164 -.116 -.358** .101 .087 .261* .050 -.243* -.189 -.137 -.274* -.156 6. Finished the first 60 ECT's of the bachelor within one academic year 1 .304* -.085 .065 -.159 .134 -.090 -.171 .201 .153 -.002 .262* 7. Faculty 1 -.051 .090 -.148 .204 -.084 -.073 .144 .132 .171 .007 8. Erasmus Rotterdam 1 -.101 -.087 -.584 ** -.050 .243* .091 -.123 .099 .003 9. Hogeschool van Amsterdam 1 -.050 -.334 ** -.028 -.017 .080 .078 -.143 .126 10. Free university Amsterdam 1 -.287 * -.024 -.183 .069 -.636 ** .034 -.044 11. University of Amsterdam 1 -.163 -.248* .111 .451** .033 .184 12. Hotelschool the hague 1 -.128 -.112 -.362 ** -.070 -.214 13. Year minor started 1 -.137 .123 .154 .008 14. Quality grade of the minor 1 .100 .238 * .727** 15. The requirement or stimulation to start a business during the minor 1 .087 .243* 16. Previous business experience before the minor 1 .216 17. Entrepreneurial Self efficacy 1
34
4.4. Regression
The multiple regression model, as described in the method section, has been found to lack predicting power due to multi collinearity. As a result, the outcomes of the multiple regression model reported in this study did not incorporate university as an independent variable.
After having addressed the encountered multi collinearity, the model reports a R square of 0.628, implying that 62.8% of the variation in the “Perceived change in entrepreneurial self-efficacy” can be explained by the independent variables. This result can be interpreted as if there is a strong relationship between the dependent and independent variables used in this model. Since the reported significance of the regression model (p < 0.001) is lower than the alpha used in this study (0.05), the model can be seen as significant.
The table below summarizes the coefficients and significance of the independent variables.
35 Unstandardized Coefficients Standardized Coefficients T Sig. B Std. Error Beta (Constant) 0.159 0.634 0.250 0.803 Gender -0.317 0.184 -0.159 -1.727 0.090 Age 18t to 20 0.235 0.478 0.051 0.492 0.625 21 to 22 -0.013 0.273 -0.007 -0.047 0.963 23 to 25 -0.051 0.221 -0.024 -0.229 0.820 Nationality 0.079 0.245 0.030 0.332 0.749
Finished the first 60 ECT’s in one study year -0.242 0.189 -0.121 -1.279 0.206
Faculty -0.342 0.182 -0.169 -1.878 0.066
Year minor started 0.140 0.227 0.073 0.616 0.541
The quality grade given to the minor (by the alumni)
0.443 0.057 0.735 7.784 0.000 **
The requirement or stimulation to start a business during the minor
0.514 0.289 0.153 1.779 0.081
Entrepreneurial Experience before the minor 0.075 0.222 0.033 0.336 0.738
Table 10: Coefficients table indicating significant positive relationship between quality grade given to the minor and the dependent variable. * indicates p < 0.05 ** indicates p < 0.01
Table 10 reports the coefficients and the significance associated with the proposed multiple regression model. The variable quality grade of the minor reports a highly significant ( p < 0.001 ) relationship (β = 0.443) with the dependent variable. Gender, Faculty and the requirement or stimulation to start a business during the minor are not found to be significant, however the p-value (respectively, 0.090, 0.066 and 0.081) are close to 0.05, suggesting marginal significance. The variables Age, Nationality, Finished the first 60 ECT in one (study) year, Year minor started and Entrepreneurial experience before the minor all are not significant because the reported p-values are (much) higher than the alpha.
36
5. Discussion
5.1. Summary of results
First of all, this study found that 73 alumni of entrepreneurial minors reported a self-perceived change in ESE mean score of 3.70, suggesting that the alumni perceived the minor to have been ESE enhancing. The reported mean score was found to be significant, tested against the null hypothesis (3.00).
Additionally, this study shows that the variable quality grade of the minor by the alumni reports a highly significant (p < 0.001) relationship (β = 0.443) with the dependent variable, suggesting that the quality grade alumni assessed to the minor by the alumni is significantly associated with the dependent variable.
Gender, Faculty and The requirement or stimulation to start a business during the minor were not found to be significant. However the reported p-value (respectively, 0.090, 0.066 and 0.081) are close to 0.05, suggesting marginal significance.
The variables Age, Nationality, Finished the first 60 ECT of the bachelor in one (study) year, Year minor started and Entrepreneurial experience before the minor all are not significant because the reported p-values are (much) higher than the alpha.
Consequently, the pivotal question that has been formulated in this study: “What factors influence the self perceived ESE in entrepreneurial education programs?” can be answered. According to the results, as derived from the survey, the quality grade of the minor
significantly influences the self perceived ESE in EEPs. Furthermore, Gender, Faculty and the requirement or stimulation to start a business during the minor are found to influence self perceived ESE marginally.
37
5.2. Interpretation
The reported mean score of 3.70 on self perceived change in ESE due to the minor is consistent with findings (Maritz and Brown, 2013) suggesting that following targeted education can be considered as a method of developing ESE levels amongst individuals.
The variable quality grade of the minor reported a positive effect size (β = 0.443) implying that, according to the proposed multiple regression model, alumni who grade the quality of the minor one extra point are estimated to report an increase of 0.443 in self-perceived change in ESE as a result of the minor. Since the quality grade of the minor seems to be positively associated with the dependent variable, one could argue that, according to students, ESE is, according to the alumni who participated, an important component of EEP. This result could provide some indirect support to arguments made byPihie and Bagheri (2010)suggesting that the curriculum and learning techniques of EEPs should be designed to effectively increase levels of ESE amongst students since a high perceived increase in ESE due to the minor is associated with a higher quality grade given by the alumni. However, one must be careful interpreting this result. Since these results do not provide evidence for causality, it cannot be interpreted as such.
Gender, Faculty and The requirement or stimulation to start a business during the minor were found to be marginally significant. Although not highly significant, they provide some useful insights. The effect size of Gender (β = - 0.317) found is in line with earlier findings (Chen et al., 1998; Chowdhury & Endres, 2005) indicating that females generally report lower levels of ESE, but contradictory to findings of change in ESE after following an EEP (Maritz and Brown, 2013; Wilson et al., 2007). Of course, this contradiction could be due to the
38
ESE for females, from EEPs being researched by Maritz and Brown (2013) and Wilson et al. (2007).
The effect size of Faculty (β= - 0.342) suggests that the perceived change in ESE due to the minor could depend on previous education. Participants who followed a bachelor program at a economics or business faculty are likely to have reported lower change in ESE due to the minor than did participants who did not followed their bachelor at a economics or business faculty. This possible relationship could be due to the fact that students from a business or economics faculty are already familiar with some aspects of the minor (for example,
accounting or finance) than students from other faculties, therefore limiting or decreasing the potential effect of the minor on ESE.
Results regarding The requirement or stimulation to start a business during the minor show a marginally significant effect size (β = 0.514) suggesting that alumni who were required or stimulated a business during the minor are likely to have reported higher perceived change in ESE than those who were not required or stimulated to do so. This could be explained by self-efficacy theory indicating that levels of self self-efficacy can be enhanced by mastery experiences (Zimmerman, 2000) which refers to repeated performance accomplishments (Bandura, 1997). The requirement or stimulation to start a business during the minor can be seen as a form of mastery experience as it provides a opportunity for students to get familiar with the process of starting a new venture, thereby potentially enhancing levels of ESE amongst alumni.
Interestingly, this has never been tested in an EEP setting and the result derived from this study is promising. Providing opportunities for students to experience in real life what it is like to start a new venture could, extending on self-efficacy theory by Bandura (1982), increase levels of ESE. It must be noted however that these results should be interpreted with caution since the variable university, due to multi collinearity, was left out of the regression model. Therefore it could be that EEPs where starting a business was required or stimulated,
39
were, generally speaking, considered to more ESE enhancing, than others. Additionally, due to the fact that less than 10 percent of the respondents were not required or stimulated to start a business during the minor, this result should be interpreted with caution.
Age has not been found to be significantly influencing change in ESE due to the minor which is contradictory to findings of Wilson et al. (2007). However, Wilson et al. (2007) made a distinction between participants younger than forty and older than forty while this study involved participants who were much younger (68.5 percent was between 21 and 25 years). Although it has not been found significant in this study, this does not imply that age is not a important factor in ESE. It could be that the age differences in this study were too small to significantly influence ESE.
Furthermore, Nationality, Finished the first 60 ECT of the bachelor in one (study) year, Year minor started and Entrepreneurial experience before the minor have not found to be a significant factor, suggesting that according to this study, they do not influence self perceived ESE as a result of following different entrepreneurial minor programs in the Netherlands.
5.3. Contribution, limitations and further research
The investigation into ESE and EEPs identified a gap in the research with few studies focusing on ESE as a outcome measure of EEPs. Consequently, this paper attempts to contribute to a better understanding of the effects of EEP on ESE by exploring the
relationship between perceived change in ESE and different university EEPs and examining which factors enhance this (possible) relationship. This study contributes to scientific
literature by supporting findings from existing literature on ESE and EEPs by indicating that alumni of the EEPs studied report a positive perceived change in ESE due to following EEPs, suggesting targeted education is a method of developing ESE. Furthermore it identified several factors as either significant, marginally significant or insignificant that could be used
40
by policymakers when considering adjusting different EEP programs to enhance ESE for different groups of students. For instance, previous business experience was not found to be a significant factor influencing the self perceived ESE as a result of following the minor
suggesting following a EEP could have the same effect on self perceived ESE for those who already started a business before the minor compared to students who did not.
One of the main limitations of this research is the limited generalisability due to the ESE measurement method in this study. First of all, the measurement of ESE is not validated from literature but self composed, limiting the comparability of the reported outcome on ESE. Additionally, there was no measurement of levels of ESE before and after the program. Due to time constraints this study opted for measuring whether or not alumni perceived a change in ESE. Furthermore, ESE was measured in retrospect, which could have had a effect on the reported levels of ESE since respondents could have been biased. Although this study, by measuring in which the year the respondent started, incorporated a control variable for this possible bias, the effect of this control variable should be interpreted cautiously due to the categorical, and therefore arguably less accurate, nature of the variable.
Secondly, the sample size is relatively low (n=73), which limits the statistical power of the reported results. For instance, only 7 respondents were not required or stimulated to start a business during the minor, which questions the statistical power with respect to this variable. Another important limitation is the low response rate (estimated on 12%), which, combined with the low sample size, could suggest the presence of a response bias. The survey could, for example, have attracted mainly enthusiastic alumni or mainly alumni who were disappointed with the content and execution of the curriculum, which could indicate the results are not representative of the population (e.g. alumni of entrepreneurial minor programs in the
Netherlands). Furthermore, for practical purposes, a non probability method of sampling was used and therefore the final results could not be viewed as representative of the relevant
41
populations. However, since information was treated anonymously, it is believed to have minimal impact.
Lastly, as mentioned before, due to multi collinearity, the variable university was excluded from the regression and age was measured categorically, limiting the accuracy of this variable. As such, interpretations of the results should be done cautiously.
Future research is needed to get a better understanding of ESE in EEP context. Although Pihie and Bagheri (2010), amongst others, emphasize that entrepreneurial education programs should be designed in order to maximize the entrepreneurial self-efficacy, ESE as an outcome measure in EEP context has not been fully grasped by studies (Wilson et al., 2007). As such, future research is needed to test the relationship between ESE as an outcome of EEPs. To develop ESE as a field however, more research is needed on the measurement of ESE since different studies use different items (Wilson et al., 2009). Although these different approaches could initially help to gain a better understanding of ESE as a phenomenon, consensus on how to measure ESE is eventually need to deal with the limited comparability of contemporary (and future) studies. Furthermore, more research is needed on different factors influencing ESE as an outcome. Additionally, this study encourages a longitudinal research approach with measurements of ESE before attending the EEP and afterwards. Lastly, studies in other countries need to be conducted and other EEPs need to be researched to pave the way for a better understanding of entrepreneurial education in the context of ESE.
42
6. References
Avis, J. 2012. “The Ambiguities of Learning in the Knowledge Economy: Transformation, Innovation and Capital.” Journal of Vocational Education & Training 64 (1): 119–125. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84, 191-215.
Bandura. A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 17(2). 122-147.
Bandura, A. (1997).Self-efficacy: The exercise of control. New York: W. H. Freeman and Company.
Chowdhury, S. & Endres, M. (2005). Gender difference and the formation of entrepreneurial self-efficacy. Presented at the United States Association of Small Business (USASBE)Annual Conference, IndianWells, CA.
Dickson, P.H., Solomon, G.T., Weaver, K.M. (2008) Entrepreneurial selection and success: Does education matter? Journal of Small Business and Enterprise Development, 15 (2) (2008), pp. 239–258
Erikson, T. (2003). “Towards a Taxonomy of Entrepreneurial Learning Experiences among Potential Entrepreneurs,” Journal of Small Business and Enterprise Development 10(1), 106– 112.
Fleming, P. (1994), “The role of structured interventions in shaping graduate
entrepreneurship”, Irish Business and Administrative Research, Vol. 15, pp. 146-57. Florin, J., R. Karri, and N. Rossiter (2007). “Fostering Entrepreneurial Drive in Business Education: An Attitudinal Approach,” Journal of Management Education 31, 17–42.
Gist, M. E.. & Mitchell. T. R. (1992). Self-efficacy: A theoretical analysis of its determinants and malleability, Academy of Management Review. 17(2), 183-211.
Hair, J.F., Jr., Anderson, R.E., Tatham, R.L. & Black, W.C. (1998). Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall.
Herron. L., & Sapienza, H. J. (1992). The entrepreneur and the initiation of new venture launch activities Entrepreneurship Theory and Practice, 17(1). 49-55.
Honig, B. (2004), “Entrepreneurship education: Toward a model of contingency-based business planning”, Academy of Management Learning and Education, Vol. 3 No. 3, pp. 258-73.
Hytti, U. and Kuopusja¨rvi, P. (2004), “Evaluating and measuring entrepreneurship and enterprise education: methods, tolls and practices”, available at: www.entreva.net
