J Knowl Econ

DOI 10.1007/s13132-014-0227-2

Attitude Toward Entrepreneurship: Structure,

Prediction from Behavioral Beliefs, and Relation
to Entrepreneurial Intention

Charalampos Botsaris & Vasiliki Vamvaka

Received: 11 February 2014 / Accepted: 13 October 2014

# Springer Science+Business Media New York 2014

Abstract The present research investigated the associations among the dimensions of
behavioral beliefs, attitude toward entrepreneurship, and entrepreneurial intention. Using
a structural equation modeling approach with a sample of 227 Greek tertiary education
students (114 males, 113 females), support is provided that entrepreneurial outcomes,
viewed as rewards, are distinguished into intrinsic (intangible/psychological/
nonpecuniary) and extrinsic (tangible/physical/pecuniary). Our results show further that
attitude toward entrepreneurship contains three separable components, one being instru-
mental or cognitive in nature, the other being experiential or affective, and the third
representing opportunity costs, that is, the personal and financial sacrifices one is willing
to incur for the sake of the entrepreneurial venture. Concerning intention for entrepre-
neurship, our findings support a two-factor structure that represents commitment to an
entrepreneurial career and nascent entrepreneurship. Evidence is also found for the
Fishbein and Ajzen’s expectancy-value model of attitudes: that attitude toward entrepre-
neurship is predicted by the expectation that entrepreneurship will be followed by a given
outcome, as well as the value that the individual assigns to that outcome. Moreover, our
results reveal that the intrinsic rewards of entrepreneurship are stronger predictors of
entrepreneurial attitude than extrinsic rewards and that the dimensions of attitude toward
entrepreneurship exert a differential impact on entrepreneurial intention, with affective
attitude appearing to be more strongly related to intention than instrumental attitude.

Keywords Entrepreneurship . Attitude . Intention . Expectancy-value model

JEL Classification L26 . J24

C. Botsaris (*) : V. Vamvaka

Department of Economic and Regional Development, Panteion University, 136 Syngrou Avenue,
Athens 176 71, Greece
e-mail: botsaris@otenet.gr
 J Knowl Econ

Introduction

Entrepreneurship is receiving an increased amount of attention in both academic

research and practice, because entrepreneurial activity has been linked with value
creation and, as such, is thought to have a significant impact on economic development,
continuous business renewal, and employment, being the engine of a nation’s economic
and social growth (Engle et al. 2010; Stam et al. 2012; Zampetakis et al. 2009). Thus,
explaining and predicting the choice of an entrepreneurial career have been and still
remains an important research issue (Pruett et al. 2009). Toward identifying the factors
that shape the entrepreneurial decision, several explanatory models of entrepreneurship
determinants have been developed (Kennedy et al. 2003; Schlaegel and Koenig 2014).
Among the models developed to explain how and why new ventures originate, the
intention-based models to predict entrepreneurial activities emerged as dominant in the
entrepreneurship literature (Hindle et al. 2009; Krueger et al. 2000; Schlaegel and
Koenig 2014). The basic rationale behind the intention models is that most behaviors of
social relevance are under volitional control and are thus predictable from intention
(Ajzen 1991, 2002; Ajzen and Fishbein 1980; Armitage and Conner 2001; Bagozzi
et al. 1989; Souitaris et al. 2007; Sutton 1998, as cited in Fini et al. 2012). The overall
tenet of the intention models is that intention is the immediate antecedent of behavior,
while in turn intention is determined by attitudes, and attitudes are affected by
exogenous influences (such as traits, demographics, and situational variables) (Ajzen
1991; Krueger et al. 2000; Shapero and Sokol 1982). A more favorable attitude would
increase the intention to carry out the intended behavior (Fini et al. 2012; Liñán 2004).
Without a positive attitude toward a behavior, one is not likely to intend to engage in
the behavior (Elfving 2008).
Many scholars argue that the decision to become an entrepreneur may be plausibly
considered as voluntary and conscious and that setting up a business involves careful
planning and a thinking process which is highly intentional (Autio et al. 2001; Bird
1988; Krueger et al. 2000). Thus, entrepreneurship has been seen as a good example of
planned intentional behavior and therefore applicable for intention models (Autio et al.
2001; Bird 1988; Davidsson 1995; Fayolle 2006, 2007; Krueger et al. 2000; Shapero
and Sokol 1982).
A variety of intention models have been proposed and tested by entrepreneurship
researchers (Schlaegel and Koenig 2014). However, the model of intentions that has
received predominate attention over the past two decades has involved the theory of
reasoned action (TRA; Ajzen and Fishbein 1980; Fishbein and Ajzen 1975), which
primarily deals with the connection between attitude and behavioral intention, and its
extension, the theory of planned behavior (TPB; Ajzen 1985, 1991). In a variety of
settings, including health promotion, new technology acceptance, social networking,
and entrepreneurship (e.g., French et al. 2005; Hales et al. 2010; Krueger et al. 2000;
Pelling and White 2009; Truong 2009; van der Heijden 2003), it has been shown
repeatedly that the TPB is successful at predicting not only intention to perform a
variety of behaviors, but also whether or not those behaviors are performed (Armitage
and Conner 2001, cited in French et al. 2005). The reasons for the TPB having become
the prevailing model for the explanation of behavioral intentions (Krueger and Brazeal
1994) are its parsimony, conceptual clarity, and ability to be used in many different
behavioral domains (French et al. 2005; Tegtmeier 2012).
J Knowl Econ

Many studies have examined the relation between attitude toward entrepreneurship
and entrepreneurial intention within the framework of the TPB. According to the TPB,
attitudes toward a behavior are predicted by salient beliefs that the behavior will lead to
particular outcomes in multiplicative combination with outcome evaluations. However,
much fewer studies have examined the more affective aspects of behavioral beliefs and
attitude itself, in addition to the more instrumental or cognitive aspects of these
constructs usually examined (Dawson et al. 2009; French et al. 2005; Horvath et al.
2012; Kolvereid 1996a, 1996b; Kolvereid and Isaksen 2006; Kraft et al. 2005; Pardo
and Ruiz-Tagley 2011). In contrast to the cognitive factors, whose importance has been
highlighted in a significant body of literature on entrepreneurship (Liñán et al. 2013),
affect has long been neglected in models of behavioral decision making (Keer 2012).
The aims of this study were four-fold:

1. To assess the dimensionality of the three TPB constructs examined in this study
(attitude toward entrepreneurship, behavioral beliefs, and entrepreneurial
intention),
2. To provide empirical evidence for the assumption of a multiplicative combination
rule for the effects of belief strength and outcome evaluation on overall attitude,
3. To evaluate the relative importance of the various outcomes of entrepreneurship in
predicting attitude, and
4. To investigate whether the affective and cognitive dimensions of attitude toward
entrepreneurship exert a differential impact on entrepreneurial intention.

Theoretical Framework and Research Hypotheses

Attitude Toward Entrepreneurship

Deeply connected to intentional and volitional behavior are beliefs and attitudes (Elfving
2008). Attitude toward a behavior is a person’s overall evaluation of performing the
behavior in question (Ajzen 1991; Fishbein and Ajzen 1975; Mather and Romo 2007;
Schwarz and Bohner 2001). Fishbein and Ajzen (1975: 6) define an attitude as “learned
predisposition to respond in a consistently favorable or unfavorable manner with respect
to a given object.” A similar definition adopt Eagly and Chaiken (1993) who view
attitudes as “psychological tendencies expressed by evaluating a particular entity with
some degree of favor or disfavor.” For Souitaris et al. (2007: 570), “attitude towards self-
employment is the difference between perceptions of personal desirability in becoming
self-employed and organizationally employed.” According to Fini et al. (2012: 390)
“attitude toward behavior, refers to the degree to which a person has a favorable or
unfavorable appraisal of the behavior under scrutiny.”
Theoretical approaches and empirical evidence suggest that attitude should not be
deemed a one-dimensional construct, as overall evaluation often contains two separable
components, one being instrumental or cognitive in nature and the other being expe-
riential or affective. The instrumental component refers to a more cognitive consider-
ation of the extent to which the behavior achieves something valuable or advantageous
and corresponds to the thoughts, knowledge, and perceptions we have about the object
 J Knowl Econ

which commonly take the form of beliefs. The experiential/affective component refers
to emotions, feelings (e.g., joy, satisfaction), and drives engendered by the prospect of
performing a behavior (Ajzen 1991; Fernandes and Proença 2013; French et al. 2005;
Kraft et al. 2005; Lawton et al. 2007; Rhodes and Courneya 2003; Trafimow and
Sheeran 1998; Yan 2014).
Based on the argumentation outlined above, the following hypothesis was
formulated and tested in the study:

H1. A model specifying instrumental and affective components of attitude provides a

better fit than a one-factor model.

Behavioral Beliefs

According to the TPB, attitude toward behavior is determined by the individual’s beliefs
about outcomes or attributes of performing the behavior (behavioral belief strength)
weighted by evaluations of those outcomes or attributes (outcome evaluation).
Entrepreneurial outcomes are viewed as rewards of one’s efforts. Entrepreneurship
research reveals that, although one of the key reasons for entering self-employment is
the potential economic benefits arising from the venture, yet entrepreneurship is not
driven solely by extrinsic, pecuniary rewards like income, but many nonmonetary,
intrinsic rewards play a significant role in motivating the entrepreneurial process (Chell
2001; Croson and Minniti 2012; Guerra and Patuelli 2013; Holland and Shepherd 2013).
Intrinsic (intangible/psychological/nonpecuniary) rewards vest in the activity itself
and include the independence, autonomy, freedom, and control gained by being one’s
own boss, an opportunity for creative expression, personal growth, recognition, chal-
lenge, excitement, satisfying a need for achievement, and self-fulfillment arising from
accomplishing personal goals and expectations. Extrinsic (tangible/physical/pecuniary)
rewards typically refer to obtaining monetary benefits and success (Aspaas 2004;
Carter et al. 2003; Chell 2001; Kuratko et al. 1997; Pardo and Ruiz-Tagley 2011;
Robichaud et al. 2001; Wiklund et al. 2003).
Based on these considerations, we put forward and tested the following hypothesis
regarding the structure of behavioral beliefs:

H2. Behavioral beliefs are best represented by a hierarchical model in which the
entrepreneurial outcomes are distinguished into intrinsic and extrinsic.

Entrepreneurial Intention

Bird (1992: 11) defined intention as “a state of mind directing a person’s attention,
experience and behavior towards a specific object or method of behaving.” According
to Boyd and Vozikis (1994: 64), an entrepreneurial intention is “the state of mind that
directs and guides the actions of the entrepreneur toward the development and
implementation of the business concept.” Along a similar line, Fini et al. (2012) suggest
to keep in mind that entrepreneurial intention reflects a state of mind directing a
person’s attention and action toward the enactment of entrepreneurial behavior.
J Knowl Econ

In an attempt to clarify the construct of intention—in particular entrepreneurial

intention—Thompson (2009: 676) suggested that “individual entrepreneurial intent is
perhaps most appropriately and practically defined as a self-acknowledged conviction
by a person that they intend to set up a new business venture and consciously plan to do
so at some point in the future.”
According to Bird (1988: 442), “entrepreneurs’ intentions guide their goal setting,
communications, commitment, organization, and other kinds of work.” In a similar
vein, Krueger et al. (2000: 412) argue that intentions are “the single best predictor of
any planned behavior, including entrepreneurship.” For Thompson (2009: 670), “en-
trepreneurial intent is substantially more than merely a proxy for entrepreneurship—it
is a legitimate and useful construct in its own right that can be used as not just a
dependent, but as an independent and a control variable.”
No hypothesis can be formulated regarding the dimensions of entrepreneurial
intention.

The Expectancy-Value Model of Attitude

The expectancy-value model of attitudes of Fishbein and Ajzen (1975) postu-

lates that attitudes toward a behavior develop from the total set of accessible
behavioral beliefs people hold about the object of the attitude, linking the
behavior to various outcomes and other attributes. Because the outcomes or
attributes linked to the behavior are already valued in some way, an attitude
toward the behavior is automatically and simultaneously created (Ajzen 1991,
2006). In other words, expectancy-value theory argues that the tendency to
behave in a certain way is predicted by the expectation that a behavior will
be followed by a given outcome, as well as the value that the individual
assigns to that outcome. Ajzen (1991, 2006) expressed this two-element process
by means of the symbolic equation
X
K
AB ∝ bk e k
k¼1

where ABis the attitude toward the behavior and bk is the strength of each belief
which is combined in a multiplicative fashion with the subjective evaluation ek
of the belief attribute, and the resulting products are summed over the K salient
beliefs.
This symbolic equation allows us to explain why persons holding different beliefs
may exhibit identical attitudes, and vice versa, as in the case of two people who may
hold an equally strong belief that entrepreneurship involves facing new challenges, but
one of them may view these challenges as very important to consider when he or she is
to decide his or her future career path while the other may consider them less important
(Moriano et al. 2012).
Evidence for the validity of the expectancy-value hypothesis would be
obtained if addition of the composite belief term in a regression analysis
accounts for additional variance in attitude, over and above the predictive
validity afforded by the belief strength scores and the evaluation scores
(Fishbein and Ajzen 2010).
 J Knowl Econ

Based on the expectancy-value approach to entrepreneurial attitude formation, the

following hypothesis was formulated and tested in the study:

H3. Belief composite accounts for additional variance in attitude, over and above the
predictive validity afforded by belief strength.

The Relative Importance of Entrepreneurial Outcomes in Predicting Attitude

The rational agent-based models used in the research on entrepreneurship and self-
employment transitions assume that better prospects of entrepreneurial earnings as
compared to wages are a major stimulus toward self-employment (Guerra and
Patuelli 2013). Recently, however, the assumption that earnings act as a proxy for
utility has been relaxed, and a number of researchers claim that it is important to
include both economic and psychological motivations in the rational decision process
of potential entrepreneurs (Croson and Minniti 2012). Hamilton (2000) concluded that
nonpecuniary benefits of entrepreneurship are substantial for the self-employed in the
USA and that most entrepreneurs enter and persist in business despite lower initial
earnings and lower earning growth with respect to wage employment, implying a
median earning differential of 35 %. Benz and Frey (2008), using three major data
sets from Germany, Great Britain, and Switzerland, showed that the self-employed
derive higher satisfaction from work than those employed in organizations, irrespective
of income gained or hours worked. Similarly, Croson and Minniti (2012) demonstrated
that newly self-employed individuals are willing to accept lower earning outcomes in
exchange for psychic benefits from self-employment.
Based on this logic and evidence, we hypothesized that

H4. Intrinsic nonmonetary rewards of entrepreneurship are more powerful predictors

of entrepreneurial attitude than extrinsic monetary rewards.

The Attitude-Intention Relation

It is well documented in research that attitude toward entrepreneurship is an important

predictor of entrepreneurial intention (Engle et al. 2010; Schlaegel and Koenig 2014;
Schwarz et al. 2009; Zampetakis et al. 2009). There is also substantial evidence that
affective and cognitive attitudes are distinguishable from each other and that both seem
to be important determinants of at least some intentions (Trafimow et al. 2004). A
number of studies examined the relative contribution of affective and cognitive attitudes
to predict behavioral intentions for a variety of behaviors (French et al. 2005; Kraft et al.
2005; Trafimow et al. 2004). In general, these studies revealed that affective attitude
tends to have a higher predictive power than cognitive attitude in more behaviors.
Based on this evidence, we put forward the following hypothesis regarding the
relationship between attitude toward entrepreneurship and entrepreneurial intention:

H5. Affective attitude is more strongly related to entrepreneurial intention than

instrumental attitude.
J Knowl Econ

Method

Sample

Two hundred twenty-seven Greek tertiary education students (114 males, 113 females)
from the University and the Technological Institute of Central Greece participated in
this cross-sectional study. The mean age and standard deviation of the students were 21
±1.8 years. Their grade point average based on a 0–10 grading scale was 6.9±0.8.
Out of the total number of participants, 106 were studying at the University of
Central Greece (52 at the Department of Economic Regional Development and 54 at
the Department of Informatics with Applications in Biomedicine). Out of the 111
remaining students, 43 were studying at the Faculty of Health and Caring Professions
and 74 were studying at the Faculty of Technological Applications of the Technological
Institute of Central Greece. Four participants did not report their department of studies.

Measures

Behavioral beliefs were assessed using the scale developed by Kolvereid (1996a,
1996b), and Kolvereid and Isaksen (2006) for tapping five outcomes of entrepreneur-
ship: economic opportunity, autonomy, authority, self-realization, and challenge. Two
items were used as indicators of each dimension. The items tapping the outcomes of
entrepreneurship are listed in Table 1.
Note that the items of the economic opportunity subscale tap monetary extrinsic
rewards of entrepreneurship, whereas all other items assess intrinsic rewards. In
specific, the items of the autonomy and the authority subscales assess the control one
feels in position of exerting over life; the items of the self-realization subscale tap the
opportunity for creative expression and the accomplishment of personal goals; and the
items of the challenge subscale assess personal challenge of an interesting and moti-
vating career.

Table 1 Items for the assessment

Economic opportunity
of the outcomes of entrepreneurship
1. To earn enough money.
2. To keep a large proportion of the result.
Autonomy
3. To have independence at work.
4. To be your own boss.
Authority
5. To have the power to make decisions.
6. To issue orders down the chain of command.
Self-realization
7. To realize your dreams.
8. To take advantage of your creative needs
Challenge
9. To have an interesting job.
10. To have a motivating job.
Items 1 and 6 are self-constructed
 J Knowl Econ

Behavioral belief strength was assessed by asking respondents to rate on a 7-point

scale, with extremely unlikely (=1) and extremely likely (=7) as the anchors, the
likelihood that each of the outcomes described in the ten statements would be an
outcome of entrepreneurial activity. Higher score on an item indicated higher likelihood
that entrepreneurship will produce the outcome in the statement.
Outcome evaluation was assessed by asking the respondents to rate on a 7-point
scale, ranging from 1 (extremely undesirable) to 7 (extremely desirable), how desirable
is for them in their future career path each of the ten outcomes of entrepreneurship, as
described in the behavioral belief strength scale. The higher the score on an item, the
higher the subjective desirability of the outcome in the statement.
According to the TPB, attitude toward behavior is predicted by a behavioral belief
composite which is the summation of all salient cognitive beliefs multiplied by the
respondents’ subjective evaluation of these beliefs and gives an indirect measure of
attitude. For both statistical and theoretical reasons, it is suggested (Ajzen 1991; Francis
et al. 2004b) that, when creating this multiplicative composite, behavioral beliefs
should be measured on a unipolar scale (+1 to +7), as belief strength is defined as
the subjective probability that a given behavior will produce a certain outcome,
constituting therefore a unidirectional concept which would seem reasonable to be
measured on a unipolar scale analogous to the 0 to l scale of objective probabilities.
In contrast, outcome evaluations should be recoded into a bipolar scale (1=−3 to 7=+3),
as they are weights and, as such, they may be negative, zero, or positive and they reflect the
differential impact of behavioral beliefs on attitudes (Francis et al. 2004b).
Attitude toward entrepreneurship was directly assessed with a scale developed
utilizing items from existing and widely used scales. The items of the scale, along
with their source, are listed in Table 2. To account for the dimensionality of attitude
toward entrepreneurship, theorists (e.g., Ajzen 2006; Francis et al. 2004a) have pro-
posed that measurement of this construct should involve the use of items which are

Table 2 Scale for the direct measurement of attitude

1. Being an entrepreneur implies more advantages than Liñán and Chen 2009
disadvantages to me.
2. A career as entrepreneur is (totally) attractive for me.
3. If I had the opportunity and resources, I would love to
start a firm (business).
4. Being an entrepreneur would entail great satisfactions for me.
5. I would rather own my own business than earn a higher Kolvereid and Isaksen 2006;
salary employed by someone else. Gundry and Welch 2001
6. I would rather own my own business than pursue another
promising career.
7. I am willing to make significant personal sacrifices in order
to stay in business.
8. I would work somewhere else only long enough to make
another attempt to establish my business.
9. Being an entrepreneur evokes mainly positive thoughts. Leroy et al. 2010
10. Entrepreneurship would present more up than downsides.
J Knowl Econ

evaluative in nature. In addition, these items should reflect the two qualitative compo-
nents of evaluation, the instrumental/cognitive and the experiential/affective.
In this scale, items 1, 9, and 10 represent the instrumental component of attitude,
while items 2, 3, and 4 represent its affective component. The remaining items of the
scale (5, 6, 7, and 8) assess the sacrifices one is willing to incur in order to become self-
employed. The personal and financial sacrifices one is willing to incur for the sake of
the entrepreneurial venture are referred to as opportunity costs (Gundry and Welch
2001). Davidsson (1995), in his economic-psychological model of determinants of
entrepreneurial intentions, included payoff among the variables of the domain attitudes.
For Davidsson (1995), payoff is a composite of beliefs concerning the workload, risk,
and financial gain to be expected by a business founder and concerns the type of
expected outcomes that would typically be included in an explanatory model based on
microeconomic utility theory. Gassar (2006, p. 611) defined opportunity costs as “the
foregone benefit of the next available alternative as a consequence of making a choice”
and provided evidence that an entrepreneur’s opportunity costs are a significant
determinant of the intended scale of venturing activity.
Respondents were asked to indicate their level of agreement with the statements of
the scale from 1 (total disagreement) to 7 (total agreement). A high score on an item,
factor, or the full scale indicated a positive attitude toward entrepreneurship.
Entrepreneurial intention was assessed using the items listed in Table 3, along with
their source. Respondents were asked to indicate their level of agreement with the
statements of the scale from 1 (total disagreement) to 7 (total agreement). A high score
on an item, factor, or the full scale indicated a stronger intention toward
entrepreneurship.

Statistical Analyses

The statistical analyses were performed with SPSS and AMOS software for structural
equation modeling (Arbuckle 2012) using the full information maximum likelihood
procedure to deal with missing values and correlation matrices.

Table 3 Entrepreneurial intention scale

1. My professional goal is to become an entrepreneur. Liñán and Chen 2009

2. I will make every effort to start and run my own firm (business).
3. I am determined to create a firm (business venture) in the future.
4. I have very seriously thought of starting a firm.
5. I consider it to be very likely that in the future I will be running van Gelderen et al. 2008
my own firm.
6. I read books on how to set up a firm. Thompson 2009
7. I spend time learning about starting a firm.
8. I plan to launch my own business some day.
9. I attend seminars and conferences that focus on a “start your McGee et al. 2009
own business planning”.
10. I participate in seminars that focus on writing a business plan.
 J Knowl Econ

All a priori factor models tested were based on the restrictive assumption that the
error terms of the items measuring each construct were uncorrelated. Error correlations
between items within the same factor were incorporated only if they could be justified
theoretically and interpreted substantively (ten Klooster 2008; Joreskog and Sorbom
1988), as in the case when similar response formats are used for survey items (Evans
et al. 2007), whereas error correlations between items across factors were added only
for items with parallel wordings (Magson et al. 2013; Marsh and Hau 1996; Marsh
et al. 1997; Xu et al. 2013).
Competing factorial or structural models were compared by performing a Chi-square
difference test, when the models compared had a nesting relation (O’Connell and
McCoach 2008), and by using Akaike’s information criterion (AIC), when the models
compared were non-nested, with smaller values of AIC indicating improvement in fit
(Hagenaars and McCutcheon 2002; Kaplan 2000). Some researchers (e.g., Chen et al.
2005; Cheung and Rensvold 2002; Vandenberg and Lance 2000) have noticed that, in
evaluating nested models, the comparative fit index (CFI) change is independent of
both model complexity and sample size and suggested that the null hypothesis of no
difference should be rejected, when the reduction in CFI is 0.01 or more.
Fit indices used to determine goodness of fit included the ratio χ2/df with values less
than 5 indicating an adequate fit and less than three a good fit (Kline 2010; Marsh and
Hovecar 1985), the CFI (Bentler 1990), and the Tucker-Lewis index (TLI; Tucker and
Lewis 1973) with values above 0.90 representing an adequate fit and above 0.95 a good
fit (Hu and Bentler 1999), and the root mean square error of approximation (RMSEA;
Browne and Cudeck 1993) with values of 0.05 or smaller representing close fit, below
0.08, reasonable fit; below 0.10, minimally acceptable fit; and above 0.10, unacceptable
fit (Browne and Cudeck 1993). RMSEA is considered one of the most useful indicators
of model fit (Byrne 2001).

Results

Dimensionality of the TPB Constructs

Dimensions of Attitude

French et al. (2005) suggest, before attempting to predict intention using measures of
instrumental and affective attitudes, to establish first that factor analytic techniques can
distinguish between items that are supposed to reflect each of the components, as, if
affective and instrumental components of attitude were genuinely distinct, one would
expect them to form different factors. Following the theoretical approaches to attitude
and the empirical evidence on its components, various competing models were tested
using a nested confirmatory factor modeling approach: a three-factor model, a
two-factor model, and a one-factor model.
We first estimated a three-factor model, M3, in which items 1, 9, and 10 were
allowed to load on only a factor labeled “instrumental attitude,” items 2, 3, and 4 were
allowed to load on only a factor labeled “affective attitude,” and items 5, 6, 7, and 8
were allowed to load on only a factor labeled “opportunity costs.” In model M3, the
correlations between the three factors were freely estimated. Next, we fitted to
J Knowl Econ

responses of a two-factor model, M2, obtained by constraining the correlation between

instrumental attitude and affective attitude at 1. Setting this correlation to one collapses
the two factors into one factor, consisting of the instrumental and affective items,
resulting in model M2 to be nested within model M3. The one-factor model, M1,
was obtained by constraining the correlations of the opportunity cost factor with the
instrumental and affective factors at one resulting in model M1 to be nested within
model M2. To accomplish this, we fixed the latent variable’s variances to one, rather
than fixing factor loadings to one. The fit indices of the above mentioned competing
models and the Chi-square difference tests conducted in order to compare alternative,
nested models, are presented in Table 4.
The data in Table 4 show that the three-factor model fitted better than the other two
models. In addition, the overall fit of the three-factor model was acceptable. To improve
the overall model fit, modification indices were then assessed, and the model has been
modified by allowing correlated residual variances between different indicators, but
within the same factor, in an effort to limit the complexity of the model (Byrne 2001).
Inspection of modification indices suggested that model fit would improve substantially
if we allowed residuals to correlate between items 5 (“I would rather own my own
business than earn a higher salary employed by someone else”) and 6 (“I would rather
own my own business than pursue another promising career”) in the opportunity cost
factor. The comparison of the two models using a Chi-square difference test revealed
that the model with the one pair of correlated error terms was a significantly better fit
than the unmodified model [Δχ2 =19.930, Δdf=1, p<0.001]. Moreover, the changes
of the standardized factor loadings were less than 0.10. The correlation between the
residuals of items 5 and 6 may be attributed to the common content and wording
(“I would rather own my own business”) shared by these two items. On the
basis of the statistical evidence, the similarities in the content and structure of
the items involved, and the fact that the correlated residuals are associated with
indicators of the same construct, we can conclude that the increased complexity
induced by incorporating these correlated errors is justified and would not alter
the meaning or interpretability of our model.
The modified model had a reasonable fit to the data [χ2/df=2.157, TLI=0.968,
CFI=0.978, RMSEA=0.072]. All loadings and correlations were substantial and
significant at p < 0.001. Moreover, the squared multiple correlations of the
indicators ranged from 0.46 to 0.83, indicating that a large portion of variance
was accounted for in each indicator by its underlying factor. High Cronbach
alphas of 0.90, 0.86, and 0.87 were obtained for the affective, instrumental, and
opportunity cost components, respectively.

Table 4 Fit statistics for competing models of the factor structure of attitude

Model χ2 df χ2/df TLI CFI RMSEA Models Δχ2 Δdf p

compared

M3. Three factors 86.793 32 2.712 0.953 0.966 0.087

M2. Two factors 112.470 33 3.408 0.933 0.951 0.104 M2-M3 25.677 1 <0.001
M1. One factor 246.703 35 7.049 0.833 0.870 0.164 M1-M3 159.909 3 <0.001
 J Knowl Econ

These results verify hypothesis H1 and are consistent with the view that attitude
should not be considered, nor measured, as a unidimensional construct.

Dimensions of Behavioral Beliefs

Based on the theoretical approaches to the rewards of entrepreneurship and the a priori
five-factor structure of the scales, alternative factor models were tested and compared
using confirmatory factor analyses. Each analysis included all 20 items from both
scales, while factorial invariance, that is, the same factor structure, was postulated
across the two scales. The models compared were as follows: (a) a model with five
first-order correlated factors: economic opportunity, autonomy, authority, self-realiza-
tion, and challenge; (b) a model with five first-order factors and two second-order
correlated factors: “self-fulfillment” consisting of the self-realization and challenge first-
order factors, and “control over career” consisting of the autonomy and authority first-
order factors; and (c) a model in which the self-fulfillment and control over career
second-order factors loaded on a third-order factor labeled “intrinsic rewards.” We
recall that a higher-order factor underlies correlations between lower order factors.
Due to the identical wording of the indicators (belief strength and outcome evalu-
ation items), the residuals between identically worded items across the two scales were
allowed to correlate in order to obtain more accurate estimates and avoid inflating the
correlation between belief strength and outcome evaluation corresponding factors.
The fit indices of the above described competing models are presented in Table 5. As
suggested by the AIC and the other data in Table 5, the third-order factor model fitted
better than the other two models. Moreover, the best fitted model had a close fit to the
data. In this model, all loadings and correlations were substantial and significant at
p<0.001. In addition, the squared multiple correlations of the indicators ranged from
0.54 to 0.82, indicating that a large portion of variance was accounted for in each
indicator by its underlying factor.
The Cronbach alphas of economic opportunity, autonomy, authority, self-realization,
and challenge dimensions were found equal to 0.79, 0.83, 0.81, 0.80, and 0.88,
respectively, for the belief strength scale, and equal to 0.75, 0.79, 0.76, 0.77, and
0.89, respectively, for the outcome evaluation scale.
These results verify hypothesis H2, showing that behavioral beliefs are best repre-
sented by a third-order factor model in which the entrepreneurial outcomes are distin-
guished into intrinsic and extrinsic.

Table 5 Fit statistics for competing models of the factor structure of the belief strength and the outcome
evaluation scales

Model χ2/df TLI CFI RMSEA AIC

Five first-order factors 1.741 0.948 0.968 0.058 390.220

Five first-order factors and two second-order 1.709 0.950 0.964 0.056 380.192
factors (self-fulfillment, control over career)
Five first-order factors, two second-order factors 1.671 0.953 0.965 0.055 373.319
(self-fulfillment, control over career) and one
third-order factor (intrinsic rewards)
J Knowl Econ

Dimensions of Intention

Since no hypothesis was made about the dimensionality of the scale assessing entre-
preneurial intention, first, an exploratory factor analysis (EFA) of the items of the scale
was performed on the first random split-half sample to identify a viable factor structure.
The EFA was performed using principal component analysis with an oblique rotation,
which assumes factors to be correlated with each another (Jennrich and Sampson
1966). The compliance of the data with the factor analysis was ascertained with
Kaiser-Meyer-Olkin (KMO) and Barlett sphericity test. KMO was found equal to
0.86, and Barlett’s sphericity test was found to be meaningful [χ2(45)=743.317,
p<0.001], indicating that the scale is in compliance with the factor analysis. The
analysis yielded two eigenvalues greater than one (equal to 5.215 and 1.940)
suggesting, according to Kaiser’s (1960) criterion, that two factors should be extracted.
These two factors explained the 71.6 % of the total variance. The factor structure of the
scale was very clear. In specific, all items had primary loadings over 0.57, with only
two cross-loadings being slightly above the recommended cutoff level of 0.30
(Tabachnick and Fidell 1996). Under this two-factor solution, items 1, 2, 3, 4, 5, and
8 were grouped in one factor which reflects “commitment to an entrepreneurial career”
and is consistent with the understanding of Krueger (1993) of entrepreneurial intention
as the commitment to start a new business, while the remaining four items loaded onto
the second factor which can be labeled “nascent entrepreneurship.”
Some researchers have challenged the view that entrepreneurial activity is a
static process with two outcomes—to engage or to refrain from it—and put
forth a dynamic approach that discriminates between prebirth, birth, and post-
natal stages of entrepreneurship, which may have different antecedents, with the
prebirth stage often being referred to as latent or nascent entrepreneurship
(Grilo and Thurik 2005, 2008; Masuda 2006; Reynolds 1997; van der Zwan
et al. 2010; van Gelderen et al. 2005). Latent entrepreneurship (hidden or
potentially existing but not yet realized) is defined as the preference for
entrepreneurship over paid employment, and it is visible in people who are
willing and able to become entrepreneurs but have not yet decided to start a
business, while nascent (emerging) entrepreneurship refers to individuals who
have made a decision and are taking some action toward creating a new
business (Blanchflower et al. 2001; Hayek 2012; Verheul et al. 2012).
The two-factor model for intention was subsequently verified through confir-
matory factor analysis conducted on the second random split-half sample. The
results of the analysis showed that the model had an acceptable fit to the data
[χ2/df=3.127, TLI=0.958, CFI=0.970, RMSEA=0.097]. These fit indices were
achieved by allowing correlated errors between the following pairs of items: (a)
item 6 (“I read books on how to set up a firm”) and item 7 (“I spend time
learning about starting a firm”), with both items referring to reading or learning;
(b) item 9 (“I attend seminars and conferences that focus on a start your own
business planning”) and item 10 (“I participate in seminars that focus on writing
a business plan”), with both items referring to attending or participating in a
seminar. Note that all these items are from the nascent entrepreneurship factor.
Incorporating these correlated errors in the model was considered acceptable on
the basis of the similarities in the content of the items involved.
 J Knowl Econ

Correlational Analysis

After the factor analyses were performed, a score was calculated for each construct by
adding together all the items comprising the corresponding subscale to obtain a total
score for each. The correlations among the constructs are presented in Table 6. These
correlations fall within the range of those reported in relevant studies and reviews in the
literature.
For example, in a study by Kolvereid (1996b), the correlations of economic
opportunity, autonomy, authority, and self-realization with intention were found equal
to 0.11, 0.32, 0.31, and 0.20, respectively. In another study by Kolvereid and Isaksen
(2006), the corresponding correlations were found equal to 0.21, 0.07, 0.13, and 0.11.
In the same study, the correlations of economic opportunity, autonomy, authority, and
self-realization with attitude, as measured by the subscale used in our study to assess
opportunity costs, were found equal to 0.26, 0.28, 0.20, and 0.23, respectively. The data
in Table 6 show, however, that the relations of behavioral belief composite with attitude
and intention are somewhat stronger than those found by previous studies. We attribute
this difference in the magnitude of correlations to the fact that the belief construct used
in these studies was outcome evaluation while the belief construct in Table 6 is belief
composite.
Consistently with the findings of previous studies, the data in Table 6 show also that
attitude is related with intention at levels ranging from r=0.47 to r=0.75. In a meta-
analysis of 23 studies, Hausenblas et al. (1997) found a weighted average correlation of
0.52 between attitude and intention. In another meta-analysis of Armitage and Conner
(2001), conducted with 185 independent empirical tests of the theories of reasoned
action and planned behavior reported in 161 journal articles and book chapters, the
average correlation between attitude and behavioral intentions was found equal to 0.49.
In the most recent meta-analysis of 98 relevant studies (123 samples, N=114.007),

Table 6 Pearson r correlations among the analysis variables

Construct 1 2 3 4 5 6 7 8 9 10

Belief composite
1. Economic opportunity 1.00
2. Autonomy 0.53 1.00
3. Authority 0.54 0.71 1.00
4. Self-realization 0.46 0.61 0.48 1.00
5. Challenge 0.41 0.57 0.44 0.67 1.00
Attitude
6. Instrumental attitude 0.42 0.44 0.42 0.38 0.53 1.00
7. Affective attitude 0.41 0.48 0.42 0.37 0.48 0.79 1.00
8. Opportunity costs 0.34 0.41 0.39 0.32 0.39 0.64 0.70 1.00
Entrepreneurial intention
9. Commitment to entrepreneurship 0.37 0.43 0.41 0.35 0.40 0.65 0.75 0.68 1.00
10. Nascent entrepreneurship 0.23 0.27 0.34 0.24 0.23 0.50 0.47 0.52 0.51 1.00

All correlations are statistically significant at the 0.001 level (two-tailed tests)
J Knowl Econ

Schlaegel and Koenig (2014) utilized meta-analytic structural equation modeling to

examine the empirical fit of the competing theories and found the correlation between
attitude toward entrepreneurship and entrepreneurial intent to be equal to 0.43.

Testing Fishbein and Ajzen’s Expectancy-Value Model of Attitudes

Toward testing whether belief strength and the corresponding belief strength/outcome
evaluation cross-product each contributes significantly, positively, and independently to
the prediction of attitude, we first recoded each outcome evaluation item into a bipolar
scale, and then, we multiplied each behavioral belief item score with its corresponding
outcome evaluation item score to obtain a composite score for each item of the scale in
Table 1. However, instead of using aggregate measures in a regression analysis
10
obtained by summing the belief strength scores ( ∑ bk ) and the product scores for each
k¼1 10
of the ten belief strength and outcome evaluation combinations ( ∑ bk ek ), we adopted a
k¼1
structural equation modeling approach where belief strength and belief composite were
treated as latent constructs to control for random measurement error and get unbiased
estimates of effects (Green et al. 1999).
Each of these two latent constructs was measured by three-item parcels as indicators:
economic opportunity, control over career, and self-fulfillment. Little et al. (2002) argue
that using item parcels as indicators produces more reliable latent variables than individ-
ual items. The economic opportunity score was produced by averaging the two items of
the corresponding subscale, while the four items of the autonomy and the authority
subscales and the four items of the self-realization and the challenge subscales were
averaged to produce a control over career score and a self-fulfillment score, respectively.
Overall attitude was also treated as latent construct formed by parceling the scale for the
direct measurement of attitude into three parcels composed of the average of the items of
the affective, the instrumental, and the opportunity costs subscales, which emerged as
distinct factors in the CFA of the scale assessing directly attitude.
The model used to test the expectancy-value approach to attitude toward entrepre-
neurship is illustrated in Fig. 1. In this model, direct paths were hypothesized from the
latent constructs of belief strength and belief composite to the latent construct of
attitude. Due to the identical wording of the item parcel indicators of belief strength
and belief composite, we allowed residuals to correlate between the matching indicators
across the two constructs. The theoretical model proposed and tested fitted the data well
(χ2/df=2.100, TLI=0.969, CFI=0.982, RMSEA=0.070). The structural coefficients for
the model are presented in Fig. 1 with standardized values. The path from belief
composite to attitude was significant at the p=0.05 level (critical ratio=2.354,
p=0.02), whereas the path from belief strength approached statistical significance
(critical ratio=1.853, p=0.06). These findings support the expectancy-value model of
attitudes of Fishbein and Ajzen (2010) and verify hypothesis H3.

Exploring the Importance of Entrepreneurial Outcomes in Predicting Attitude

In order to identify the entrepreneurial outcomes that contribute mostly to entrepreneurial

attitude formation, a structural equation model was fit for each of the three components of
attitude which was used as the dependent variable, while the three belief composites
 J Knowl Econ

1
Self-fulfillment 0.83

1 Beliefs
Control over career 0.85
strength
Instrumental 1
0 .72 attitude
1
Economic opportunity
0.30 0.87
0.12

0.87 Affective 1
0.48 Attitude 0.91 attitude

0 .54 0.40
0.76
1
Economic opportunity
Opportunity 1
costs
0.68
1
Control over career 0.81 Beliefs
composite
0.75
1
Self-fulfillment

Fig. 1 Path model predicting the effects of belief strength and belief composite on attitude. Note: latent
constructs are represented as ovals and multiple indicators of the latent constructs as rectangles. Single-
arrowed paths represent hypothesized directional relationships. Curved double arrow-headed lines represent
correlations between constructs or between construct residuals. The numbers by the single arrow-headed lines
are standardized regression weights. The dotted curved double arrow-headed line represents nonsignificant
correlation. Solid lines represent significant coefficients at the p=0.05 level

(economic opportunity, control over career, and self-fulfillment) were the independent
variables. The scores of the autonomy and authority first-order factors and the scores of
the self-realization and challenge first-order factors were used as observed indicators of
the control over career and self-fulfillment second-order constructs, respectively, while
economic opportunity was measured by the two indicators of the corresponding subscale.
Affective attitude and instrumental attitude were each measured by the three items of the
corresponding subscale, whereas the construct of opportunity costs was formed by
parceling the items of the corresponding subscale into two parcels.
All three models fitted the data very well (affective attitude χ2/df=1.804, TLI=
0.964, CFI=0.974, RMSEA=0.060; instrumental attitude χ2/df=2.101, TLI=0.949,
CFI=0.963, RMSEA=0.070; opportunity costs χ2/df=1.741, TLI=0.967, CFI=0.977,
RMSEA=0.058). The results obtained showed that the most important predictor of
affective attitude was control over career (Beta=0.353, critical ratio=2.871, p=0.004),
followed by self-fulfillment (Beta=0.286, critical ratio=2.252, p=0.024), whereas the
most important predictor of instrumental attitude was self-fulfillment (Beta=0.498,
critical ratio=4.913, p<0.001), followed by economic opportunity (Beta=0.212, crit-
ical ratio=2.314, p=0.021). Control over career was the only significant predictor for
opportunity costs (Beta=0.517, critical ratio=5.834, p<0.001).
These results verify hypothesis H4, showing that the intrinsic nonmonetary rewards
of entrepreneurship are more important in predicting attitude than the extrinsic mone-
tary rewards.

Exploring the Attitude-Intention Relation

The conceptual framework for exploring the relations among behavioral beliefs,
attitude, and intention is presented in Fig. 2. In this model, behavioral beliefs, attitude,
and intention were treated as latent constructs. Consistent with the TPB and the results
J Knowl Econ

obtained when testing the Fishbein and Ajzen’s expectancy-value model of attitudes,
behavioral beliefs were represented by the latent construct of belief composite mea-
sured by three-item parcels as indicators: economic opportunity, control over career,
and self-fulfillment. To investigate whether the attitude dimensions have a differential
impact on intention, attitude was represented by three latent constructs: affective
attitude, instrumental attitude, and opportunity costs, measured as in the previous
section. Entrepreneurial intention and nascent entrepreneurship were each measured
also by two indicators formed by aggregating the items of the corresponding subscale
into two parcels.
In the model illustrated in Fig. 2, direct paths were allowed from belief composite,
which was treated as the independent variable, to affective attitude and instrumental
attitude. Noting that opportunity costs, as measured in our work, represent one’s
preference for entrepreneurship, affective attitude and instrumental attitude were, in
turn, allowed to impact opportunity costs, as suggested by Ajzen’s (1991) TPB. By
further viewing preference for entrepreneurship as reflecting what Ajzen and Fishbein
(1980) referred to as “choice intentions,” a direct path was postulated from opportunity
costs to commitment to entrepreneurship, since the effect of choice intentions is
predominant over behavioral intentions (Kolvereid 1996a; Verheul et al. 2012).
We also hypothesized that commitment to entrepreneurship directly affects nascent
entrepreneurship. This direct path is in accordance with the TPB that intention is
assumed to be the immediate antecedent of behavior and the view that the most
proximal measure of entrepreneurial behavior is that of nascent gestation behaviors,
such as writing a business plan (Martin 2012). Finally, we allowed residuals to correlate

1 1 1

#1 #9 #10
1 1 1
0.74 0.91 0.80 P1 P1 P1
1
Self-fulfillment
Instrumental 0.88
0.75 0.95
0.55 attitude
0.96 0.27 .88
1 Commitment to .65 Nascent
0.44 Economic opportunity Beliefs Opportunity
0 .49 entrepreneurship entrepreneurship
composite costs
0 .94 0.63
0.56 Affective 0.87 0.96 0.57
1 attitude
Control over career
P2 P2 P2
0.89 0.81 0.89
1 1 1
#2 #3 #4

1 1 1

Fig. 2 Conceptual path model of expected relationships among belief composite, attitude dimensions,
intention, and nascent entrepreneurship. Note: latent constructs are represented as ovals and measured
variables and multiple indicators of the latent constructs as rectangles. Single-arrowed paths represent
hypothesized directional relationships. The curved double arrow-headed line represents correlation between
construct residuals. The dotted single arrow-headed line represents significant coefficient at the p=0.10 level.
Solid lines represent significant coefficients at the p<0.001 level. The numbers by the single arrow-headed
lines are standardized regression weights. The numbers within the rectangles representing the indicators of
instrumental and affective attitude correspond to the item numbers of the attitude scale in Table 2. The
indicators of opportunity costs, intention, and nascent entrepreneurship are item parcels. For opportunity costs,
parcel P1 consists of items 5 and 6, and parcel P2 consists of items 7 and 8 of the attitude scale in Table 2. For
commitment to entrepreneurship, parcel P1 consists of items 1, 2, and 3, and parcel P2 consists of items 4, 5,
and 8 of the intention scale in Table 3. For nascent entrepreneurship, parcel P1 consists of items 6 and 7, and
parcel P2 consists of items 9 and 10 of the intention scale in Table 3
 J Knowl Econ

between control over career and self-fulfillment, as these two constructs share a higher-
order factor of intrinsic rewards.
The theoretical model proposed and tested fitted the data well (χ2/df=2.735, TLI=
0.922, CFI=0.939, RMSEA=0.088). Moreover, none of the direct paths not included in
the model had a modification index greater than 5. However, in this model, the direct
path from instrumental attitude to opportunity costs was not significant although
approached significance (critical ratio = 1.687, p = 0.09). The total effect of
instrumental attitude on both commitment and nascent entrepreneurship was roughly
half as great as that of affective attitude (total effect of affective attitude on
commitment=0.55, total effect of instrumental attitude on commitment=0.28, total
effect of affective attitude on nascent entrepreneurship = 0.39, total effect of
instrumental attitude on nascent entrepreneurship = 0.20). Overall, the model
explained the 76.5 % of the variance in commitment to an entrepreneurial career and
the 42.7 % of the variance in nascent entrepreneurship.
These results verify hypothesis H5 and are consistent with the research findings from
the literature that affective attitude tends to be more strongly connected to intention than
instrumental attitude (French et al. 2005; Kraft et al. 2005; Trafimow et al. 2004) and
that attitudes explain over 50 % of the variance in intentions and intentions account for
over 28 % of the variance in behavior (Kim and Hunter 1993; Sheeran 2002).
More recently, Zampetakis et al. (2009), Iakovleva et al. (2011), and Liñán et al.
(2013) conducted studies using structural equation modeling to test the TPB in the
prediction of entrepreneurial intention. In the study of Zampetakis et al. (2009), the
direct effect of attitude to intention was found equal to 0.92; in the study of Iakovleva
et al. (2011), this effect was equal to 0.57, while Liñán et al. (2013) obtained a direct
effect equal to 0.60. We note that these studies incorporated all three antecedents of
intention, namely, attitude, subjective norms, and perceived behavioral control. In our
study, the direct effect of opportunity costs to intention was found equal to 0.88
(standardized value).

Discussion

This study focused on three of the TPB constructs within the context of entrepreneur-
ship, namely, entrepreneurial behavioral beliefs, attitude toward entrepreneurship, and
entrepreneurial intention, by specifically examining the dimensionality of these con-
structs and evaluating structural relations among their components using structural
equation modeling via AMOS.
Consistent with prior research which suggests that entrepreneurial outcomes can be
both extrinsic, such as income, personal wealth, value creation, innovation, growth, and
profit, and intrinsic, such as satisfaction, independence, excitement, and challenge
(Petkova 2009), our findings revealed that entrepreneurial outcomes, as assessed by
the self-report measure of our study, are not a unitary construct but consist of five
distinct components: economic opportunity, autonomy, authority, self-realization, and
challenge. However, our study goes one step further in that it provided strong support
for a hierarchical structure of entrepreneurial outcomes.
In common with previous research examining the dimensions of attitude toward
entrepreneurship, the data in our study supported a two-factor model consisting of both
J Knowl Econ

instrumental and affective attitudes than a model consisting of only a single attitude
factor. Our findings add to the literature regarding the dimensionality of attitude toward
entrepreneurship, which provides adequate evidence for the distinction of attitude into
two subtypes: instrumental and affective (Ajzen 1991; Fernandes and Proença 2013;
French et al. 2005; Kraft et al. 2005; Lawton et al. 2007; Rhodes and Courneya 2003;
Trafimow and Sheeran 1998; Yan 2014). It should, however, be noted that, although
instrumental and affective attitude emerged as distinct factors in our factor analytic
investigation of the data, these factors are both evaluative in nature, and as such, it is
expected to be interrelated.
Having established the multidimensional nature of attitude, it is reasonable to
explore how each of the components of attitude relates to behavioral beliefs and to
entrepreneurial intention.
Consistent with previous studies, our findings suggest that affective attitude is a
stronger predictor of intention than instrumental attitude. Attitude is one of the central
constructs in the theories of reasoned action and planned behavior. As implied by its
name, the TRA suggests that cognitive attitudes should be more strongly related to
intentions, as intentions are presumed to be based on an expectancy-value model of
attitudes, which takes into account both the subjective probability and the value of
various potential outcomes produced by a given behavior (Trafimow et al. 2004). On
the other hand, research from an evolutionary perspective (e.g., Damasio 1994;
Johnston 1999) suggests that affect provides the motivational power for behaviors
and should, therefore, be more important than cognition for predicting most intentions.
Intention is the personal, deliberate, goal-oriented, proactive aspect of behavior and
is closely related to the concept of volition and intrinsic motivation (Fernandes and
Proença 2013), which refers to the individual’s desire to engage in a behavior because it
is personally rewarding, essentially, performing an activity for its own sake and not in
order to attain some extrinsic goal (Bénabou and Tirole 2003; Kivetz 2005). As positive
affect enhances intrinsic motivation (Isen and Reeve 2005), affectively based attitudes
tend to be more persistent and resistant to change despite situational conditions or
rational arguments, while instrumental attitudes are based on calculative considerations
of the advantageous or disadvantageous outcomes of a certain behavior, and as such,
they are more vulnerable to external incentives and can be more easily invalidated and
changed (Fernandes and Proença 2013; Heath and Nairn 2005; LaBarge 2007). This
argument offers an explanation for the primacy of affective attitude over instrumental
attitude in the prediction of intention,
Another explanation can be derived from the work of those researchers who
suggested that the relative importance of affective attitude and instrumental attitude
in the prediction of intention depends either on the individual’s personality or on the
nature of the behavior. Thus, Trafimow et al. (2004) noted that some people are more
under affective control, across behaviors, whereas other people are more under cogni-
tive control, and concluded that the predictive validity of affect and cognition in
determining entrepreneurial intention should depend upon whether participants in a
study are affectively or cognitively controlled. However, other researchers (e.g., Yan
2014) argue that the impact of the instrumental and affective components of attitude on
the formation of the behavioral intention is moderated by the nature of the behavior and
that affective attitude should be a more powerful predictor of intention for behaviors
which are under the volitional control of the individual. The favorable feelings or
 J Knowl Econ

emotions associated with a volitional behavior, such as entrepreneurship, obviously

have a strong impact on intention (Yan 2014).
Our study has further revealed that the intrinsic rewards of entrepreneurship exert a
stronger influence on all attitude components than the extrinsic rewards. In particular,
the control over career second-order factor, consisting of autonomy and authority, was
the strongest predictor of both affective attitude and opportunity costs, while the self-
fulfillment second-order factor, consisting of self-realization and challenge, was the
strongest predictor of instrumental attitude. This is an important finding since affective
attitude was the most influential predictor of opportunity costs and entrepreneurial
commitment, while opportunity costs, as measured in our work, reflect choice
intentions.
Hamilton (2000) provided empirical evidence that self-employment offers signifi-
cant nonpecuniary benefits, such as “being your own boss.” Kato (2013) concluded that
entrepreneurship is a process of self-fulfillment and search for happiness, and it is not
always about making a profit or rapidly growing a business. Benz and Frey (2008),
using job satisfaction as a proxy measure for utility from work, showed that the self-
employed enjoy considerably higher job satisfaction than employees, and that the self-
employment-job satisfaction effect can to a large extent be explained by procedural
aspects of work, namely, higher autonomy and more rewarding work content. Many
studies in the economic discourse have documented that the self-employed enjoy
higher job satisfaction than employees and that the impact of self-employment on job
satisfaction is mediated by greater procedural freedom and autonomy (Lange 2012).
Croson and Minniti (2012) mention a number of studies which suggest that the self-
employed systematically differ from the organizationally employed with respect to
career anchors and that the self-employed are more likely to have autonomy, indepen-
dence, and entrepreneurial creativity anchors.
Croson and Minniti (2012) offered a multifactor utility formulation formalizing the
notion of an explicit, autonomy-based preference for self-employment and argued that
their model provides a direct link to Deci and Ryan (2000) self-determination theory
(SDT). Along a similar line, Benz and Frey (2008) noted that individuals derive
procedural utility from being self-employed because it gives them a higher measure of
self-determination and freedom. SDT maintains that an understanding of human moti-
vation requires a consideration of innate psychological needs for competence, autonomy,
and relatedness. Social contexts and individual differences that support satisfaction of the
basic needs facilitate natural growth processes including intrinsically motivated behavior
and integration of extrinsic motivations, whereas those that forestall autonomy, compe-
tence, or relatedness are associated with poorer motivation, performance, and well-being.
Another interesting finding of our study is that belief composite was able to explain
additional variance in attitude beyond that explained by belief strength. Commenting
on those investigators who challenged the assumption of a multiplicative combination
rule for the effects of belief strength and outcome evaluation on overall attitude,
Fishbein and Ajzen (2010, p. 117) argue that “on strictly theoretical grounds, there is
no logic in predicting attitude from belief strength without taking the evaluative
implication of the belief into account or in predicting attitude from outcome evaluation
without knowing whether the outcome is considered likely or unlikely.”
A number of authors argue that using multiplicative composites is unsatisfactory
from a statistical point of view and give a number of reasons why the multiplicative
J Knowl Econ

approach is flawed (for a review, see Francis et al. 2004b). French and Hankins (2003)
recommend either the use of alternative analytic strategies, such as experimental
designs and ANOVA analyses, or the use of individually salient beliefs rather than
“modal” beliefs. This view has been disputed by Francis et al. (2004b) who see it as
leading to an uncomfortable pair of options.
We argue that multiplying the strength of each belief by the corresponding subjec-
tive evaluation, and summing these cross-products over all salient beliefs to obtain on
overall index of belief composite can obscure important distinctions between subtypes
of behavioral beliefs. On these grounds, in our study, we adopted a structural equation
modeling approach where belief strength and belief composite were treated as latent
constructs measured by use of multiple indicators to control for random measurement
error and get unbiased estimates of effects, but also to analyze data at the item
or item-parcel level.

Contributions, Implications, and Limitations

Our study contributes to the literature on entrepreneurship examining the relations

among behavioral beliefs, attitude, and intention within the TPB in several ways. To
our knowledge, the work reported in this paper is one of the first attempts—if not the
first one—to disentangle the dimensions of the constructs under investigation and to
explore the associations among these dimensions using both direct and indirect mea-
sures of attitude and adopting a structural equation modeling approach. Using structural
equation modeling and treating the variables under investigation as latent constructs
inferred from multiple indicators allows controlling for random measurement error and
getting unbiased estimates of effects and, thus, for more robust and sophisticated testing
in order to better understand these effects (Green et al. 1999; Liñán et al. 2013). In this
respect, the present study addressed a limitation of existing research in TPB—not just
for entrepreneurship—that has used linear regression models, despite the risk of
missing indirect and complex effects (Liñán et al. 2013).
Our results revealed that affective attitude exerted twice as much effect on intention
compared to instrumental attitude. This finding echoes the suggestions of scholars that
it is important and meaningful to distinguish the affective and instrumental components
of attitude in TPB research and to identify the unique contribution of each in predicting
intentions. This is because, when conceptualizing attitude as unidimensional construct,
the paths whereby affect and cognition exert their impact on intention are obscured and
the large direct effect of affect on intention is neglected (French et al. 2005; Keer 2012;
Kraft et al. 2005; Yan 2014). We note that most of the scales commonly used in the
literature to measure attitude include both affective and cognitive considerations, but
their items are either summed to produce a total attitude score which is included in a
regression model or they are used as indicators of a latent global attitude construct
(Engle et al. 2010; Iakovleva et al. 2011; Liñán et al. 2013).
In addition, this study further adds to the utility-based literature on the trade-off
between increased autonomy from self-employment and the generally higher income
that traditional employment offers (Croson and Minniti 2012; Lange 2012) by showing
that the control over career composite, consisting of autonomy and authority, was the
most significant predictor of both affective attitude and opportunity costs/choice
 J Knowl Econ

intention. Finally, our study adds to the existing evidence for the expectancy-value
model of attitude justifying the suggestion that both types of measures (direct and
indirect) should be used in TPB questionnaires, as by doing so, it will be possible to
explain more variance in the model (Francis et al. 2004b).
The study’s findings have implications for theory and practice. At a theoretical level,
our work highlights the emotional aspect of entrepreneurship process showing that
emotions play an important role in determining and regulating entrepreneurs’ cogni-
tions, behaviors, and venture efforts (Cardon et al. 2009; Foo et al. 2009). Thus, our
study suggests that in order to effectively increase entrepreneurial intention, it is
important to target both the cognitive and affective bases of attitudes toward entrepre-
neurship (Zampetakis et al. 2009).
Our study has also implications for educators and policy makers. Programs to
enhance and develop traits that are associated with entrepreneurial success and provide
skills that entrepreneurs will need latter, with the aim of encouraging young people to
start their own businesses, have been spreading over the last decades at a considerable
pace and courses are being implemented in universities, secondary schools, and even
primary ones (do Paço et al. 2011; Liñán 2004). The European Commission notes that
the “primary purpose of entrepreneurship education [at higher education level] is to
develop entrepreneurial capacities and mindsets” (European Commission 2008: 11)
and recommends integrating entrepreneurship more fully into university curricula. The
results from this study suggest that entrepreneurship education should not be confused
with general business and economic studies and focus only on the learning of entre-
preneurial competencies, such as business planning; entrepreneurship education should
always include affective learning goals and develop personal attributes and skills that
form the basis of an entrepreneurial mindset and behavior (creativity, sense of initiative,
autonomy, etc.). (European Commission 2008; Müller 2011).
On a policy level, the results obtained here suggest that a government should create
and support an environment that fosters entrepreneurship through less restrictive
legislation and regulations, lower bureaucratic barriers for engaging in self-employ-
ment, and sufficient access to credit for enterprises, because this would provide
individuals with added procedural utility resulting from the individual’s sense of control
over career (Benz and Frey 2008).
However, this study is not without its limitations which need to be brought to the
reader’s attention. First, the study relies on student participants who were drawn from
two higher education institutions, located in two neighboring cities of Central Greece.
However, although the sample size (N=227) was relatively small, it was greater than
the suggested for SEM minimum sample size of 200. In addition, the number of cases
per estimated parameter, equal to 6.14 for the model with the greater number of
parameters in Fig. 2, falls within the typical range from 5 to 20 (Brown 2006;
Jackson 2003; Weston and Gore 2006). Therefore, caution must be applied in
interpreting our results and generalizing the findings to other cohorts of students.
Future research may explore the hypotheses put forth in this study with a larger and
more representative sample of students from various disciplines and different Greek
higher education institutions, or even with respondents other than students. Replication
of this study with such a sample would add weight to the reported results.
Another limitation of the study is its cross-sectional design, which limits the
assessment of causality. Attitude theory researchers have long argued that cognition
J Knowl Econ

and affect interact in a reciprocal process in attitude formation and that instrumental
attitude and affective attitude are both a cause and an effect of each other (Fernandes
and Proença 2013). In order to establish a causal relationship between these two
components of attitude, longitudinal data are needed.
Future research may also attempt to include beliefs of the outcomes of being organiza-
tionally employed along with the beliefs for entrepreneurial outcomes, as by disentangling
the reasons for preferring self-employment from those of organizational employment; a
better understanding of the former can be achieved (Guerra and Patuelli 2013).

Acknowledgments The authors would like to thank the two anonymous reviewers for their constructive and
supportive comments that helped improve the quality of the paper.

References

Ajzen, I. (1985). From intentions to actions: a theory of planned behavior. In J. Kuhl & J. Beckman (Eds.),
Action-control: from cognition to behaviour (pp. 11–39). Heidelberg: Springer.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes,
50(2), 179–211.
Ajzen, I. (2002). Perceived behavioral control, self-efficacy, locus of control, and the theory of planned
behavior. Journal of Applied Social Psychology, 32(4), 665–683.
Ajzen, I. (2006). Constructing a TPB questionnaire: conceptual and methodological considerations. Available
at: http://www.people.umass.edu/aizen/pdf/tpb.measurement.pdf. Accessed 4 October 2013.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs:
Prentice Hall.
Arbuckle, J. L. (2012). Amos 21 user’s guide. Spring House: Amos Development Corporation.
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behavior: a meta-analytic review.
British Journal of Social Psychology, 40(4), 471–499.
Aspaas, H. R. (2004). Minority women’s microenterprises in rural area of the United States of America:
African American, Hispanic American and Native American case studies. GeoJournal, 61(3), 281–289.
Autio, E., Keeley, R. H., Klofsten, M., Parker, G. G. C., & Hay, M. (2001). Entrepreneurial intent among
students in Scandinavia and in the USA. Enterprise and Innovation Management Studies, 2(2), 145–160.
Bagozzi, R. P., Baumgartner, J., & Yi, Y. (1989). An investigation into the role of intentions as mediators of the
attitude-behavior relationship. Journal of Economic Psychology, 10(1), 35–62.
Bénabou, R., & Tirole, J. (2003). Intrinsic and extrinsic motivation. Review of Economic Studies,
70(3), 489–520.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238–246.
Benz, M., & Frey, B. S. (2008). Being independent is a great thing: subjective evaluations of self-employment
and hierarchy. Economica, 75(298), 362–383.
Bird, B. (1988). Implementing entrepreneurial ideas: the case for intention. Academy of Management Review,
13(3), 442–453.
Bird, B. (1992). The operation of intentions in time: the emergence of the new venture. Entrepreneurship:
Theory and Practice, 17(1), 11–20.
Blanchflower, D. G., Oswald, A., & Stutzer, A. (2001). Latent entrepreneurship across nations. European
Economic Review, 45(4–6), 680–691.
Boyd, N., & Vozikis, G. (1994). The influence of self-efficacy on the development of entrepreneurial
intentions and actions. Entrepreneurship: Theory and Practice, 18(4), 63–77.
Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York: Guilford Press.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long
(Eds.), Testing structural equation models (pp. 136–162). Newsbury Park: Sage.
Byrne, B. M. (2001). Structural equation modeling with AMOS: basic concepts, applications and program-
ming. Mahwah: Lawrence Erlbaum Associates, Inc.
Cardon, M. S., Wincent, J., Singh, J., & Drnovsek, M. (2009). The nature and experience of entrepreneurial
passion. Academy of Management Review, 34(3), 511–532.
 J Knowl Econ

Carter, N. M., Gartner, W. B., Shaver, K. G., & Gatewood, E. J. (2003). The career reasons of nascent
entrepreneurs. Journal of Business Venturing, 18(1), 13–39.
Chell, E. (2001). Entrepreneurship: globalization, innovation and development. Stamford: Thomson Learning.
Chen, F. F., Sousa, K. H., & West, S. G. (2005). Testing measurement invariance of second-order factor
models. Structural Equation Modeling, 12(3), 471–492.
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement
invariance. Structural Equation Modeling, 9(2), 233–255.
Croson, D. C., & Minniti, M. (2012). Slipping the surly bonds: the value of autonomy in self-employment.
Journal of Economic Psychology, 33(2), 355–365.
Damasio, A. R. (1994). Descartes’ error: emotion, reason, and the human brain. New York: Putnam.
Davidsson, P. (1995). Determinants of entrepreneurial intentions. Paper presented at the RENT IX
Conference, 23–24 November, 1995, Piacenza Italy.
Dawson, C., Henley, A., & Latreille, P. (2009). Why do individuals choose self-employment? IZA Discussion
Paper, No. 3974. Available at: http://ftp.iza.org/dp3974.pdf. Accessed 29 November 2013.
Deci, E., & Ryan, R. (2000). The “what” and “why” of goal pursuits: human needs and the self-determination
of behavior. Psychological Inquiry, 11(4), 227–268.
do Paço, A. M. F., Ferreira, J. M., Raposo, M., Rodrigues, R. G., & Dinis, A. (2011). Behaviors and
entrepreneurial intention: empirical findings about secondary students. Journal of International
Entrepreneurship, 9(1), 20–38.
Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth: Harcourt Brance Javanovich.
Elfving, J. (2008). Contextualizing entrepreneurial intentions: a multiple case study on. entrepreneurial
cognition and perception. Turku: Åbo Akademi Förlag.
Engle, R. L., Dimitriadi, N., Gavidia, J. V., Schlaegel, C., Delanoe, S., Alvarado, I., He, X., Buame, S., &
Wollf, B. (2010). A twelve-country evaluation of Ajzen’s model of planned behavior”. International
Journal of Entrepreneurial Behavior and Research, 16(1), 33–57.
European Commission (2008). European Commission Expert Group: entrepreneurship in higher education,
especially within non-business studies. Available at: http://ec.europa.eu/enterprise/ newsroom/cf/
itemshortdetail.cfm?item_id=3366. Accessed 30 November 2009.
Evans, B., Glendon, A. I., & Creed, P. A. (2007). Development and initial validation of an aviation safety
climate scale. Journal of Safety Research, 38(6), 675–682.
Fayolle, A. (2006). Effect and counter effect of entrepreneurship education and social context on student’s
intention. Estudios de Economica Aplicada, 24(2), 509–523.
Fayolle, A. (2007). Handbook of research in entrepreneurship education: a general perspective. Cheltenham:
Edward Elgar.
Fernandes, T., & Proença, J. (2013). Reassessing relationships in consumer markets: emotion, cognition, and
consumer relationship intention. Journal of Relationship Marketing, 12(1), 41–58.
Fini, R., Grimaldi, R., Marzocchi, G. L., & Sobrero, M. (2012). The determinants of corporate entrepreneurial
intention within small and newly established firms. Entrepreneurship: Theory and Practice, 36(2), 387–414.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: an introduction to theory and
research. Reading: Addison-Wesley.
Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: the reasoned action approach. New
York: Taylor and Francis.
Foo, M.-D., Uy, M. A., & Baron, R. A. (2009). How do feelings influence effort? An empirical study of
entrepreneurs’ affect and venture effort. Journal of Applied Psychology, 94(4), 1086–1094.
Francis, J. J., Eccles, M. P., Johnston, M., Walker, A., Grimshaw, J., Foy, R., Kaner, E. F. S., Smith, L., &
Bonetti, D. (2004a). Constructing questionnaires based on the theory of planned behaviour: a manual for
health services researchers. Newcastle: Centre for Health Services Research, University of Newcastle.
Francis, J. J., Eccles, M. P., Johnston, M., Walker, A., Grimshaw, J., Foy, R., Kaner, E. F. S., Smith, L., &
Bonetti, D. (2004b). Measurement issues in the theory of planned behaviour: a supplement to the manual
for constructing questionnaires based on the theory of planned behaviour. Newcastle: Centre for Health
Services Research, University of Newcastle.
French, D. P., & Hankins, M. (2003). The expectancy-value muddle in the theory of planned behaviour—and
some proposed solutions. British Journal of Health Psychology, 8(1), 37–55.
French, D. P., Sutton, S. R., Hennings, S. J., Mitchell, J., Wareham, N. J., Griffin, S., Hardeman, W., &
Kinmonth, A. L. (2005). The importance of affective beliefs and attitudes in the theory of planned behavior:
predicting intentions to increase physical activity. Journal of Applied Social Psychology, 35(9), 1824–1848.
Gassar, C. (2006). Opportunity costs and intended venture growth. Journal of Business Venturing, 21(5), 610–632.
Green, D. P., Salovey, P., & Truax, K. M. (1999). Static, dynamic, and causative bipolarity of affect. Journal of
Personality and Social Psychology, 76(5), 856–867.
J Knowl Econ

Grilo, I., & Thurik, A. R. (2005). Entrepreneurial engagement levels in the European Union. International
Journal of Entrepreneurship Education, 2, 143–168.
Grilo, I., & Thurik, A. R. (2008). Determinants of entrepreneurial engagement levels in Europe and the US.
Industrial and Corporate Change, 17(6), 1113–1145.
Guerra, G., & Patuelli, R. (2013). The role of job satisfaction in transitions into self-employment. Quaderni
DSE Working Paper N° 849. Available at: http://papers.ssrn.com /sol3/papers.cfm?abstract_id=2151872.
Accessed 11 July 2014.
Gundry, L. K., & Welch, H. P. (2001). The ambitious entrepreneur: high growth strategies of women-owned
enterprises. Journal of Business Venturing, 16(5), 453–470.
Hagenaars, J. A., & McCutcheon, A. L. (2002). Applied latent class analysis. Cambridge: Cambridge
University Press.
Hales, D., Evenson, K. R., Wen, F., & Wilcox, S. (2010). Postpartum physical activity: measuring theory of
planned behavior constructs. American Journal of Health Behavior, 34(4), 387–401.
Hamilton, B. H. (2000). Does entrepreneurship pay? An empirical analysis of the returns to self‐employment.
Journal of Political Economy, 108(3), 604–631.
Hausenblas, H. A., Carron, A. V., & Mack, D. E. (1997). Application of the theories of reasoned action and
planned behavior to exercise behavior: a meta-analysis. Journal of Sport and Exercise Psychology, 19(1),
36–51.
Hayek, M. (2012). Control beliefs and positive psychological capital. Journal of Management Research,
13(1), 3–13.
Heath, R. G., & Nairn, A. C. (2005). Measuring emotive advertising—implications of low attention process-
ing on recall. Journal of Advertising Research, 45(2), 269–281.
Hindle, K., Klyver, K., & Jennings, D. F. (2009). An ‘informed’ intent model: incorporating human capital,
social capital and gender variables into the theoretical model of entrepreneurial intentions. In A. Carsrud
& M. Brännback (Eds.), Understanding the entrepreneurial mind: opening the black box (pp. 35–50).
Heidelberg: Springer.
Holland, D. V., & Shepherd, D. A. (2013). Deciding to persist: adversity, values, and entrepreneurs’ decision
policies. Entrepreneurship: Theory and Practice, 37(2), 331–358.
Horvath, C., Lewis, I., & Watson, B. (2012). The beliefs which motivate young male and female drivers to
speed: a comparison of low and high intenders. Accident; Analysis and Prevention, 45, 334–341.
Hu, L., & Bentler, P. M. (1999). Cut-off criteria for fit indexes in covariance structure analysis: conventional
criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55.
Iakovleva, T., Kolvereid, L., & Stephan, U. (2011). Entrepreneurial intentions in developing and developed
countries. Education and Training, 53(5), 353–370.
Isen, A., & Reeve, J. (2005). The influence of positive affect on intrinsic and extrinsic motivation: facilitating
enjoyment of play, responsible work behavior, and self-control. Motivation and Emotion, 29(4), 295–323.
Jackson, D. L. (2003). Revisiting sample size and the number of parameter estimates: some support for the n:q
hypothesis. Structural Equation Modeling: A multidisciplinary Journal, 10(1), 128–141.
Jennrich, R. I., & Sampson, P. F. (1966). Rotation for simple loadings. Psychometrika, 31(3), 313–323.
Johnston, V. S. (1999). Why we feel: the science of emotions. Reading: Helix Books.
Joreskog, K. G., & Sorbom, D. (1988). LISREL 7: a guide to the program and applications. Chicago: SPSS, Inc.
Kaiser, H. F. (1960). The application of electronic computer to factor analysis. Educational and Psychological
Measurement, 20(1), 141–151.
Kaplan, D. (2000). Structural equation modeling: foundations and extensions. London: Sage.
Kato, S. (2013). Entrepreneurship as a process of self-fulfillment: well-being, affect, and behavioral strategies.
PhD dissertation, Syracuse University.
Keer, M. (2012). Persuasion through facts and feelings: integrating affect and cognition into behavioral
decision models and health messages. PhD thesis, Universiteit van Amsterdam.
Kennedy, J., Drennan, J., Renfrow, P., & Watson, B. (2003, September). Situational factors and entrepre-
neurial intentions. Paper presented at the Small Enterprise Association of Australia and New Zealand 16th
Annual Conference, 28 September-1 October, 2003, Ballarat, Australia.
Kim, M. S., & Hunter, J. E. (1993). Relationships among attitudes, behavioral intentions, and behavior: a
meta-analysis of past research, part 2. Communication Research, 20(3), 331–364.
Kivetz, R. (2005). Promotion reactance: the role of effort-reward congruity. Journal of Consumer Research,
31(4), 725–736.
Kline, R. B. (2010). Principles and practice of structural equation modelling (3rd ed.). New York: The
Guildford Press.
Kolvereid, L. (1996a). Organizational employment versus self-employment: reasons for career choice inten-
tions. Entrepreneurship: Theory and Practice, 20(3), 23–31.
 J Knowl Econ

Kolvereid, L. (1996b). Prediction of employment status choice intentions. Entrepreneurship: Theory and
Practice, 21(1), 47–57.
Kolvereid, L., & Isaksen, E. (2006). New business start-up and subsequent entry into self-employment.
Journal of Business Venturing, 21(6), 866–885.
Kraft, P., Rise, J., Sutton, S., & Roysamb, E. (2005). Perceived difficulty in the theory of planned
behaviour: perceived behavioural control or affective attitude? British Journal of Social Psychology,
44(3), 479–496.
Krueger, N. F. (1993). The impact of prior entrepreneurial exposure on perceptions and new venture feasibility
and desirability. Entrepreneurship: Theory and Practice, 18(1), 5–21.
Krueger, N. F., & Brazeal, D. (1994). Entrepreneurial potential and potential entrepreneurs. Entrepreneurship:
Theory and Practice, 18(3), 91–104.
Krueger, N. F., Reilly, M. D., & Carsrud, A. L. (2000). Competing models of entrepreneurial intentions.
Journal of Business Venturing, 15(5–6), 411–432.
Kuratko, D. F., Hornsby, J. S., & Naffziger, D. W. (1997). An examination of owners’ goals in sustaining
entrepreneurship. Journal of Small Business Management, 35(1), 24–33.
LaBarge, M. C. (2007). Integral affect and attitude strength in health communications. PhD dissertation,
University of Oregon.
Lange, T. (2012). Job satisfaction and self-employment: autonomy or personality? Small Business Economics,
38(2), 165–177.
Lawton, R., Conner, M., & Parker, D. (2007). Beyond cognition: predicting health risk behaviors from
instrumental and affective beliefs. Health Psychology, 26(3), 259–267.
Leroy, H., Maes, J., Sels, L., & Kickul, J. R. (2010, August). Gender differences in entrepreneurial intentions:
a TPB multi-group analysis at factor and indicator level. Paper presented at the 2010 DIANA
International Conference on Women’s Entrepreneurship Research, 3–4 August, 2010, Banff, Canada.
Liñán, F. (2004). Intention-based models of entrepreneurship education. Piccolla Impresa/Small Business,
3(1), 11–35.
Liñán, F., & Chen, Y.-W. (2009). Development and cross-cultural application of a specific instrument to
measure entrepreneurial intentions. Entrepreneurship: Theory and Practice, 33(3), 593–617.
Liñán, F., Nabi, G., & Krueger, N. (2013). British and Spanish entrepreneurial intentions: a comparative study.
Revista de Economía Mundial, 33, 73–103.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: exploring
the question, weighing the merits. Structural Equation Modeling, 9(2), 151–173.
Magson, N. R., Bodkin-Andrews, G. H., Craven, R. G., Nelson, G. F., & Yeung, A. S. (2013). Questioning
new directions in understanding student motivation: an investigation into the domain specificity of
motivational goals. The Australian Educational and Developmental Psychologist, 30(2), 171–190.
Marsh, H. W., & Hau, K. T. (1996). Assessing goodness of fit: is parsimony always desirable? Journal of
Experimental Education, 64(4), 364–390.
Marsh, H. W., & Hovecar, D. (1985). Application of confirmatory factor analysis to the study of self-concept:
first-and higher order factor models and their invariance across groups. Psychological Bulletin, 97(3),
562–582.
Marsh, H. W., Roche, L. A., Pajares, F., & Miller, D. (1997). Item-specific efficacy judgments in mathematical
problem solving: the downside of standing too close to trees in a forest. Contemporary Educational
Psychology, 22(3), 363–377.
Martin, B. C. (2012). Entrepreneurship as a means of improving the social and economic conditions of
persons with disabilities. Unpublished PhD dissertation, McMaster University.
Masuda, T. (2006). The determinants of latent entrepreneurship in Japan. Small Business Economics, 26(3),
227–240.
Mather, R. D., & Romo, A. (2007). Automaticity and cognitive control in social behavior. Southlake:
Fountainhead.
McGee, J. E., Peterson, M., Mueller, S. L., & Sequeira, J. M. (2009). Entrepreneurial self-efficacy: refining the
measure. Entrepreneurship: Theory and Practice, 33(4), 965–988.
Moriano, J. A., Gorgievski, M., Laguna, M., Stephan, U., & Zarafshani, K. (2012). A cross-cultural approach
to understanding entrepreneurial intention. Journal of Career Development, 39(2), 162–185.
Müller, S. (2011). Increasing entrepreneurial intention: effective entrepreneurship course characteristics.
International Journal of Entrepreneurship and Small Business, 13(1), 55–74.
O’Connell, A. A., & McCoach, D. B. (2008). Multilevel modeling of educational data. Charlotte: IAP.
Pardo, C., & Ruiz-Tagley, J. (2011). The dynamic role of specific experience in the selection of self-
employment versus wage-employment. Working Paper, Chile: Mimeo, Centro de Microdatos,
Universidad de Chile.
J Knowl Econ

Pelling, E. L., & White, K. M. (2009). The theory of planned behavior applied to young people’s use of social
networking websites. Cyberpsychology and Behavior, 12(6), 755–759.
Petkova, A. P. (2009). A theory of entrepreneurial learning from performance errors. International
Entrepreneurship Management Journal, 5(4), 345–367.
Pruett, Μ., Shinnar, R., Toney, Β., Llopis, F., & Fox, J. (2009). Explaining entrepreneurial intentions of
university students: a cross-cultural study. International Journal of Entrepreneurial Behaviour and
Research, 15(6), 571–594.
Reynolds, P. D. (1997). Who starts new firms? Preliminary explorations of firms-in-generation. Small Business
Economics, 9(5), 449–462.
Rhodes, R. E., & Courneya, K. S. (2003). Relationships between personality, an extended theory of planned
behaviour and exercise behaviour. British Journal of Health Psychology, 8(1), 19–36.
Robichaud, Y., McGraw, E., & Roger, A. (2001). Toward the development of a measurement instrument for
entrepreneurial motivation. Journal of Developmental Entrepreneurship, 6(2), 189–202.
Schlaegel, C., & Koenig, M. (2014). Determinants of entrepreneurial intent: a meta-analytic test and
integration of competing models. Entrepreneurship: Theory and Practice, 38(2), 291–332.
Schwarz, N., & Bohner, G. (2001). The construction of attitudes. In A. Tesser & N. Schwarz (Eds.),
Intrapersonal processes: Blackwell handbook of social psychology (pp. 436–457). Oxford:
Blackwell.
Schwarz, E. J., Wdowiak, M. A., Almer-Jarz, D. A., & Breitenecker, R. J. (2009). The effects of attitudes and
perceived environment conditions on students’ entrepreneurial intent: an Austrian perspective. Education +
Training, 51(4), 272–291.
Shapero, A., & Sokol, L. (1982). The social dimensions of entrepreneurship. In C. A. Kent, D. L. Sexton, & K.
H. Vesper (Eds.), Encyclopedia of entrepreneurship (pp. 72–90). Englewood Cliffs: Prentice Hall.
Sheeran, P. (2002). Intention–behavior relations: a conceptual and empirical review. In W. Stroebe & M.
Hewstone (Eds.), European review of social psychology Vol. 12 (pp. 1–36). Chichester: Wiley.
Souitaris, V., Zerbinati, S., & Al-Laham, A. (2007). Do entrepreneurship programmes raise entrepreneurial
intention of science and engineering students? The effect of learning, inspiration and resources. Journal of
Business Venturing, 22(4), 566–591.
Stam, E., Bosma, N., Van Witteloostuijn, A., De Jong, J., Bogaert, S., Edwards, N., & Jaspers, F. (2012).
Ambitious entrepreneurship. A review of the academic literature and new directions for public policy.
Report for the advisory council for science and technology policy (AWT) and the Flemish council for
science and innovation (VRWI). The Hague: Advisory Council for Science and Technology Policy
(AWT).
Sutton, S. (1998). Predicting and explaining intentions and behavior: how well are we doing? Journal of
Applied Social Psychology, 28(15), 1317–1338.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: HarperCollins.
Tegtmeier, S. (2012). Empirical implications for promoting students’ entrepreneurial intentions. Journal of
Enterprising Culture, 20(2), 151–169.
ten Klooster, P. M. (2008). Interpreting patient reported outcomes in rheumatology. PhD. dissertation,
University of Twente.
Thompson, E. K. (2009). Individual entrepreneurial intent: construct clarification and development of an
internationally reliable metric. Entrepreneurship: Theory and Practice, 33(3), 669–694.
Trafimow, D., & Sheeran, P. (1998). Some tests on the distinction between cognitive and affective beliefs.
Journal of Experimental Social Psychology, 34(4), 378–397.
Trafimow, D., Sheeran, P., Lombardo, B., Finlay, K. A., Brown, J., & Armitage, C. J. (2004). Affective and
cognitive control of persons and behaviours. British Journal of Social Psychology, 43(2), 207–224.
Truong, Y. (2009). An evaluation of the theory of planned behavior in consumer acceptance of online video
and television services. The Electronic Journal Information Systems Evaluation, 12(2), 177–186.
Tucker, L. R., & Lewis, C. (1973). A reliability coefficient for maximum likelihood factor analysis.
Psychometrika, 38(1), 1–10.
van der Heijden, H. (2003). Factors influencing the usage of websites: the case of a generic portal in The
Netherlands. Information and Management, 40(6), 541–549.
van der Zwan, P., Thurik, A. R., & Grilo, I. (2010). The entrepreneurial ladder and its determinants. Applied
Economics, 42(17), 2183–2191.
van Gelderen, M., Thurik, A. R., & Bosma, N. (2005). Success and risk factors in the pre-start up phase. Small
Business Economics, 24(4), 365–380.
van Gelderen, M. W., Brand, M., van Praag, M., Bodewes, W., Poutsma, E., & van Gils, A. (2008). Explaining
entrepreneurial intentions by means of the theory of planned behaviour. Career Development
International, 13(6), 538–559.
 J Knowl Econ

Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature:
suggestions, practices, and recommendations for organizational research. Organizational Research
Methods, 3(1), 4–69.
Verheul, I., Thurik, A. R., Grilo, I., & van der Zwan, P. (2012). Explaining preferences and actual involvement
in self-employment: gender and the entrepreneurial personality. Journal of Economic Psychology, 33(2),
325–341.
Weston, R., & Gore, P. A. (2006). A brief guide to structural equation modeling. The Counseling Psychologist,
34(5), 719–751.
Wiklund, J., Davidsson, P., & Delmar, F. (2003). What do they think and feel about growth? An expectancy–
value approach to small business managers’ attitudes towards growth. Entrepreneurship: Theory and
Practice, 27(3), 247–270.
Xu, M. K., Marsh, H. W., Hau, K.-T., Ho, I. T. F., Morin, A. J. S., & Abduljabbar, A. S. (2013). The internal/
external frame of reference of academic self-concept: extension to a foreign language and the role of
language of instructio. Journal of Educational Psychology, 105(2), 489–503.
Yan, Z. (2014). Predicting teachers’ intentions to implement school-based assessment using the theory of
planned behaviour. Educational Research and Evaluation: An International Journal on Theory and
Practice, 20(2), 83–97.
Zampetakis, L. A., Kafetsios, K., Bouranta, N., Dewett, T., & Moustakis, V. S. (2009). On the relationship
between emotional intelligence and entrepreneurial attitudes and intentions. International Journal of
Entrepreneurial Behaviour and Research, 15(6), 595–618.