Success And Risk Factors In The Pre Startup Phase

Description
In this particular brief criteria in regard to success and risk factors in the pre startup phase.

Success and Risk Factors in
the Pre-Startup Phase
Marco van Gelderen
Roy Thurik
Niels Bosma
ABSTRACT. Why does one person actually succeed in
starting a business, while a second person gives up? In
order to answer this question, a sample of 517 nascent
entrepreneurs (people in the process of setting up a busi-
ness) was followed over a three-year period. After this
period, it was established that 195 e?orts were successful
and that 115 startup e?orts were abandoned. Our research
focuses on estimating the relative importance of a variety
of approaches and variables in explaining pre-startup suc-
cess. These in?uences are organized in terms of Gartner’s
(Academy of Management Review 10(4), 696–706 [1985])
framework of new venture creation. This framework sug-
gests that start-up e?orts di?er in terms of the characteris-
tics of the individual(s) who start the venture, the
organization that they create, the environment surround-
ing the new venture, and the process by which the new
venture is started. Logistic regression analyses are run for
the sample as a whole as well as for subgroups within the
sample, namely for those with high ambition vs. low
ambition and for those with substantial vs. limited experi-
ence. The results point to the importance of perceived risk
of the market as a predictor of getting started vs. aban-
doning the startup e?ort.
KEY WORDS: performance, survival, nascent entrepre-
neurs, start-ups
JEL CODE: M13
1. Introduction
The ?rst success of a ?rm is its birth. A signi?-
cant portion of those attempting to establish a
business fails. In this paper the person undertak-
ing activities to create a business is referred to as
the nascent entrepreneur, and the founding e?ort
is called nascent entrepreneurship (Reynolds and
White, 1992). Relatively few attempts have been
made to study nascent entrepreneurship empiri-
cally. One important reason is the lack of a rep-
resentative sample: nascent entrepreneurs are
unregistered, which makes them di?cult to sam-
ple in comparison to small business owners (Rey-
nolds, 1997). As a consequence, many questions
about nascent entrepreneurship remain unan-
swered. One question is addressed in this paper:
Which factors contribute to success or failure in
starting a business? This question is vital for
several stakeholders.
First, people considering starting a business
have an interest in kowledge about factors that
contribute to success or failure in the pre-start-up
phase. Armed with this knowledge, they can eval-
uate their own prospects and potential pitfalls.
Second, knowledge of the behavior of nascent
entrepreneurs is important for those involved in
creating and maintaining policy measures on a
macro-economic level. A high level of entrepre-
neurial activity has been shown to contribute to
innovative activities, competition, economic
growth and job creation (Carree and Thurik,
2003). Promotion of entrepreneurship can bene?t
Final version accepted on November 2004
Marco van Gelderen
Department of Management and International Business
Albany Campus
Massey University
Private Bag 102
904 North Shore MSC
Auckland
New Zealand
E-mail: [email protected]
Roy Thurik
Rotterdam School of Economics
Erasmus University Rotterdam
PO box 1738 Rotterdam
The Netherlands
E-mail: [email protected]
and EIM Business and Policy Research Zoetermeer and
Max Planck Institute for Research into Economic Systems
Jena
Niels Bosma
EIM Business and Policy Research PO box 7001
2701 AA Zoetermeer
The Netherlands
E-mail: [email protected].
Small Business Economics (2005) 24: 365–380 Ó Springer 2005
DOI 10.1007/s11187-004-6994-6
from insight into the factors that contribute to
success or failure in the pre-start-up phase.
Third, there is a gap in scienti?c knowledge con-
cerning this issue. The study of success in the
pre-start-up phase borders on two large streams
of entrepreneurship research. The ?rst stream
consists of comparisons between entrepreneurs
and non-entrepreneurs. The second stream con-
cerns comparisons between successful entrepre-
neurs and less successful entrepreneurs. The
study of success in the pre-start-up phase is a
mixture of both.
2. Models
While empirical work on success and risk factors
in nascent entrepreneurship is scarce, there is an
abundance of conceptual work modeling (parts
of) the pre-start-up process (e.g. Bhave, 1994;
Busenitz and Lau, 1996; Greenberger and Sexton,
1988; Herron and Sapienza, 1992; Johnson, 1990;
Kamm and Nurick, 1993; Larson and Starr,
1993; Learned, 1992; Na?ziger, et al., 1994; Starr
and Fondas, 1992; Vanderwerf, 1993). Some
models are based on a single approach, such as a
motivational model (Na?ziger et al., 1994), a cog-
nitive model (Busenitz and Lau, 1996), or a net-
work model (Larson and Starr, 1993). Most
models build on a variety of approaches. Usually,
there is also a temporal aspect to the models.
Some authors describe the process of setting up a
business as entailing the execution of a number of
activities, with high variation in the sequence and
amount of activities (Carter et al., 1995; Reynolds
and Miller, 1992). While acknowledging this vari-
ation, some authors still discern sub-phases in the
pre-start-up process (Bhave, 1994; Kamm and
Nurick, 1993).
Four phases are often mentioned. The ?rst
phase concerns the development of an intention to
start an enterprise (Krueger et al., 2000; Shapero
and Sokol, 1982). In the second phase an entrepre-
neurial opportunity is recognized and a business
concept is developed. In the third phase resources
are assembled and the organization is created. In
the ?nal phase the organization starts to exchange
with the market. Nascent entrepreneurship is con-
sidered the active pursuit of organization creation
(phases two and three); therefore criteria are
needed to demarcate nascent entrepreneurship
from the ?rst phase (potential entrepreneurs), and
from the fourth phase (starting entrepreneurs).
This is a thorny issue that is discussed in the
method section.
Given the scarcity of empirical work on suc-
cess and risk factors in nascent entrepreneurship,
our research is exploratory. Success factors in
phases one and four cannot be considered evi-
dence for success in phases two and three. A suc-
cess factor in one phase might very well be a
failure factor in another phase. For example,
Tiessen (1998) argues that individualistic tenden-
cies are conducive to intentions towards self-
employment, but interfere with the process of
resource acquisition where active cooperation
with other people is vital. Also, some variables
may be more important in one phase and less
important in another phase. For example, the
psychology of the entrepreneur has been found
to be more important in predicting the chances
to start a business than in predicting the chances
of the success of a business (Rauch and Frese,
2000). Although not investigated empirically in
this study, success and risk factors might even
vary between phases two and three. Our study
will be guided by the conceptual work mentioned
above. The variety of approaches and variables
that are possible in?uences will be organized in
terms of Gartner’s (1985) framework of new ven-
ture creation. This framework suggests that start-
up e?orts di?er in terms of the characteristics of
the individual(s) who start the venture, the orga-
nization which they create, the environment sur-
rounding the new venture, and the process by
which the new venture is started. We derive pos-
sible success and risk factors from each of these
dimensions.
3. Approaches
This section describes the four main approaches
in Gartner’s framework – the individual, the
environment, the process and the organization, –
in more detail. Approaches that concern the indi-
vidual can be divided into two types of variables:
human capital and psychological individual dif-
ferences. Human capital variables include knowl-
edge, education, skills and experience (Deakins
and Whittam, 2000). Human capital variables are
likely to in?uence the development of a business
366 Marco van Gelderen et al.
idea and the organization of resources. For
example, start-up experience provides the nascent
entrepreneur with learning opportunities that can
be exploited; work experience provides skills that
might function in the accomplishment of the
many tasks that setting up a business entails;
industry experience can be helpful in the percep-
tion and valuation of new business ideas.
Psychological individual di?erences concern dif-
ferences in personality characteristics, cognitive
characteristics, and motivational patterns. Research
on personality characteristics relates dispositions
such as risk-taking, locus of control, and need for
achievement to the emergence and the success of
entrepreneurship (for an overview, see Rauch and
Frese, 2000). These characteristics might also in?u-
ence success in the pre-start-up phase.
In cognitive approaches, the manner in which
individuals process information is central. Cogni-
tive characteristics concern individual di?erences
in attributions and in perceptions. Di?erences in
attributions concern how people explain events
or outcomes of events. Di?erences in perceptions
concern the study of how people perceive them-
selves or their environment. Cognitive psychology
repeatedly shows that people are not fully rational
but rather make extensive use of heuristics, result-
ing in cognitive biases (Kahneman et al., 1982).
Entrepreneurs have been shown to be prone espe-
cially to cognitive biases, enabling them to con?-
dently take risks (but not perceiving these risks as
such) (Simon et al., 2000). Other examples of cogni-
tive variables that have been found to distinguish
entrepreneurs from non-entrepreneurs are perceived
self-e?cacy (Chen et al., 1998), and counterfactual
thinking (thinking about what-might-have-been)
(Baron, 1999) (for an overview see Baron, 2004).
Finally, di?erences in motivation might in?u-
ence success in the pre-start-up phase. People
have di?erent motives for setting up a business.
Gatewood et al. (1995) have studied di?erences
in motives as a success factor in nascent entrepre-
neurship. They ?nd that women who start for
internally oriented reasons (such as need for auton-
omy), and men who start for externally oriented
reasons (like perceiving a need in the market) have
greater chances of successfully completing the pre-
start-up phase. Another common distinction is
between push and pull motives, push motives being
reasons that force people into entrepreneurship
(such as lack of alternatives), and pull motives
being reasons that attract people to entrepreneurship
(e.g. challenge or autonomy). Economic approaches
suggest that entrepreneurial motivation is based
on the di?erence in expected utility between self-
employment and organizational employment
Campbell, 1992).
Approaches that take the environment into
account can be divided into network, ?nancial,
and ecological approaches. In network approaches
the emphasis is on relationships between people.
Ties can di?er in diversity and emotional strength.
Diversity of ties means that one knows people
that do not know each other. Emotional strength
can vary from strong to weak (Aldrich, 1999).
Aldrich expects successful nascent entrepreneurs
to have a diverse network with many strong ties.
Such a network is important, as an individual
does not set up a ?rm solely by himself or herself.
In the opportunity recognition and business idea
development phase, one depends upon the envi-
ronment for information; in the resource assem-
bling and organization phase, one depends upon
the environment for resources.
The ?nancial approach is concerned with the
sources and size of capital of the new ?rm. Most
?rms start out with a small amount of capital pro-
vided by the ?rm founder(s) (Aldrich, 1999). Lack
of funding might be a reason for nascent entrepre-
neurs to abandon the start-up process (Blanch?ower
and Oswald, 1998; Holtz-Eakin et al., 1994). There-
fore, a large amount of start-up capital to be pro-
vided by a bank or business angel might be risk
factor in nascent entrepreneurship. This is in contra-
diction to the post-start-up phase where it is usually
found to be a success factor.
In ecological approaches attention is given to
the environmental conditions that generate varia-
tions in the number of start-ups over time
(Aldrich, 1990). Within an industry, the carrying
capacity or muni?cence is an important variable
explaining success in the pre-start-up phase
(Specht, 1993). In an industry with many oppor-
tunities and many resources, chances of getting
started are relatively high. On the institutional
level, factors such as political turbulence, culture,
and the media, in?uence rates of organizational
emergence (Aldrich, 1990).
The ?nal two dimensions of Gartner’s frame-
work are the characteristics of the process and
367 Success and Risk Factors
the intended organization. With regard to process
it may matter how aggressively people pursue the
completion of start-up activities, whether they
work on their start-up e?ort full-time or part-time,
and whether they work with a business plan or
not. Carter et al. (1995) report that both individu-
als who start their business as well as individuals
who give up the start-up e?ort undertake more
activities to realize their business than people who
are still trying to set up their business. Therefore,
the authors recommend individuals considering a
business start-up to pursue opportunities aggres-
sively in the short term, in order not to ?nd them-
selves perpetually in the pre-startup phase.
With respect to the intended organization, the
nature of the opportunity is important, for exam-
ple regarding its degree of technological innova-
tion. Other examples of relevant variables are the
intended size of the ?rm, and whether there is
team or individual leadership. As this overview
indicates, there are many potentially relevant
in?uences. In the next session we will discuss
these approaches and dimensions in more detail
and explore their relationships with success and
failure in the pre-startup phase.
4. Exploring success and failure
Our research focuses on estimating the relative
importance of each of the approaches described
in the previous section in explaining perfor-
mance. However, while the approaches discussed
above conceptually explain pre-start-up success,
we will not make predictions at the level of the
particular variables measured due the lack of pre-
vious empirical work in this area. Instead, argu-
ments are given pro and contra the in?uence of
each variable as a success factor in the pre-start-
up phase. See Table I, for an overview of the
approaches and variables used in this study. Only
risk of the market can safely be assumed to be
negatively related to success in the pre-startup
phase. This applies both if risks are concrete or
only perceived. If the market is really risky,
chances of actually getting started are lower, as
the nascent entrepreneur will abort the startup
process when he learns that the prospects for his
?rm are poor. If the amount of risk is instead a
question of risk perception, then also a high
amount of perceived risk is indicative of failure,
as entrepreneurs are assumed to perceive less
risk. The relationship between experience vari-
ables and pre-startup success may be curvilinear,
as either a limited or an extended amount of
(work-, management-, industry-) experience
might prove to be harmful. In sum, our research
design is exploratory in establishing which vari-
ables are relevant for explaining success or failure
in the pre-startup phase.
5. Design, sample, variables, and analyses
The design of this study was developed by the
Entrepreneurial Research Consortium (ERC),
initiated and directed by Paul Reynolds. The
ERC is an international research e?ort (including
as participants among others the United States,
Sweden, Norway and The Netherlands) in which
each country investigates a random and represen-
tative sample of nascent entrepreneurs during
the start-up process. See Reynolds (2000), and
Reynolds et al., (2004) for details on the research
design. The data collection method of the ERC is
the general public survey. In the fall of 1998, a
random Dutch sample of 49,936 phone numbers
was dialed. An interview was held with 21,393
persons (43%) aged between 18 and 65 years.
The remaining 57% roughly consisted of refusals
(14.000), too young/too old (10.000), and other
(4,500). The person picking up the phone was
asked: ‘Are you, alone or with others, currently
setting up a business?’
If the person answers a?rmatively, two possi-
ble exclusions are made. First, it is essential to
have an active and manifest desire to set up a
business. If the respondent is only dreaming
about starting up a business, he or she is con-
sidered a potential entrepreneur instead of a
nascent entrepreneur. Persons indicating that
they have not yet undertaken activities yet in
pursuing their start-up are thus not included in
the sample. Second, someone who has set up a
business that is already operational, even though
in a start-up phase, is considered an entrepre-
neur instead of a nascent entrepreneur. The lat-
ter exclusion was not made in the initial
screening but rather in the follow up interview.
The question ‘Are you currently starting a busi-
ness?’ turned out to be quite ambiguous as a
number of people consider themselves still in a
368 Marco van Gelderen et al.
TABLE I
Variables representing the approaches and predicted sign of success (N = 271, initial measure)
Approach Variable Categories % N Arguments
Individual
Demographics Gender (n = 271) Female
Male
26
74
Females often face several
entry barriers, but may also
start less ambitious ventures
Age (n = 263) Age 18–24
Age 25–34
Age 35–44
Age 45–54
Age 55–64
7
41
35
14
3
More energy
More life experience
Human capital Work experience
(n = 256)
0–3 years
4–10 years
11–20 years
>20 years
5
33
38
24
More energy, less experience
More experience, but rigid?
Management experience
(n = 255)
0–1 year
2–5 years
6–10 years
>10 years
26
33
21
20
More energy, less experience
More experience, but rigid?
Experience in ?rm
founding (n = 271)
No
Yes
79
21
Experience of startup process,
but in case of previous failure
same ?aws might show
Education (n = 266) Low/middle education
High education
50
50
Higher educated seem
to have advantage, but also have
more alternative opportunities
Motivation Push motivation
(n = 307)
No push motivation
Push motivation
81
19
Push motivation better motivated
(a must) but may stop in case of
alternative employment
Ambition become
rich (n = 264)
To earn a living
To become rich
86
14
High materialistic expectations
can motivate but can also
be a source of disillusionment
Process
Business plan
(n = 307)
No business plan
business plan
40
60
Business plan seems to be
advantageous,
but experienced people as well as
very simple ?rms may not need
a business plan
Information
and guidance
(n = 255)
Makes no use of it
receives inf. and sup.
23
77
Information and guidance seems to be
advantageous, but experienced
people as well as simple ?rms may
not need support
Environment
Financial Third party money
(n = 254)
Only own money
makes a loan
59
41
With a loan better capitalization but
with own money obtaining ?nance is
no obstacle
Start-up capital
(n = 255)
0–10,000
10,001–50,000
50,001–200,000
>200,001
33
34
17
16
Easy to start
better capitalization
Network Industry experience
(n = 256)
0–1 year
2–5 years
6–10 years
>10 years
26
20
26
28
More energy, less experience
more experience, but rigid?
369 Success and Risk Factors
starting phase whereas their business is already
operational.
Those who in the initial wave state that they
are setting up a business, and who in the follow
up state that their business is operational and run-
ning, are asked to provide the startup date. The
follow up status assessment procedure is described
below. If the date is prior to the initial interview,
they are excluded from the sample (148 persons).
This set of protocols resulted in a sample of 517
nascent entrepreneurs (2,4% of the sample, which
indicates a prevalence rate of 2,4% within the
Dutch population between 18 and 65 years old).
This prevalence rate is comparable with Scandina-
vian countries but much lower than that in the
United States (Delmar and Davidsson, 2000). In
comparison with a control group (N ¼ 586) taken
from the 21.393 persons who state that they are
not currently setting up a business, the typical
nascent entrepreneur is male, young, has pursued
higher education and earns a higher income.
5.1. Dependent variable
Follow-up interviews were scheduled at a six
month, one year, two year and three year interval
(follow-up 1,2,3 and 4) after initial screening.
They included an assessment of the current status
of the start-up e?ort. Respondents were asked:
‘How would you classify your ?rm? Is it (1) oper-
ational and running; (2) are you still setting up
the business; (3) have you temporarily delayed
your start-up e?ort; or (4) have you completely
abandoned your start-up e?ort?’ After three
years, it could be established that 187 persons
had succeeded in starting their business, and that
105 persons had abandoned their start-up e?ort.
Table II presents ?gures for the number of start-
ups and abandoned e?orts that accrued during
the four assessment periods. 116 Persons were
never reached after the initial phone interview.
The remaining 109 nascent entrepreneurs were
still trying to set up their business the last time
we contacted them (follow-up 1, 2, 3 and/or 4).
Thus, a minimum of 36% of the sample started
and a minimum of 20% abandoned the startup
e?ort during the three-year period under study.
Of the remaining 44% we do not have data
about their eventual startup status. If we com-
pare the 116 persons of whom no follow up
information is available with the 292 persons
who either started or quitted their e?ort we ?nd,
TABLE I
Continued
Approach Variable Categories % N Arguments
Ecological Risk of the market
(n = 262)
Hardly any risk
A little bit of risk
Quite some risk
High risk
18
68
12
2 Higher risk means higher chances
of failure
Intended
organization
Ambition to grow
large (n = 267)
To stay small
To grow large
81
19
High growth ambitions
give motivation but can also
be a source of disillusionment
Start out part-
or full-time
(n = 258)
Part-time
Full-time
52
48
Part-time less risk but less committed
full-time more committed
but more risk
Techno nascent
(n = 271)
No
Yes
86
14
Higher risk, but also higher chances
of success
Team (n = 266) Solo
Team
63
37
Lack of team complementarity
Risk of team disagreements
Industry type
(dummy variables)
Manufacturing
Trade
Business services
Consumer services
10
17
34
16
Serious entrepreneurs,
but high entry barriers
Easy to start, easy to fail
Easy to start, easy to fail
Easy to start, easy to fail
370 Marco van Gelderen et al.
surprisingly, the people that took part in follow
ups tended to be less educated and higher in
push motivation than those that dropped out of
the study.
In our design it is the entrepreneur himself
who de?nes whether his business is actually
started or still in the start-up phase. This implies
that entrepreneurs can use di?erent criteria to
judge whether they consider themselves started or
not. In fact, the question why someone consid-
ered him- or herself started gave rise to a pleth-
ora of answers. This is consistent with results
found by Reynolds and Miller (1992). In
Table III these answers are classi?ed using the
properties of emerging organizations given by
Katz and Gartner (1988). This heterogeneity is,
in fact, an argument to take the judgment of the
nascent entrepreneur as the key criterion for
start-up. Only in this way the particular situation
of each nascent entrepreneur will be re?ected.
We choose to use the subjective measure as we
feel that the application of uniform, objective
measures creates a degree of arbitrariness. For
example, ?rst sales might be taken as an indica-
tor of being started, but many people start a
business based on an activity for which they pre-
viously informally received money (for example,
bookkeeping or repairing computers). In an
opposite fashion, some ?rms start out with
investing, and only after a certain period do ?rst
sales come in. So when interpreting the results,
one has to bear in mind the underlying heteroge-
neity in the performance measure. In fact, in a
di?erent study using the present data set, the
application of theory driven measures of whether
a business actually started resulted in somewhat
di?erent explanatory success factors (Van Gelderen,
2001).
We limit our analysis to a comparison between
those who succeed in starting a business and
those who abandon the startup e?ort. The cate-
gory of people still trying to start a business is
not analyzed in this paper. There are two reasons
for this decision. First of all, in the last follow up
interview only a few people are still trying to
start a business. Nascent entrepreneurs still trying
who were reached in earlier follow up interview
might very well either have started or stopped
their e?ort after three years. Second, since the
initial sample was collected at one point in time,
people who were about to set up shop were rep-
resented in the sample as well as people who had
just begun the startup process. So even if one
wants to analyze the group of still trying, one
TABLE II
Moments of getting started/e?ort abandoned
(t0) t1 t2 t3 t4 Total
Half year One year Two years Three years
Started 129 33 15 15 192
Abandoned 42 38 20 15 115
Total 171 71 35 30 307
(Unavailable/still trying) (346) (275) (240) (210) (210)
(Grand total) (517) (517) (517) (517) (517)
TABLE III
Di?erent de?nitions of start-up moments
Intention Boundary Resources Exchange
Wish or desire Registration ch. comm. Arranged ?nance First customer
Idea Sign at magistracy Hired personnel First cash ?ow
Resolution O?cial address Arranged housing Acceptation in market
Ambition Business cards Production of goods A certain scale
Gave up job O?cial opening Bought inventory To derive income
Searched information Bank account Got license To buy stock
371 Success and Risk Factors
should correct for the fact that the still trying
group shows an overrepresentation of people who
have only lately begun to set up their business.
5.2. Comparison with international e?orts
The e?ort made in the Netherlands is compara-
ble to those in the U.S. and Sweden in terms of
sample size. However, for budgetary reasons the
Dutch research used only a sample of the ERC
phone- and interview questions. Amongst others,
most questions on start-up activities and their
timing are left out. Also, not all approaches dis-
cussed in our theory section are measured, such
as the cognitive characteristics and personality
traits. However, given the lack of empirical data
about nascent entrepreneurship, in our opinion
the data still warrant thorough investigation and
publication. Moreover, while topics covered in
the ERC questionnaire were assigned to particu-
lar persons and institutions (for example, an
exclusive right to publish about birth order and
its relationships), this was not the case in the
Dutch study. This makes it possible to prevent
an overview of success and risk factors in nascent
entrepreneurship and an estimation of their rela-
tive strength. Finally, publication of the present
results enables international comparison.
5.3. Independent variables
The independent variables that were used to
establish the success and risk factors in nascent
entrepreneurship are listed in Table I, together
with their descriptive statistics. Some of the vari-
ables are only rough approximations of the
approaches. For example, years of industry expe-
rience is taken as a proxy for the network of the
nascent entrepreneur. However, industry experi-
ence is biased in favor of older people, and does
not describe the amount, strength or diversity of
the ties that a person has with the industry. Four
continuous variables (work experience, manage-
ment experience, industry experience, and desired
start-up capital) are recoded into categories to
mitigate the e?ects of very large numbers. The
categories become larger as the average value of
the categories increases in order to re?ect dimin-
ishing marginal returns. Age is recoded into cate-
gories to obtain insight into the relations of the
di?erent age categories with the other variables.
Industry sector is recoded into four dummy vari-
ables (manufacturing, trade, business services,
and consumer services). The individuals that are
not assigned to any of these four industry vari-
able represent start-up e?orts in agriculture,
artists, repair shops, etcetera and used as the
base case in the regression analysis (Table IV).
Most independent variables have some missing
data, though it never exceeds 10% of the cases.
Missing values sometimes occur when persons do
not know the value of the variable at the time of
the ?rst interview. The experience variables are
ascertained for the ?rst time in a follow up inter-
view, which also generated some missing values
(since not all people participated in a follow-up
interview). For the multivariate analyses an
expected maximization procedure is used to
replace missing data based on underlying data
TABLE IV
Description of change variables used in the regressions
Variable Changed category %N
Business plan Business plan ? no
business plan
No business plan ? business
plan
10 (4)
12 (4)
Start full-time/
part-time
Full-time ? part-time
Part-time ? full-time
6 (2)
8 (3)
Team Team ? solo
Solo ? team
6 (2)
9 (3)
Startup capital Less capital two
ordinal points
Less capital one
ordinal point
More capital one
ordinal point
More capital two
ordinal points
1 (1)
11 (1)
15 (1)
1 (1)
Risk of market Perceived less risk
three ordinal points
Perceived less risk
two ordinal points
Perceived less
risk one
ordinal point
Perceived more
risk one ordinal point
Perceived more
risk two ordinal points
Perceived more
risk three ordinal points
2 (1)
2 (1)
8 (3)
34 (13)
12 (5)
10 (4)
372 Marco van Gelderen et al.
patterns, while keeping means and standard
deviations constant (Dempster et al., 1977). The
independent variables are checked on possible
multicollinearity.
Some frequency distributions of the indepen-
dent variables are striking. Only a minority
(19%) of the nascent entrepreneurs prefers to
grow large or to become rich. Many distributions
of variables are skewed suggesting limited under-
lying heterogeneity. Table I reports the initial
values provided by the respondents. However,
some variables changed over time, while others
remained constant. In terms of approaches, the
variables concerning the individual remain con-
stant, while the other variables may have devel-
oped during the startup process. We analyze
success and failure using initial values as well as
using the values reported in later follow up inter-
views. In order to limit the loss of degrees of
freedom, only variables for which 5% or more of
the respondents report changes are analyzed.
These are business plan, risk of the market, team
vs. solo, startup capital, and means of ?nancing
(third party loan vs. own money). For develop-
ment variables we use the data reported in the
follow-up interview preceding the follow-up inter-
view in which someone stated they had started or
stopped.
Again it should be noted that the initial sam-
ple was collected at one point in time. This
means that people who were about to open up
shop are in the sample as well as people who had
just begun the startup process. With regard to
the ?rst group, no development data were avail-
able as many of them started within six months
(before the ?rst follow up). In fact, by the time
of follow up interview 1 already 56% of the
respondents had either started or given up. On
the other hand, there were also people who had
begun their pre-startup activities a long time ago
(the maximum is 84 months in our sample). With
these people, the observed values at the initial
interview may already have developed since the
very beginning of the start-up e?ort. However,
we do not know with which values they started
initially. In order to make the sample somewhat
less heterogeneous, we decided to restrict the
sample to those people who had started setting
up their business in the preceding year. There
were 36 respondents who had been preparing
their business for more than a year. The ?nal
sample consisted of 307 ) 36 ¼ 271 persons, of
whom 174 got started (64%) and 97 abandoned
the startup process (36%).
5.4. Statistical strategy
As our explanatory variables are a mixture of
categorical and ordinal variables, and our crite-
rion measure is categorical, we use logistic regres-
sion analysis. Development variables are included
in the second step. First, we report on success
and failure characteristics for the total sample.
However, this goes against the spirit of Gartner’s
framework since his main purpose was to show
the large variety in startup e?orts. Therefore, we
also conduct analyses for subgroups. In order to
?nd subgroups, we conducted a PRINCALS
analysis to identify independent factors of com-
pounded characteristics. PRINCALS, an acro-
nym for PRINcipal Components analysis by
Alternating Least Squares, is a non-linear princi-
pal component analysis method (Gi?, 1990). The
PRINCALS method enables handling ordinal
data.
A principal component analysis describes a
number of variables with a smaller number of
variables, termed the principal components, that
still contain as much information, exhibited in
the original variables, as possible. The results of
the analysis reveal that two factors can be distin-
guished. The ?rst factor can be typi?ed as a fac-
tor measuring ambition, with the amount of
start-up capital, ambition getting large, a full-
time start-up, and ambition getting rich having
component loadings above 0.50. The second factor
can be typi?ed as a factor measuring experience,
with work experience, management experience
and industry experience having component load-
ings above 0.50. Using standardized scores, these
variables are combined into a single index. The
sample is split by the mean in order to investi-
gate success and risk factors within subgroups
(low vs. high ambition and low vs. high in experi-
ence). The PRINCALS analysis showed gender
(being male) to belong to the ambition factor,
and age to belong to the experience factor. These
demographics were not made part of the col-
lapsed variable but were left out of the respective
regression analyses.
373 Success and Risk Factors
6. Results
The logistic regression results for the complete
sample are presented in Table V. The dependent
variable distinguishes between a successful start-up
(regardless of the ?rm’s performance after the
start-up) (N ¼ 174, 64%) and an abandonment
of the start-up e?ort (regardless of whether or
not the nascent entrepreneur succeeds in setting
up another business) (N ¼ 97, 36%). Three vari-
ables directly relate to success both with regard
to the initial (t0) measures as well as in a
dynamic sense (the development variables). As
stated in the method section, 56% of the nascent
entrepreneurs start or stop before the follow-up
interview; so their initial (t0) measure may also
be an ‘‘end-measure’’. The observed initial value
is not always truly initial. The ?rst variable
related to pre-startup success is perceived risk of
the market. There is a circularity to this ?nding:
people who perceive less risk will start earlier,
whether their risk perception is accurate or not.
The same reasoning applies to starting full-time
vs. part-time: the decision to switch from part-
time to full-time may be grounded on clear indi-
cations that the entrepreneur can indeed start the
business. Nascent entrepreneurs who intend to
use more start-up capital have lower probabilities
to get their business running. Change in required
start-up capital (along the process) also has a sig-
ni?cant e?ect, in other words those who lowered
their capital requirements increased their chances
of getting started. Nascent entrepreneurs wishing
TABLE V
Estimation results full sample (N=271, 174 started, 97 stopped)
Values in 1998 Development
values
Without
dev. values
Coef. S.E. Coef. S.E. Coef. S.E.
Gender female/male 0.22 0.40 0.00 0.35
Age young/old )0.37 0.23 )0.24 0.22
Push motivation )0.51 0.40 )0.03 0.36
Education low/high 0.45 0.34 0.21 0.31
Work experience 0.11 0.27 0.06 0.25
Management experience 0.18 0.19 0.15 0.17
Experience in setting up 0.14 0.40 0.36 0.37
Business plan 0.04 0.34 0.58 0.54 0.11 0.30
Information and guidance 0.57 0.39 0.61* 0.36
Start part-time/full-time 0.77** 0.36 2.37** 0.89 0.68** 0.33
Industry experience 0.23 0.15 0.24* 0.14
Start up capital )0.39** 0.19 )0.81* 0.44 )0.26 0.17
Third party loan )0.24 0.38 )0.37 0.35
Risk of the market )0.92** 0.29 )0.89** 0.22 )0.51** 0.24
Dummy manufacturing 1.38** 0.70 1.19* 0.66
Dummy trade )0.09 0.48 )0.37 0.43
Dummy business services 0.49 0.44 0.26 0.40
Dummy consumer services 0.31 0.53 0.29 0.46
Ambition becoming rich 0.15 0.47 0.20 0.44
Ambition becoming large )0.45 0.42 )0.42 0.39
Techno nascent )0.14 0.48 )0.25 0.44
Solo-team )0.27 0.35 )0.38 0.68 )0.20 0.31
Constant 1.19 1.69 )0.22 1.47
Chi-square model test St 69.83** 37.57**
Nagelkerke R
2
0.31 0.18
Note: Values relate to initial questionnaire, developments relate to last moment before reported start-up or abandonment
Respondents that were preparing the business for more than a year at the moment that they were contacted for the initial survey
were excluded.
** p < 0.05.
* p < 0.10.
374 Marco van Gelderen et al.
to start out in manufacturing have a higher prob-
ability of success in comparison with nascent
entrepreneurs in other sectors.
A common supplementary measure for assess-
ing the ?t of the model in these kinds of appli-
cations is the Hosmer and Lemeshow test,
where the probability of an outcome is speci?ed
rather than the actual occurrence of an out-
come. This test rejects the hypothesis that the
model does not ?t well. For this test, cases are
sorted by predicted outcome and then divided
into 10 subgroups of equal sizes. For each sub-
group, the numbers of observed and expected
successes and failures are compared. The p-value
associated with the associated Chi-square test
statistic equals 0.384, which provides support
for using this particular model. Also, inclusion
of the development of some explanatory vari-
ables results in a signi?cant improvement as
regards to overall model-?t. The associated
increase in likelihood (leading to a likelihood
ratio statistic of 32.26) is signi?cant at the 5%
level, using the Chi-squared distribution with 5
(number of additional parameters) degrees of
freedom. We also checked for the presence of
curvilinear relationships as regards to the ordi-
nal explanatory variables that are included in
the analysis. There was no evidence for such
curvilinear relationships.
The last two columns of Table V give the
results of a logistic regression model when the
change variables are not taken into account.
After all, risk perception, time investment, and
capital requirements may be changed in the pro-
cess, but the potential entrepreneur wants to
know his chances at the very beginning. Two
additional variables taking on a signi?cant posi-
tive impact are industry experience and exposure
to guidance and advice agencies. Summarizing
the results we conclude that few of the nascent
entrepreneurs’ characteristics are directly associ-
ated with success (a start-up). Most of the signi?-
cant ?ndings relate to the entrepreneur’s
environment: start-up capital and risk of the mar-
ket are seen to be the most important features.
As characteristics of the intended organization,
starting a manufacturing ?rm and of starting
full-time are also important. Indicators of the
followed process were not signi?cant. Moreover,
none of the included individual characteristics
seem to distinguish successful nascent entrepre-
neurs from the unsuccessful ones. Of course, we
only investigated direct e?ects, and there may be
indirect (mediated or moderated) e?ects of these
variables.
As stated in the design section of the method
paragraph, we investigate success and risk factors
for subgroups in the total sample. Two measures
that collapse a number of variables are derived
from a PRINCALS analysis and used as a basis
for categorization. The ambition score is a stan-
dardized mean of the variables ‘wish to grow
large’, ‘start full-time’, and ‘start-up capital’, and
‘third party loan’. Gender is left out of the analy-
sis. Splitting the sample by the mean, we identify
a ‘limited ambition group’ (164 cases) and a ‘high
ambition group’ (107 cases). We apply the same
regression to both ambition subgroups as we did
earlier to the total sample. Some interesting
results emerge in Table VI.
Among the nascent entrepreneurs showing
limited ambition in our sample, older people are
less likely to get the business started. Interest-
ingly, a business plan works positively for
nascents with limited ambition but negatively for
nascents with high ambitions. Those with high
ambitions who write a business plan later on,
have increased chances of success. Management
experience is also useful for those with high
ambitions. From the characteristics that turned
out to be signi?cant in the entire sample, only
the negative e?ect of market risk remains
(although less signi?cant). Among nascent entre-
preneurs revealing high ambitions, risk of the
market seems just as important as for those with
lower ambitions. A push motivation combined
with high ambition leads to a lower propensity
to start-up.
Similarly, two subgroups distinguishing experi-
ences of the nascent entrepreneurs are identi?ed
(141 with limited experience, 130 with high expe-
rience). The experience score is a standardized
mean of the variables ‘work experience’, ‘man-
agement experience’, and ‘industry experience’.
Age is left out of the analysis. The corresponding
results are shown in Table VII. It appears that,
apart from perceived market risk, there is not
much to predict on the chances of success for
nascent entrepreneurs with high degrees of expe-
rience. Among nascent entrepreneurs with limited
375 Success and Risk Factors
experience, there is more variation that can be
explained by the characteristics distinguished.
Interestingly, making use of information and
guidance increases the chances of success among
less experienced business founders. People with
experience in setting up a business but who have
relatively little experienced otherwise also have
an advantage.
7. Discussion
We have studied a range of approaches and their
associated variables in their relationship with
pre-startup success. For the full sample, the
results show four variables to be signi?cantly
related, three of them both in their initial values
and as change variables. The association with
success of starting part-time or full-time may
appear to be a circular ?nding, as the amount of
time that one can put in the business is a success
measure by itself. Still, the position that it is eas-
ier to start a part-time business because of pre-
sumably smaller scale and ?nancial risk receives
no support, using our particular sample. Setting
up a business part-time may be a disadvantage
because there isn’t a single focus on the business.
One is distracted as an entrepreneur if not work-
ing full-time.
TABLE VI
Estimation results separating low and high ambition
Limited ambition High ambition
Value in 1998 Development Value in 1998 Development
Coef. S.E. Coef. S.E. Coef. S.E. Coef. S.E.
Gender female/male
Age young/old )0.73** 0.33 0.13 0.38
Push motivation 0.60 0.55 )1.51** 0.71
Education low/high 0.02 0.45 0.53 0.54
Work experience 0.01 0.35 )0.20 0.48
Management experience 0.27 0.25 0.54* 0.31
Experience in setting up 0.70 0.61 )0.06 0.61
Business plan 0.84* 0.44 0.27 0.65 )1.16* 0.64 2.13* 1.22
Information and guidance 0.44 0.54 0.97 0.66
Start part-time/full-time
Industry experience 0.28 0.20 0.14 0.27
Start up capital
Third party loan
Risk of the market )0.75* 0.41 )0.81** 0.27 )1.31** 0.52 )1.03** 0.41
Dummy manufacturing 0.94 0.98 2.24* 1.21
Dummy trade )0.49 0.74 )0.26 0.64
Dummy business services 0.19 0.63 0.20 0.75
Dummy consumer services 0.11 0.71 0.17 0.80
Ambition becoming rich )0.40 0.72 )0.36 0.61
Ambition becoming large
Techno nascent )0.23 0.69 )0.70 0.74
Solo - team )0.47 0.46 )0.21 0.49
Constant 0.71 2.22 3.79 2.77
Chi-square Model Test St 40.75** 31.55**
Nagelkerke R
2
0.31 0.34
Notes: Values relate to initial questionnaire, developments relate to last moment before reported start-up or abandonment.
Limited ambition: N = 164, 70% start, 30% stopped. High ambition: N = 107, 55% start, 45% stopped.
Respondents that were preparing the business for more than a year at the moment that they were contacted for the initial survey
were excluded. Developments on the variables ‘start part-time/full-time’ and ‘solo/team’ were excluded because of the low variation
in subgroups.
** p < 0.05.
* p < 0.10.
376 Marco van Gelderen et al.
Similarly, the results with respect to perceived
risk of the market cannot be considered trivial.
They show the importance of risk management,
as the e?ective use of risk reduction techniques
will lead to lower perceived risk. Moreover, the
central importance of market risk is reassuring in
the sense that this is the result that one would
prefer to ?nd. In the end business success should
be primarily a question of market selection and
not of other factors. Still, risk perception may
also be considered a mediator as conceptually all
remaining variables may in?uence success
through the heightening or lowering of the
perception of risk.
The third variable directly and negatively
a?ecting performance proves to be the amount
of intended startup capital. This shows that it is
easier to start with a small amount of capital.
Amount of intended startup capital will be
related to intended size, and companies are easier
to get started when they are smaller. Another
explanation is that smaller capital may be often
obtainable without having to go through the
?nancial lending system and its risk evaluation
criteria. A greater proportion of funds for smal-
ler projects can be obtainable by pre-startup sav-
ings, loans from friends or family, other partners
in the venture, etc. For those who want to start
TABLE VII
Estimation results separating low and high experience
Limited experience Substantial experience
Value in 1998 Development Value in 1998 Development
Coef. S.E. Coef. S.E. Coef. S.E. Coef. S.E.
Gender female/male 0.64 0.50 )0.60 0.67
Age young/old
Push motivation )0.16 0.57 )0.85 0.57
Education low/high 0.23 0.46 0.20 0.49
Work experience
Management experience
Experience in setting up 1.27* 0.69 )0.60 0.54
Business plan 0.13 0.51 0.93 0.72 0.28 0.50 0.61 0.93
Information and guidance 0.97* 0.59 )0.23 0.56
Start part-time/full-time 0.68 0.52 0.52 0.51
Industry experience
Start up capital )0.51* 0.28 )1.01 0.68 )0.20 0.29 )0.67 0.59
Third party loan )0.38 0.52 0.23 0.59
Risk of the market )0.91** 0.41 )0.82** 0.30 )1.13** 0.47 )0.98** 0.32
Dummy manufacturing 1.83 1.29 1.38 0.86
Dummy trade )0.49 0.68 0.35 0.79
Dummy business services 0.35 0.63 0.74 0.65
Dummy consumer services 0.01 0.79 0.29 0.75
Ambition becoming rich 0.13 0.66 0.05 0.74
Ambition becoming large )0.93 0.57 0.36 0.71
Techno nascent )1.12 0.71 0.68 0.76
Solo/team )0.23 0.47 )0.48 0.55
Constant )0.18 2.33 4.98** 2.46
Chi-square Model Test St 43.45** 25.16
Nagelkerke R
2
0.36 0.25
Notes: Values relate to initial questionnaire, developments relate to last moment before reported start-up or abandonment.
Limited experience: N = 141, 59% start, 41% stopped. Substantial experience: N=130, 70% start, 30% stopped.
Respondents that were preparing the business for more than a year at the moment that they were contacted for the initial survey
were excluded. Developments on the variables ‘start part-time/full-time’ and ‘solo/team’ were excluded because of the low variation
in subgroups.
** p < 0.05.
* p < 0.10.
377 Success and Risk Factors
with a high amount of capital but fail to do so,
di?erent processes may be responsible. On the
one hand dreamers fail to gather their intended
amount of startup capital and are rightfully
rejected by ?nanciers; on the other hand people
with a sound business concept may be unjustly
rejected by ?nanciers. It may also be that some
of these people calculate their prospects carefully
and back o? if risks cannot be reduced (Carter
et al., 1995). Unfortunately we cannot assess the
quality of the opportunity and of the business
concept with our data.
A fourth ?nding is that people in manufac-
turing more often got started. An explanation
may be that comparative to the other sectors it
is less easy to start in manufacturing, as knowl-
edge about and capital for production are
required. This may have a selection e?ect on
those who want to become nascent entrepre-
neurs. Another explanation may be that manu-
facturing enterprises are likely to invest capital
in assets that can be sold in case of bankruptcy.
This means that investing in those kind of com-
panies can be less risky for ?nanciers. Investors
may anyhow be more likely to invest in ‘‘tradi-
tional’’ companies than in a service company. It
is easier for ?nanciers to assess the real worth of
a company (and issue loans covered accordingly)
as tangible assets are of much easier valuation
than intangibles such as human capital skills.
On the approach level, direct e?ects are associ-
ated with environmental variables (risk of the
market, startup capital), and with characteristics of
the intended organization (starting full time, start-
ing in manufacturing). Approaches and variables
not having a direct e?ect on pre-startup success
can still be in?uential. In this paper only direct
e?ects are studied. Their e?ects may be moderated
or mediated by other variables. For example, with
regard to mediators, the correlations indicate that
those who wish to start out full-time are charac-
terized by being male, the ambition to grow large,
and the intended use of a large amount of startup
capital. With regard to moderators, it may be
argued that being female and having experience
will interact positively in predicting success, just
as push motivation and team startup will interact
negatively. Mediators and moderators need to be
speci?ed beforehand, and the results of this study
give input for the derivation of hypotheses.
An issue that was taken up in this paper was
the search for relatively homogeneous subsam-
ples. Acknowledging heterogeneity in the sample
of nascent entrepreneurs proved to be bene?cial
for predicting chances of success. In this particu-
lar sample, the success and failures of less experi-
enced business founders and the business
founders with high ambitions can be reasonably
predicted. This is less so for their counterparts,
i.e., the highly experienced business founder and
the business founder with limited ambitions. Still,
some interesting ?ndings emerged. The results
con?rm that those with limited experience bene?t
from information and guidance. Apparently this
is a fertile target group for guiding agencies.
Push motivation works negatively in combina-
tion with high ambitions. If forced to start a
business, and on the lookout for organizational
employment, it is advisable to start an operation
limited in scope and scale. Finally, the writing of
a business plan works out di?erentially for those
with limited and high ambitions: for the limited
ambition group it correlates positively, while for
the high ambition group it correlates negatively.
An interpretation is that for those who start a
small scale business writing and having a plan
helps them to structure and focus their activities.
Those who start a large-scale business without a
plan may be people who are so knowledgeable and
experienced that they do not need a plan. Then,
later in the startup process, writing a business plan
after turns out to be bene?cial. This runs some-
what counter to Delmar and Shane’s (2004) ?nd-
ings, also using the ERC design but with di?erent
measures and analytical techniques, that a business
plan is particularly useful in the early beginnings of
the startup phase as a means to obtain legitimacy.
The present study has a number of limitations.
First, as stated above, only direct e?ects are stud-
ied. Second, not all approaches proposed in our
theory section are studied, e.g. psychological
approaches. Moreover, of those that are studied
the operationalizations are sometimes crude.
Third, the data prove to be far from perfect.
Because the sample of nascent entrepreneurs is
collected at a single point in time they vary in
the number of months they had been preparing
their business. Thus, both the initial values and
the change variables are confounded: for some
the initial values are in reality end values; and
378 Marco van Gelderen et al.
end values could be computed for a limited sub-
set of the sample only. Fourth, our survey study
does not shed light on variables that are less eas-
ily accessible. The so-called ‘‘how’’ variables
(VanderWerf, 1993) are not taken into account,
for example how resources are developed, how
relationships are maintained, and how informa-
tion is gained (Cooper, 1993).
Government policy in the old, managed econ-
omy was largely about control. High certainty
dictated that it was known what to produce, how
it should be produced, and who would produce
it. The role of government was to constrain the
power of large corporations, which were needed
for e?ciency under mass-production, but posed a
threat to democracy through their concentration
of power (Chandler, 1977, 1990). Under the old,
managed economy the policy debate centered on
competition policies (antitrust), regulation and
public ownership of business (Teece, 1993). In
the new, entrepreneurial economy these con-
straining policies have become increasingly irrele-
vant. The central role of government policy in
the new, entrepreneurial economy is enabling in
nature. The focus is to foster the production and
commercialization of knowledge. Rather than
focus on limiting the freedom of ?rms to contract
through antitrust, regulation and public owner-
ship, government policy in the new, entrepreneur-
ial economy targets education, increasing the
skills and human capital of workers, and facili-
tating the mobility of workers and their ability to
start new ?rms (Audretsch and Thurik, 2001).
Knowledge of relevant factors and in?uences in
the pre-start-up phase is essential for creating a
portfolio of new enabling policies. Therefore, we
believe that e?orts to understand predictors of
pre-start-up performance are an important part
of entrepreneurship research. Characteristics of
nascents, i.e., people who are in the process of
setting up a business, are hardly dealt with in the
area of entrepreneurship research. The present
study is one of the ?rst to contribute to this new
area. We hope the simple models described here
will encourage the work yet to be done.
Acknowledgements
We express our thanks to Paul Reynolds and the
ERC for the design, to Mathilde Blok for assist-
ing in the development of the theoretical frame-
work, and to Luca Bregoli, Paul Jansen, Gaetan
Laederich, Niels Plum, and Lorraine Uhlaner for
suggestions. This research project was funded by
the Dutch Ministry of Economic A?airs in an
assignment given to EIM Zoetermeer. The
research design was developed by the ERC,
headed by Professor Paul Reynolds.
References
Aldrich, E. A., 1990, ‘Using an Ecological Perspective to
Study Organizational Founding Rates’, Entrepreneur-
ship, Theory and Practice spring 1, 7–24.
Aldrich, H. E., 1999, Organization Evolving, London: Sage
Publications.
Audretsch, D. B. and A. R. Thurik, 2001, ‘Sources of
Growth: The Entrepreneurial versus the Managed Econ-
omy’, Industrial and Corporate Change 10(1), 267–315.
Baron, R. A., 1999, ‘Counterfactual Thinking and Ven-
ture Formation: The Potential E?ects of Thinking
About ‘‘What Might Have Been’’’, Journal of Business
Venturing 15, 79–91.
Baron, R. A., 2004, ‘The Cognitive Perspective: A Valu-
able Tool for Asking Entrepreneurship’s Basic ‘‘why’’
Questions’, Journal of Business Venturing 19, 221–239.
Bhave, P. B., 1994, ‘A Process Model of Entrepreneurial
Venture Creation’, Journal of Business Venturing 9,
223–242.
Blanch?ower, D. G. and A. J. Oswald, 1998, ‘What
Makes an Entrepreneur?’ Journal of Labor Economics
16(1), 26–60.
Busenitz, L. W. and C. M. Lau, 1996, ‘A Cross-Cultural
Cognitive Model of New Venture Creation’, Entrepre-
neurship, Theory and Practice 20, 25–39.
Campbell, C. A., 1992, ‘A Decision Theory Model for
Entrepreneurial Acts’, Entrepreneurship, Theory and
Practice 17, 21–27.
Carree, M. A. and A. R. Thurik, 2003, ‘The Impact of
Entrepreneurship on Economic Growth’, in D. B. Au-
dretsch and Z. J. Acs (eds.), Handbook of Entrepreneur-
ship Research, Boston/Dordrecht: Kluwer Academic
Publishers, forthcoming.
Carter, N. M., W. B. Gartner and P. D. Reynolds, 1995,
‘Exploring Start-up Event Sequences’, Journal of Busi-
ness Venturing 11, 151–166.
Chandler, A. D., 1977, The Visible Hand: The Managerial
Revolution in American Business, Cambridge: Harvard
University Press.
Chandler, A. D., 1990, Scale and Scope: The Dynamics of
Industrial Capitalism, Cambridge: Harvard University
Press.
Chen, C. C., P. G. Greene and A. Crick, 1998, ‘Does
Entrepreneurial Self-E?cacy Distinguish Entrepreneurs
from Managers?’ Journal of Business Venturing 13,
295–316.
Cooper, A. C., 1993, ‘Challenges in Predicting New Firm
Performance’, Journal of Business Venturing 8, 231–253.
379 Success and Risk Factors
Deakins, D. and G. Whittam, 2000, ‘Business Start-up:
Theory, Practice and Policy’, in S. Carter and D. en
Jones-Evans (eds.), Enterprise and Small Business,
Essex: Pearson.
Delmar, F. and P. Davidsson, 2000, ‘Where Do They
Come From? Prevalence and Characteristics of Nas-
cent Entrepreneurs’, Entrepreneurship and Regional
Development 12, 1–23.
Delmar, F. and S. Shane, 2004, ‘Legitimizing First: Orga-
nizing Activities and The Survival of New Ventures’,
Journal of Business Venturing 19, 385–410.
Dempster, A. P., N. M. Laird and D. B. Rubin, 1977,
Maximum Liklihood From Incomplete Data Via the
Em Algorithm’, Journal of the Royal Statistical Society
B, 39, 1–38.
Gartner, W. B., 1985, ‘A Conceptual Framework for
Describing the Phenomenon of New Venture Crea-
tion’, Academy of Management Review 10(4), 696–706.
Gatewood, E. J., K. G. Shaver and W. B. Gartner, 1995,
‘A Longitudinal Study of Cognitive Factors In?uenc-
ing Start-up Behaviours and Success at Venture Crea-
tion’, Journal of Business Venturing 10, 371–391.
van Gelderen, M. W. 2001, ‘Ondernemers Vo´ o´ r De Start:
De?nitiekwesties En Succesfactoren’, Bedrijfskunde 73,
50–56.
Gi?, A., 1990, Nonlineair Multivariate Analysis, Chichester:
Wiley.
Greenberger D. B. and D. L. Sexton, 1988, ‘An Interac-
tive Model of New Venture Initiation’, Journal of
Small Business Management 26, 1–7.
Herron, L. and H. J. Sapienza, 1992, ‘The Entrepreneur
and the Initiation of New Venture Launch Activities’,
Entrepreneurship, Theory and Practice 17, 49–55.
Holtz-Eakin, D., D. Joulfaian and H. S. Rosen, 1994,
‘Sticking It Out: Entrepreneurial Survival and Liquidity
Constraints’, Journal of Political Economy 102(1), 53–76.
Hosmer, D. W. and S. Lemeshow, 1989, Applied Logistic
Regression, New York: John Wiley and Sons.
Johnson, B. R., 1990, ‘Toward a Multidimensional Model
of Entrepreneurship: The Case of Achievement Moti-
vation and the Entrepreneur’, Entrepreneurship Theory
and Practice spring 39–54
Kahneman, D., P. Slovic and A. Tversky, 1982 Judgment
Under Uncertainty: Heuristics and Biases. Cambridge:
Cambridge University Press.
Kamm, J. and A. Nurick, 1993, ‘The Stages of Venture
Formation: A Decision-Making Model’, Entrepreneur-
ship Theory and Practice 17(2), 17–28.
Katz, J. and W. B. Gartner, 1988, ‘Properties of Emerging
Organizations’, Academy of Management Review 13,
429–441.
Krueger, N. F. jr., M. D. Reilly and A. L. Carsrud, 2000,
‘Competing Models of Entrepreneurial Intentions’,
Journal of Business Venturing 15, 411–432.
Larson, A. and J. A. Starr, 1993, ‘A Network Model of
Organization Formation’, Entrepreneurship, Theory and
Practice 18, 5–15.
Learned, K. E., 1992, ‘What Happened Before the Orga-
nization? A Model of Organization Formation’, Entre-
preneurship, Theory and Practice 17, 39–48.
Na?ziger, W. N., J. S. Hornsby and D. F. Kuratko, 1994,
‘A Proposed Model of Entrepreneurial Motivation’,
Entrepreneurship, Theory and Practice 18, 29–42.
Rauch, A. and M. en Frese, 2000, ‘Psychological
Approaches to Entrepreneurial Success: A General
Model and an Overview of Findings’, International
Review of Industrial and Organisational Psychology, 15,
101–143.
Reynolds, P. D., 1997, ‘Who Starts New Firms? – Preli-
minary Explorations of Firms-in-Gestation’, Small
Business Economics 9, 449–462.
Reynolds, P. D., 2000, ‘National Panel Study of U.S.
Business Startups: Background and Methodology’, in
Katz (ed.), Advances in Entrepreneurship, Firm Emer-
gence and Growth, Vol. 4: Databases for the Study of
Entrepreneurship, Amsterdam: Elsevier Science.
Reynolds, P., N. Carter, W. Gartner and P. Greene, 2004,
‘The Prevalence of Nascent Entrepreneurs in the Uni-
ted States: Evidence From The Panel Study of Entre-
preneurial Dynamics’, Small Business Economics 23,
263–284.
Reynolds, P. and B. Miller, 1992, ‘New Firm Gestation:
Conception, Birth, and Implications for Research’,
Journal of Business Venturing 7, 405–417.
Reynolds, P. and S. White, 1992, ‘Finding the Nascent
Entrepreneur: Network Sampling and Entrepreneur-
ship Gestation,’ in N. Churchill, et al. (eds.), Frontiers
in Entrepreneurship Research, Wellesley: Babson Col-
lege.
Shapero, A. and L. Sokol, 1982, ‘Social Dimensions of
Entrepreneurship’, in C. Kent, D. Sexton and K. Vesper
(eds.), The Encyclopedia of Entrepreneurship. Englewood
Cli?s: Prentice-Hall, 72–90.
Simon, M., S. M. Houghton and K. Aquino, 2000, ‘Cog-
nitive Biases, Risk Perception, and Venture Formation:
How Individual Decide to Start Companies’, Journal
of Business Venturing 15, 113–134.
Starr, J. E. and N. Fondas, 1992, ‘A Model of Entrepre-
neurial Socialization and Organization Formation’,
Entrepreneurship, Theory and Practice 17, 67–76.
Teece, D. J., 1993, ‘The Dynamics of Industrial Capital-
ism: Perspectives on Alfred Chandler’s Scale and
Scope’, Journal of Economic Literature 31, 199–225.
Tiessen, J. H., 1998, ‘Individualism, Collectivism, and
Entrepreneurship: A Framework for International
Comparative Research’, Journal of Business Venturing
12, 367–384.
VanderWerf, P. A., 1993, ‘A Model of Venture Creation
in New Industries’, Entrepreneurship, Theory and Prac-
tice 18, 39–47.
380 Marco van Gelderen et al.

doc_231082825.pdf