Int. J. Entrepreneurship and Small Business, Vol. X, No. Y, xxxx 1


Copyright © 200x Inderscience Enterprises Ltd.













Entrepreneurship education and the propensity for
business creation: testing a structural model
Ricardo G. Rodrigues*, Mário Raposo,
João Ferreira and Arminda do Paço
Department of Business and Economics
University of Beira Interior
Estrada do Sineiro, 6200-209 Covilhã, Portugal
E-mail: rgrodrigues@ubi.pt
E-mail: mraposo@ubi.pt
E-mail: jjmf@ubi.pt
E-mail: apaco@ubi.pt
*Corresponding author
Abstract: This study aims to identify the factors that contribute the most to the
intention to start up a business. The research also aims to identify the profile of
student who is a potential entrepreneur concerning several aspects: personal
attributes, family, demographic variables and motivations.
Based on a sample of university students, a structural model was tested.
Research findings include the idea that entrepreneurship education is the most
relevant factor in the propensity for business creation. On the other hand,
personal characteristics have an important role in shaping the motivation to
start up a business and perceived hurdles have a negative impact on the
intention to start one up.
Keywords: entrepreneurship; education; universities; start-ups; structural
equation modelling; SEM.
Reference to this paper should be made as follows: Rodrigues, R.G.,
Raposo, M., Ferreira, J. and do Paço, A. (xxxx) ‘Entrepreneurship education
and the propensity for business creation: testing a structural model’, Int. J.
Entrepreneurship and Small Business, Vol. X, No. Y, pp.000–000.
Biographical notes: Ricardo G. Rodrigues is a Professor at the University of
Beira Interior, Portugal. His research interests are entrepreneurial marketing,
entrepreneurship education and market orientation.
Mário Raposo is a Full Professor at the University of Beira Interior, Portugal.
He is the Vice-rector and Director of the Regional Studies Center and the
Office of Technology Transference. His research interests are marketing,
strategy, and competitiveness and entrepreneurship.
João Ferreira is a Professor at the University of Beira Interior, Portugal. His
research interests are strategy and entrepreneurship.
Arminda do Paço is a Professor at the University of Beira Interior, Portugal.
Her research interests are entrepreneurship education, public and nonprofit
marketing, and social marketing.

2 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















1 Introduction
Most economies realise that ad hoc efforts to expose the youth to entrepreneurship will
not be sufficient to build a strong entrepreneurial culture. Entrepreneurship education
must be integrated in the school curriculum at all levels from primary school through
to university (Lundström and Stevenson, 2001). According to Veciana et al. (2002),
institutional theory seems to have drawn the attention to the institutional or contextual
factors (cultural, social, political and economic) as determinants of entrepreneurship.
Among the formal environmental variables, the legal rules, government support measures
and procedures have widely been described as critical in the decision to start a firm
(Stephen et al., 2005).
The theory that the entrepreneur is a result of heredity or that entrepreneurship is an
innate characteristic of some individuals no longer seems to have many followers in the
scientific literature. Li (2006) argues that the theory of planned behaviour is very useful
and it provides a sound theoretical framework towards understanding the antecedents of
entrepreneurial intentions. But the stream that defends that it is possible to learn to be an
entrepreneur seems to be gaining strength.
An entrepreneur is usually identified as an individual who knows how to
identify business opportunities and to define contexts for an innovative job. However,
our educational system is structured in such a way as to emphasise the domain of
analytical subjects.
The development of entrepreneurial talent is important to sustain competitive
advantages in an economy that is driven by innovation. The role of education and training
in entrepreneurship, in the identification of the endowment of entrepreneurial potential at
a young age, has become evident for students, politicians and educators (Rasheed, 2000).
The studies that have been developed have demonstrated that educational preparation
can contribute to the increase in the number of entrepreneurs through the development of
a positive perception concerning the need for and the practicability of entrepreneurship,
as well as the development of competences that facilitate the entrepreneurial process.
Entrepreneurship education is rooted in a theory of solid learning – that it will
develop entrepreneurs through the increase in knowledge of businesses and the
promotion of the psychological attributes associated with entrepreneurs, such as
self-trust, self-esteem and capability for concretisation (Kourilsky and Walstad, 1998).
The effects of entrepreneurship education and the creation of businesses can reflect
at both the level of cognitive development and the level of the youngest psychological
development. Empirical evidence exists concerning the impact of this education type on
adult life; in other words, this tool can influence adults’ attitudes and direct them towards
entrepreneurship (Hansemark, 1998; Hatten and Ruhland, 1995).
The present research aims to identify the factors that explain and influence the
propensity for the creation of own business. In this sense, a conceptual model is proposed
which is evaluated according to models of structural equations.

Please clarify
the highlighted
text.

Entrepreneurship education and the propensity for business creation 3


















2 Literature review
The belief that entrepreneurs have distinctive psychological characteristics has a
long tradition in entrepreneurship research (Gartner, 1988). Several studies focused
on personality dimensions that could relate to entrepreneurial behaviour through the
formation of entrepreneurial intentions (Kennedy et al., 2003; Brice, 2004; Liñán-Alcalde
and Rodríguez-Cohard, 2004; Barahona and Escudero, 2005; Asián, 2005; Li, 2006).
There is a more or less consensus in the literature regarding the study of the
entrepreneurs’ profile. According to Ferreira (2003) and Rodrigues (2004), this approach
involves aspects such as personal characteristics, risk propensity, need of achievement,
self-control, self-confidence and optimism, profit motivation and personal values.
Concerning personal characteristics, more specifically the demographic
characteristics, these seem neither to influence nor to predict the predisposition for
entrepreneurship (Robinson et al., 1991), as opposed to the psychological attributes.
However, according to Ferreira (2003), age seems to be negatively related to innovation
and growth orientation (Ferreira, 2003). On the other hand, when analysing the
entrepreneurship indicators, Davidsson (1989) found a positive relationship between
factors of educational formation, previous experience and growth aspirations, despite
having also found entrepreneurs with a low level of education.
As to risk propensity, Davidsson (1989) states that most empirical studies confirm the
theory that entrepreneurs (of small firms) do not have favourable attitudes related to risk,
nor are they considered as risk takers.
Several psychological characteristics have been suggested as being good predictors
of entrepreneurial behaviour. According to Kourilsky (1980), the more relevant
psychological attributes are:
• the need for self-achievement
• creativity and initiative
• the propensity for risk
• self-confidence and being the locus of control
• independence and autonomy
• motivation, energy and commitment
• persistence.
Gorman (1997) adds values, attitudes and personal objectives. Robinson et al. (1991)
refer to self-esteem and innovation as being more relevant than the need for self-
achievement. Davidsson (1989) found some evidence of a relationship between the need
for achievement and the individual entrepreneurial behaviour. However, Davidsson and
Wiklund (1999) concluded that the need for achievement does not constitute a relevant
explanation for entrepreneurial behaviour, since the concept lacks precision and therefore
is difficult to measure.
In relation to locus of control, Brockhaus (1980) realised a longitudinal study and
found evidence of a positive correlation between internal orientation and entrepreneurial
success. High self-confidence has been pointed out, in several studies, as a typical trait of

4 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















entrepreneurs. Optimism seems to be a characteristic present in the start-up phase of
businesses (Davidsson, 1989). Optimism can facilitate action but it does not necessarily
lead to positive results.
Through the psychological characteristics related to motivations, it is possible to
corroborate that these are important for the determination of the direction, persistence and
intensity of the action and performance (Davidsson and Wiklund, 1999). The processes of
choosing tasks to be carried out, and assigning them time and energy, depend particularly
on the degree of motivation that the individual possesses.
Empirical investigation indicates that, usually, entrepreneurs consider the economic
motivation less important than pursuing other objectives (Baumol, 1983). However, in
the first phases of development of a business, the profit motivation can completely drive
the behaviour, while other objectives can be sacrificed (Davidsson, 1989). Autonomy
is another characteristic pointed to as relevant to the propensity for businesses creation.
The reasons for the need for autonomy can range from the importance attributed to
independence, to the frustration resulting from previous experiences, the low level of
requested support, the rigid style, and the desire to do a pleasing job (Brockhaus, 1980).
Some investigations have come to support the idea that the psychological attributes
related to entrepreneurship can be culturally acquired (Vesper, 1990). To this extent, it
also seems pertinent to conduct an analysis concerning the contribution of education to
the fomentation of entrepreneurship. The entrepreneurship education based on a theory of
solid learning can contribute to increase knowledge management and to promote the
psychological attributes associated with entrepreneurs.
According to Hansemark (1998), entrepreneurship education has been defined as
an education that intends to create a new product or service that can result in a high
economic value. Entrepreneurship education is also focused on the knowledge and study
of the creation of small business or on the creation of own employment, as well as on the
entrepreneurial capabilities and attributes.
There is some empirical evidence that a formal entrepreneurship education affects
the attitudes of secondary school students, influencing them in the direction of their
future career, and affects their propensity for entrepreneurship when they become adults.
Kourilsky and Walstad (1998) indicate that a very early stimulus of entrepreneurial
attitudes, even before high school, can encourage entrepreneurship as a career option,
although they have not tested this assertion empirically.
In general, the results of the previous researches suggest that the exposure to certain
education types in the area of entrepreneurship can contribute to build the intention to
start a business. In particular, practical programmes that provide a real experience
seem to be particularly useful to increase the desire to make a reality the creation of a
business (Honig, 2004). In this sense, at the education level, active experimentation
should be balanced with abstract conceptualisation, contributing to infuse in the students
a deeper propensity for entrepreneurship. Entrepreneurship education can include
behavioural simulations and study areas such as negotiation, leadership, creative thought,
technological innovation, the development of new products, the detection and exploration
of new business opportunities, and long-term business planning, among others
(McMullan and Long, 1987; Vesper and McMullan, 1988; Stumpf et al., 1991).

Entrepreneurship education and the propensity for business creation 5


















3 Proposed conceptual model
Usually the division of a model involves a measurement model and a structural model
(Anderson and Gerbing, 1988). The measurement model refers to the indicators and/or
subconstructs that reflect the relevant constructs, while the structural model addresses the
relationships between constructs.
The proposed conceptual model (Figure 1) considers a group of variables that
are likely to influence the propensity for business creation. It is made up of various
constructs, each one measured by several indicators. In this context, the model includes
the relationships between the following constructs:
• Personal Attributes
• Family (existence of entrepreneurs in the family)
• Demographics
• Field of Training
• Education
• Obstacles
• Motivation
• Propensity to Start Up a Firm.
Figure 1 Proposed conceptual model
PERSONAL
ATTRIBUTES
FAMILY
DEMOGRAPHICS
FIELD OF
TRAINING
OBSTACLES
EDUCATION
MOTIVATION
PROPENSITY TO
START UP A
FIRM

6 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















Each construct is composed of a group of indicators. The construct Personal Attributes
has 15 indicators; the construct Family has 1 indicator; the construct Demographics has
3 indicators; the construct Field of Training has 1 indicator; the construct Education has
5 indicators; the construct Obstacles has 15 indicators. In turn, the construct Motivation
has 13 indicators and 4 subconstructs. Finally, the construct Propensity to Start Up a Firm
has 6 indicators.
Due to the fact that the propensity for business creation is a not directly observable
variable, it paves the way for an analysis based on structural equations. This modelling
technique allows incorporating variables that are not directly observable (latent variables
or constructs) into the models. The constructs may be measured by indicators (observable
variables whose values reflect the value of the construct) or even by subconstructs.
4 Methodology
In order to test the proposed model, a sample was obtained from a population of
students who were attending a course (first degree) at the University of Beira Interior
(UBI) in Portugal at the time of data collection. An effort was made to target mainly
students in their last years. They were directly approached by the interviewers, who
visited several classrooms throughout the university. The sample was composed of a total
of 316 UBI students.
For this study, the method of data collection was the survey by self-administered
questionnaire.
After collection, data was statistically analysed and interpreted using the statistical
software SPSS. The Partial Least Squares (PLS) technique was also used to test the
proposed model. The PLS method consists of a statistical modelling-based technique
through structural equations that allows for the simultaneous estimation of a group of
equations by measuring the concepts (measurement model) and the relationships between
them (structural model).I It has the capacity to address concepts that are not directly
observable. Thus, PLS aims to maximise the variance explained for the indicators and
latent variables, making it possible to examine the relations and the R-squared (R2). A
series of iterative factorial analyses was performed through the Ordinal Least Squares
(OLS) estimation technique, combining linear and multiple regression for path analyses
(Duarte, 2005).
Table 1 shows the main methodological aspects related to the investigation.
Table 1 Synthesis of methodological aspects
Time basis Cross-section
Sampling unit Undergraduate students
Population 1417 individuals
Sample 316 individuals
Response rate 100%
Sample error 5,51%
Research method Self-administered questionnaire
Time period March 2005
Statistical analysis Bivariate, multivariate – PLS

Entrepreneurship education and the propensity for business creation 7


















5 Results
The sample was composed of 316 individuals who were attending acourse at UBI at the
time. After the questionnaires were completed, it was possible to obtain a sample with the
demographic and geographic characteristics shown in Table 2.
Table 2 Characterisation of the sample
Sex Percentage (%) Place of birth Percentage (%)
Female 51.9 Norte 43.7
Male 47.5 Centro 25.6
No answer 0.6 Lisboa e Vale do Tejo 9.8
Age Percentage (%) Alentejo 3.5
Under 20 24.4 Algarve 5.7
21–25 63.0 Açores 0.6
26–30 10.8 Madeira 0.3
31–35 0.3 Rest of Europe 5.7
Above 36 0.9 USA 1.6
No answer 0.6 Africa 3.2
Scientific area of the course Percentage (%) No answer 0.3
Economic and business sciences 32.3 Region Percentage (%)
Other social and human sciences 15.5 Beira Interior 38.0
Engineering sciences 36.1 Rest of the country 61.7
Health sciences 0.6 No answer 0.3
Exact sciences 7.3
Others 4.7
Arts and humanities 1.9
No answer 1.6

A PLS model is composed of two parts: the outer model and the inner model. The outer
model describes the relationships between the observable variables or indicators and the
latent variables or constructs. The inner model shows the direct and indirect relationships
among the constructs (Chin, 1998).
In the early stages, the estimation of the model is performed by computing the latent
variables through an iterative procedure that requires the regression of the variables of
the outer and inner models, with the parameters of one part of the model being fixed
while estimating those of the other part. In a later stage, the relationships of the outer
and inner models are estimated through OLS noniterative regression. The quality of the
model is determined by the observation of the R2, or by the Stone-Geisser test, and by the
significance of the structural relationships using the Jackknife and Bootstrap techniques
(Chin, 1998).

8 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















The measurement model is composed of 59 indicators which measure 8 constructs.
The constructs may be measured by reflective indicators (measuring the same construct
and representing the construct’s visible part or its symptoms) and/or formative indicators
(with the ability to bring about or to create the nonobserved theoretical construct) (see
Duarte, 2005). In our model all the indicators are of a formative nature.
According to Duarte (2005), with respect to formative indicators, the analysis process
includes the examination of the weights, multicollinearity analysis and the statistical
significance of weights.
When analysing constructs with formative indicators, the examination should focus
on the weights of each indicator to form the construct. According to Chin (1998), the
loadings analysis may be misleading, since the correlations among the indicators in the
same construct are not considered in the estimation process. Thus, the analysis should
entail the weights on the PLS output and the respective significance statistics, calculated
by means of a bootstrap process (Chin, 1998).
To evaluate multicollinearity, an evaluation of both the tolerance value and the
Variance Inflation Factor (VIF) is performed. These measures give us the degree to
which each independent variable is explained by the others (Hair et al., 1998). Table 3
gathers these statistics.
Table 3 Collinearity statistics
Construct Indicator Tolerance VIF
ATRIB1 0.571 1.751
ATRIB2 0.519 1.928
ATRIB3 0.601 1.665
ATRIB4 0.562 1.781
ATRIB5 0.563 1.777
ATRIB6 0.659 1.517
ATRIB7 0.708 1.412
ATRIB8 0.545 1.836
ATRIB9 0.611 1.636
ATRIB10 0.624 1.602
ATRIB11 0.700 1.428
ATRIB12 0.694 1.440
ATRIB13 0.538 1.858
ATRIB14 0.621 1.611
Personal Attributes
ATRIB15 0.692 1.445
Family FAMILY 1.000 1.000
SEX 0.990 1.010
NACION 0.924 1.082
Demographics
REGION 0.926 1.080
Field of Training MAJOR 1.000 1.000
PREP1 0.324 3.088
PREP2 0.294 3.403
POSSIB1 0.726 1.378
DIFICUL 0.995 1.005
Education
COURSE 0.925 1.081

Entrepreneurship education and the propensity for business creation 9


















Table 3 Collinearity statistics (continued)
Construct Indicator Tolerance VIF
OBST1 0.533 1.876
OBST2 0.531 1.884
OBST3 0.567 1.764
OBST4 0.698 1.433
OBST5 0.804 1.243
OBST6 0.529 1.890
OBST7 0.520 1.922
OBST8 0.622 1.607
OBST9 0.634 1.577
OBST10 0.673 1.487
OBST11 0.564 1.772
OBST12 0.514 1.945
OBST13 0.571 1.751
OBST14 0.561 1.782
Obstacles
OBST15 0.658 1.520
MOTIV1 0.589 1.699
MOTIV2 0.634 1.578
MOTIV3 0.659 1.517
MOTIV4 0.529 1.890
MOTIV5 0.506 1.974
MOTIV6 0.502 1.990
MOTIV7 0.521 1.918
MOTIV8 0.441 2.270
MOTIV9 0.714 1.400
MOTIV10 0.631 1.586
MOTIV11 0.732 1.366
MOTIV12 0.829 1.206
Motivation
VAL 0.936 1.068
WISH1 0.562 1.779
WISH2 0.486 2.056
EMPRE1 0.348 2.871
EMPRE2 0.275 3.630
CONSIDER 0.641 1.560
Propensity to Start Up a Firm
POSSIB2 0.537 1.860
As can be seen in Table 3, the indicators have no multicollinearity problems, as there are
no tolerance values close to zero and the VIF values are close to 1 (VIF = 1 means the
absence of multicollinearity).
To test the significance of the weights, we used the bootstrapping technique, which
consists in generating a large number of subsamples from the original sample through the
systematic deletion of observations. The model is recomputed for each subsample and
the resulting weights are averaged. The resulting mean of weights is compared with the
original weight. In this case 100 valid subsamples were extracted (see Table 4).

10 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















Table 4 Statistical significance of the weights
Estimate
Indicator Initial sample Bootstrap
Standard
deviation t Sig.
Personal Attributes
ATRIB1 0.112 0.113 0.011 10.586 0.000
ATRIB2 0.118 0.119 0.013 9.182 0.000
ATRIB3 0.119 0.119 0.014 8.483 0.000
ATRIB4 0.142 0.141 0.014 10.212 0.000
ATRIB5 0.126 0.127 0.015 8.468 0.000
ATRIB6 0.096 0.096 0.016 6.056 0.000
ATRIB7 0.099 0.099 0.018 5.629 0.000
ATRIB8 0.088 0.089 0.013 6.861 0.000
ATRIB9 0.091 0.088 0.016 5.550 0.000
ATRIB10 0.132 0.129 0.012 10.784 0.000
ATRIB11 0.061 0.063 0.017 3.684 0.000
ATRIB12 0.096 0.100 0.017 5.712 0.000
ATRIB13 0.125 0.123 0.014 9.087 0.000
ATRIB14 0.126 0.123 0.014 8.792 0.000
ATRIB15 0.118 0.117 0.015 7.859 0.000
Family
FAMILY 10 000 10 000 0.0000 0.000 1.000
Demographics
SEX 0.802 0.432 0.538 1.492 0.137
NACION –0.446 –0.377 0.452 0.987 0.324
REGION –0.187 –0.179 0.303 0.617 0.538
Field of Training
MAJOR 10 000 10 000 0.0000 0.000 1.000
Education
PREP1 0.320 0.314 0.022 14.517 0.000
PREP2 0.376 0.370 0.018 20.473 0.000
POSSIB1 0.487 0.490 0.039 12.389 0.000
DIFICUL 0.072 0.077 0.052 1.375 0.170
COURSE –0.040 –0.042 0.045 0.880 0.380
Obstacles
OBST1 0.127 0.114 0.097 1.315 0.189
OBST2 0.080 0.082 0.106 0.754 0.452
OBST3 –0.044 –0.035 0.134 0.325 0.746
OBST4 0.045 0.038 0.124 0.360 0.719
OBST5 0.158 0.115 0.128 1.233 0.218
OBST6 0.166 0.150 0.084 1.977 0.049
OBST7 0.307 0.252 0.150 2.051 0.041
OBST8 –0.019 –0.020 0.128 0.150 0.881
OBST9 –0.112 –0.084 0.165 0.679 0.498
OBST10 –0.076 –0.084 0.159 0.481 0.631

Entrepreneurship education and the propensity for business creation 11


















Table 4 Statistical significance of the weights (continued)
Estimate
Indicator Initial sample Bootstrap
Standard
deviation t Sig.
OBST11 0.262 0.222 0.121 2.170 0.031
OBST12 0.200 0.175 0.105 1.907 0.057
OBST13 0.283 0.223 0.141 2.010 0.045
OBST14 0.075 0.067 0.100 0.749 0.454
OBST15 0.100 0.079 0.095 1.057 0.292
Motivation
MOTIV1 0.149 0.147 0.017 8.557 0.000
MOTIV2 0.169 0.165 0.015 11.326 0.000
MOTIV3 0.128 0.128 0.019 6.934 0.000
MOTIV4 0.199 0.198 0.019 10.486 0.000
MOTIV5 0.141 0.143 0.017 8.398 0.000
MOTIV6 0.143 0.145 0.015 9.260 0.000
MOTIV7 0.118 0.122 0.016 7.321 0.000
MOTIV8 0.132 0.131 0.016 8.497 0.000
MOTIV9 0.047 0.042 0.023 2.027 0.044
MOTIV10 0.174 0.171 0.016 10.831 0.000
MOTIV11 0.103 0.104 0.020 5.150 0.000
MOTIV12 0.056 0.055 0.022 2.547 0.011
VAL 0.098 0.098 0.026 3.735 0.000
Propensity to Start Up a Firm
WISH1 0.154 0.157 0.025 6.192 0.000
WISH2 0.203 0.199 0.017 11.796 0.000
EMPRE1 0.264 0.262 0.020 13.183 0.000
EMPRE2 0.279 0.278 0.018 15.406 0.000
CONSIDER 0.185 0.184 0.022 8.562 0.000
POSSIB2 0.271 0.276 0.017 16.141 0.000
Note: Statistically significant with α = 0.05, df = 99.
The results show that all the indicators from Demographics are significant with α = 0.05,
as well as most of the Motivation indicators. As for the other indicators, besides four
from Personal Attributes, three from Education, eight from Obstacles and four from
Propensity to Start Up a Firm, all the others are nonsignificant with α = 0.05. This
suggests that the nonsignificant indicators should not be included in the model. However,
Chin (1998) recommends that indicators should be kept in the model as long as their
deletion implies the loss of useful information. That is the case for the indicators of
Demographics, and so these indicators remain in the model.
The structural model specifies the dependency relationships between the constructs.
To assess this, two criteria can be used: the explanatory capacity of the model and the
structural coefficients’ statistical significance.

12 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















In the first criterion, the R2 for each dependent variable is obtained from the PLS
output. The R2 represents the part of the dependent constructs’ variance that is explained
by the independent variables in the model. Bigger values of R2 indicate better models.
The second criterion consists in considering all the structural coefficients. Chin
(1998) notes that relationships between constructs with structural coefficients bigger than
0.2 should be considered as robust. It should be noted that the total effect of an
independent variable over a dependent variable is bigger than the direct effect, because of
the indirect effects.
There are two structural coefficients (direct effects) with an absolute value bigger
than 0.2 – the effect of Personal Attributes on Motivation and the effect of Education on
Propensity to Start Up a Firm. The direct, indirect and total effects on the Propensity to
Start Up a Firm are shown in Table 5.
Table 5 Direct, indirect and total effects on Propensity to Start Up a Firm
Effects
Construct Direct Indirect Total
Personal Attributes 0.122 0.106 0.228
Family 0.054 n.s. 0.054
Demographics –0.080 n.s. –0.080
Field of Training n.s. n.s. n.s.
Education 0.514 – 0.514
Obstacles –0.173 – –0.173
Motivation 0.178 – 0.178
Note: n.s. = nonsignificant with α = 0.05.
The analysis of the total effects shows that the construct Personal Attributes has a total
effect over Propensity to Start Up a Firm that is bigger than 0.2.
In order to complete the model evaluation, it is necessary to assess its explanatory
capacity, given by the proportion of the total variance of each dependent variable
explained by the model, the R2 (Table 6).
Table 6 Explained variance
Dependent latent variables R2
Motivation 0.390
Propensity (to start up) 0.487
Taking into account the previous results, the initial model was purified by eliminating
the relationships that were not supported by data. These were the relationship between the
constructs Family and Motivation, the relationship between the constructs Demographics
and Motivation, the relationship between the constructs Field of Training and Motivation,
and the relationship between the constructs Field of Training and Propensity to Start Up
a Firm.
The final model with direct effects is presented in Figure 2.

Entrepreneurship education and the propensity for business creation 13


















Figure 2 Final model and direct effects
6 Conclusion
The main objective of this study was to identify the factors that explain and influence the
propensity for the creation of own business.
The proposed model included the following constructs: Personal Attributes, Family,
Demographics, Field of Training, Education, Obstacles, Motivation and Propensity to
Start Up a Firm. It was concluded that the Propensity to Start Up a Firm is a consequence
of all the constructs, with the exception of the Field of Training.
The results of the structural analysis show that only personal attributes explain the
motivation to create a business.
Concerning the construct Propensity to Start Up a Firm, it has weak relations with
Family and Demographics. More relevant is the negative effect of Obstacles, which has a
similar intensity but a contrary effect to Motivation and Personal Attributes.
But the most important effect in the model is that of Education on Propensity
to Start Up a Firm. The results point to the importance of entrepreneurship education
and the promotion of the entrepreneurial intention. That is, the entrepreneurial intention
was significantly predicted by entrepreneurship education. These conclusions find
support in other studies (Kennedy et al., 2003; Brice, 2004; Li, 2006). In light of this
it should be further developed and researched in the context of entrepreneurship

-0,080
0,081
PERSONAL
ATTRIBUTES
DEMOGRAPHICS
OBSTACLES
EDUCATION
MOTIVATION
PROPENSITY TO
START UP A
FIRM
FAMILY
0,122
0,598
0,178
-0,173
0,514

14 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















education in Portuguese universities, in order to understand the antecedents of the
entrepreneurial intention to create a new business and to provide a favourable climate for
entrepreneurship to flourish.
Acknowledgement
The authors are thankful for the funding provided by the INTERREG IIIA programme,
which supported the research project OBSEREGIO.
References
Anderson, J.E. and Gerbing, D.W. (1988) ‘Structural equation modelling in practice: a review and
recommended two-step approach’, Psychological Bulletin, Vol. 103, No. 3, pp.411–423.
Asián, R. (2005) ‘Measuring the entrepreneurial capacity of business men from Coria del Rio:
perceived self-efficacy [in Spanish: Medición de la Capacidad Emprendedora de los
Empresarios de Coria del Rio: La Autoficacia Percibida]’, Selected papers from the
XV Spanish-Portuguese Meeting of Scientific Management, Cities in Competition, Seville.
Barahona, J. and Escudero, A. (2005) ‘Is the entrepreneur born or made? An analysis of
entrepreneurial spirit determinants [in Spanish:¿El Emprendedor Nace o se Hace? Un
Análisis de los Determinantes del Espíritu Emprendedor]’, Selected papers from the
XV Spanish-Portuguese Meeting of Scientific Management, Cities in Competition, Seville.
Baumol, W.J. (1983) ‘Toward operational models of entrepreneurship’, in J. Ronen (Ed.)
Entrepreneurship, Lexington Books, D.C. Heath, pp.29–47.
Brice, J. (2004) The Role of Personality Dimensions on the Formation of Entrepreneurial
Intentions, USASBE Small Business Advancement National Center, University of Central
Arkansas, USA.
Brockhaus, R.H. (1980) ‘Risk taking propensity of entrepreneurs’, Academy of Management
Journal, Vol. 23, No. 3, pp.509–520.
Chin, W. (1998) ‘The partial least squares approach to structural equation modelling’, in
G.A. Marcoulides (Ed.) Modern Methods for Business Research, New Jersey: Laurence
Erlbaum Associates, pp.295–336.
Davidsson, P. (1989) Continued Entrepreneurship and Small Firm Growth, Stockholm School of
Economics, Stockholm, Sweden.
Davidsson, P. and Wiklund, J. (1999) ‘Suitable approaches for studying small firm growth. The
role of entrepreneurship and small and medium enterprises’, Proceedings of the 44th ICSB
World Conference, Naples, Italy, 20–23 July.
Duarte, P. (2005) ‘Brand preference – a study of the factors contributing to the formation of
preference [in Portuguese: A Preferência pela Marca – Estudo dos Factores que Contribuem
para a Formação da Preferência]’, Unpublished doctoral thesis, University of Beira Interior,
Covilhã, Portugal.
Ferreira, J. (2003) ‘Study of growth and performance of the small firm: the influence of
entrepreneurial strategic orientation [in Portuguese: Estudo do Crescimento e Desempenho das
Pequenas Empresas: A Influência da Orientação Estratégica Empreendedora’, Unpublished
doctoral thesis, University of Beira Interior, Covilhã, Portugal.
Gartner, W. (1988) ‘Who is an entrepreneur? is the wrong question’, American Journal of Small
Business, Vol. 12, No. 4, pp.11–33.

Entrepreneurship education and the propensity for business creation 15


















Gorman, G. (1997) ‘Some research perspectives on entrepreneurship education, enterprise
education and education for small business management: a ten-year literature review’,
International Small Business Journal, Vol. 15, No. 3, pp.56–79.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W. (1998) Multivariate Data Analysis,
5th ed., Englewood Cliffs, NJ: Prentice Hall Inc.
Hansemark, O. (1998) ‘The effects of an entrepreneurship programme on need for achievement and
locus of control of reinforcement’, International Journal of Entrepreneurship Behaviour and
Research, Vol. 4, No. 1, pp.28–50.
Hatten, T. and Ruhland, S. (1995) ‘Students attitudes toward entrepreneurship as affected
by participation in an SBI program’, Journal of Education for Business, Vol. 70, No. 4,
pp.224–227.
Honig, B. (2004) ‘Entrepreneurship education: toward a model of contingency-based business
planning’, Academy of Management Learning and Education, Vol. 3, No. 3, pp.258–273.
Kennedy, J., Drennan, J., Renfrow, P. and Watson, B. (2003) ‘Situational factors and
entrepreneurial intentions’, Paper for the Small Enterprise Association of Australia and
New Zealand, 16th Annual Conference, Ballarat, 28 September–1 October.
Kourilsky, M.L. (1980) ‘Predictors of entrepreneurship in a simulated economy’, Journal of
Creative Behavior, Vol. 14, No. 3, pp.175–199.
Kourilsky, M.L. and Walstad, W.B. (1998) ‘Entrepreneurship and female youth: knowledge,
attitudes, gender differences and educational practices’, Journal of Business Venturing,
Vol. 13, No. 1, pp.77–88.
Li, W. (2006) Entrepreneurial Intention among International Students: Testing a Model
of Entrepreneurial Intention’, USASBE Small Business Advancement National Center,
University of Central Arkansas.
Liñán-Alcalde, F. and Rodríguez-Cohard, J. (2004) ‘Entrepreneurial attitudes of Andalisian
university students’, http://ideas.repec.org/p/wiw/wiwrsa/ersa04p161.html (accessed
20 January 2007).
Lundström, A. and Stevenson, L. (2001) ‘Patterns and trends in entrepreneurship. SME policy and
practice in ten economies’, The Swedish Foundation for Small Business Research,
Örebro: FSF.
McMullan, W.E. and Long, W.A. (1987) ‘Entrepreneurship education in the nineties’, Journal of
Business Venturing, Vol. 2, No. 3, pp.261–276.
Rasheed, H.S. (2000) Developing Entrepreneurial Potential in Youth: The Effects of
Entrepreneurial Education and Venture Creation, University of South Florida.
Robinson, P.B., Stimpson, D.V., Huefner, J.C. and Hunt, H.K. (1991) ‘An attitude approach to
the prediction of entrepreneurship’, Entrepreneurship Theory and Practice, Vol. 15, No. 4,
pp.13–31.
Rodrigues, R.G. (2004) ‘Market orientation, entrepreneurial orientation and performance in
Portuguese manufacturing SMEs: exploring the relationship between marketing and
entrepreneurship [in Portuguese: Orientação para o Mercado, Orientação Empreendedora
e Desempenho nas PME Industriais Portuguesas: Exploração da Relação Entre o Marketing
e o Empreendedorismo]’, Unpublished doctoral thesis, University of Beira Interior,
Covilhã, Portugal.
Stephen, F., Urbano, D. and Hemmen, S. (2005) ‘The impact of institutions on entrepreneurial
activity’, Managerial and Decision Economics, Vol. 26, pp.413–419.
Stumpf, S.S., Dunbar, R.L. and Mullen, T.P. (1991) ‘Simulations in entrepreneurship education:
oxymoron or untapped opportunity’, Frontiers in Entrepreneurship Research, pp.681–694.
Veciana, J.M., Aponte, M. and Urbano, D. (2002) ‘Institutions and support programmes for
entrepreneurship: a two countries comparison’, http://www.igw.unisg.ch/rencontres/band2002/
B_05_Veciana.pdf (accessed 20 January 2007).

16 R.G. Rodrigues, M. Raposo, J. Ferreira and A. do Paço

















Vesper, K.H. (1990) New Ventures Strategies, Englewood Cliffs, NJ: Prentice Hall.
Vesper, K.H. and McMullan, W.E. (1988) ‘Entrepreneurship: today courses, tomorrow degrees?’,
Entrepreneurship Theory and Practice, Vol. 13, No. 1, pp.7–13.
Walstad, W.B. and Kourilsky, M.L. (1998) ‘Entrepreneurial attitudes and knowledge of black
youth’, Entrepreneurship Theory and Practice, Vol. 23, No. 2, pp.5–19.