Description
On this brief description concerning smells like team spirit how entrepreneurial founding team motivations affect new venture.
STEUNPUNT ONDERNEMEN EN REGIONALE ECONOMIE
NAAMSESTRAAT 61 – BUS 3550
BE-3000 LEUVEN
TEL +32 16 32 66 61 | FAX +32 16 37 35 11
[email protected]
www.steunpuntore.be
Beleidsrapport STORE-B-14-011
Smells like team spirit? How entrepreneurial
founding team motivations affect new venture
financial and innovation performance
J onas Debrulle
1,a
en J ohan Maes
a
a
IESEG School of Management
23 december 2014
1
De resultaten in dit rapport geven de mening van de auteurs weer en niet deze van de Vlaamse overheid: de
Vlaamse Gemeenschap/het Vlaams Gewest is niet aansprakelijk voor het gebruik dat kan worden gemaakt van de
in deze mededeling of bekendmaking opgenomen gegevens.
1
Smells like team spirit? How entrepreneurial founding team
motivations affect new venture financial and
innovation performance
Jonas Debrulle
Johan Maes
Abstract
This study investigates the relationship between entrepreneurial founding teams’ (EFTs)
entrepreneurship motivations and new venture financial and operational performance,
measured as ‘return on assets’ and ‘firm innovation’. We consecutively introduce the level
and heterogeneity of EFTs’ autonomous and controlled entrepreneurship motivation as
potential drivers of new venture performance. Our analyses are based on a sample of 66
teams representing 142 founders. We observe that EFTs’ level of autonomous and controlled
entrepreneurship motivation contribute to new venture return on assets. We also find support
for the negative impact of EFTs’ controlled entrepreneurship motivation on new venture
innovation. Introduction of the moderating motivation heterogeneity variables uncovers
complex interactions between EFTs’ levels of entrepreneurship motivation, team member
motivation heterogeneity, and this study’s dependents. Our findings reveal that motivation
heterogeneity acts as a positive moderator of the relationship between EFTs’ level of
autonomous/controlled entrepreneurship motivation and new venture innovation, while it
emerges as a negative moderator of the relationship between EFTs’ controlled motivation and
new venture financial performance. Implications for further research are highlighted.
Introduction
New venture creation and development rarely involves a “lonely hero” exploiting a lucrative
opportunity. Rather, evidence supports the idea that entrepreneurship commonly represents a
collective activity, whereby teams of entrepreneurs are responsible for creating and managing
new ventures (McMullen et al., 2008). Ever since Gartner et al. (1994) emphasized that
organizations are in fact social entities, and that the “entrepreneur in entrepreneurship” is
more likely to be plural than singular, an increasing number of researchers have shifted their
attention to the entrepreneurial team phenomenon. Building on prior contributions, Kamm et
al. (1990) defined an entrepreneurial founding team as “two or more individuals who jointly
establish a business in which they have an equity interest” (p. 7). Ucbasaran et al. (2003)
extended this definition to include individuals who “have a key role in the strategic decision-
making of the venture at the time of founding” (p. 109). This study focuses on entrepreneurial
founding teams (EFTs) of which all members meet these conditions and continue to impact
the strategic development of the new venture.
So far, two types of team heterogeneity have dominated research on EFTs. On the one
hand, prior work has emphasized demographic differences between entrepreneurs. These
differences are conceptualized as “surface-level” diversity or heterogeneity (Harrison et al.,
1998). Particular interest has been directed to age and/or gender diversity between business
founders (e.g., Chandler et al., 2005; Chowdhury, 2005). On the other hand, task- or skill-
related heterogeneity (Milliken and Martins, 1996) has equally emerged as a chief topic.
Though a regularly debated topic, extant work suggests that diversity of skills within EFTs
increases the pool of cognitive resources, which enables the team to better handle the
2
complexities inherent to new venture emergence and development, thus stimulating new
venture performance. Widely held indicators of task and skill heterogeneity include
differences in educational specialization, functional expertise, and industry experience (e.g.,
Chandler et al., 2005).
In sharp contrast to these demographic diversity and skill differences, for which a link
between team heterogeneity and firm performance has been established in the context of
EFTs and top management teams (TMTs) in more established organizations, the issue of
more psychological characteristics has received far less research attention. Recent
contributions have, nevertheless, argued that such characteristics are just as likely to affect
team effectiveness and organizational outcomes (Chowdhury, 2005; Van Knippenberg et al.,
2004). In an empirical study of new venture TMT composition, Ensley and Pearce (2001) and
Ensley and Hmieleski (2005) shed light on the implications of team cohesion for new venture
performance. They found that contrary to affective conflicts, cognitive conflicts between
entrepreneurial team members contribute directly to new venture success. Likewise, when
evaluating the theoretical basis for demographic diversity in EFTs, Chowdhury (2005)
established that entrepreneurial team effectiveness builds on team member commitment and
comprehensiveness in strategic decision-making. This study extends the field of deep-level
(directly unobservable) team diversity by exploring the effects of individual motivation
heterogeneity between entrepreneurial team members on new venture performance. All in all,
however, motivation remains a highly neglected variable in team heterogeneity research (Van
Knippenberg et al., 2004).
According to Self-Determination Theory (SDT) (Gagné and Deci, 2005), people can
adhere to a task with eagerness and volition because they genuinely enjoy doing it, or because
the task at hand feels personally important (autonomous motivation). In contrast, they can
engage in an activity with a sense of pressure, because they wish to attain certain rewards that
are dependent upon task completion, or because they strive for outside approval (controlled
motivation). Hence, in an entrepreneurial setting, ‘autonomous entrepreneurship motivation’
causes individuals to engage in new firm formation and development simply because they
find this activity enjoyable and derive satisfaction from it, or because they are deeply
committed to becoming an entrepreneur and wish to maintain their business. ‘Controlled
entrepreneurship motivation’, on the contrary, provokes individuals to create and develop
new ventures because of the anticipated rewards (including social rewards), or because they
appreciate the status associated with being an entrepreneur while evading the stigma of
business failure. Though autonomous and controlled motivation both elicit and sustain
purposeful behavior, they are governed by distinctive mechanisms (Gagné and Deci, 2005)
and are aimed at different objectives (Deci et al., 1999) that arise from a particular ‘locus of
causality’ (DeCharms, 1968). Despite these inconsistencies, in the research specific to
entrepreneurial team heterogeneity little is known about the effect of motivational team
composition on new venture performance.
This study aims to make the following two contributions to existing literature. First, it
encompasses a theoretical perspective. It draws on motivation theory, specifically SDT
(Gagné and Deci, 2005), to introduce autonomous and controlled motivation as EFT
psychological driving forces of new venture performance. By exploring motivational
differences in the context of EFTs, we wish to corroborate the idea that factors other than
demographic diversity and skill differences affect team effectiveness and organizational
outcomes (Chowdhury, 2005). This research gap not only concerns research on private firms
owned by EFTs, but also that addressing TMTs in general. While we acknowledge that the
relationships found in this study, due to differences in individual risk bearing, ownership and
decision-making power (Ucbasaran et al., 2003), may not necessarily hold for TMTs in
established firms, they could inspire researchers and policy-makers to broaden their view of
3
team-based venture determinants to include diversity in psychological and personality
characteristics. A second contribution of the current study encompasses an empirical
perspective. How to define and measure new venture performance has long been debated.
Both financial and operational conceptualizations have been advocated. It is therefore
appropriate to be skeptical about how to best capture new venture performance. We address
this issue by investigating two performance outcomes: one grounded in audited financial
information, and another in operational data. From a financial standpoint, we define new
venture performance in terms of the company’s profitability relative to its total assets (return
on assets), whereas from an operational standpoint we focus on new ventures’ innovation
performance.
This article proceeds as follows: First, we capture the notion of autonomous and
controlled entrepreneurship motivation and discuss its anticipated model impacts. We
identify gaps in existing research and formulate five hypotheses to be tested. Next, we
describe our research methodology, with special emphasis on our sampling procedures,
measures and measurement validity tests. Hypotheses developed in this study were tested
using data from 142 individual entrepreneurs representing 66 EFTs. Finally, we present and
discuss our findings, after which we conclude with some caveats and opportunities for future
research.
Theoretical background and hypotheses
Motivation is the internal disposition that initiates, reinforces and maintains goal-oriented
behaviors. It is the psychological drive that encourages individuals to take action, whether it
is to eat when we feel hungry, or to enroll in a university to obtain a degree. A considerable
part of motivation research interest has been aimed at understanding work-related behavior.
As classics in organizational behavior literature, the theories of Maslow (1954), Herzberg
(1966), Alderfer (1972), and McClelland (1961) are generally referred to as ‘content theories
of motivation’ as they attempt to explain ‘what’ motivates behavior, yet fail to shed light on
‘how’ motivation occurs. ‘Process theories of motivation’, on the other hand, do not primarily
deal with the energizers of motivation (i.e., needs-based approach), but pay attention to ‘how’
individuals direct work-related behavior (e.g., Adams, 1963; Porter and Lawler, 1968;
Vroom, 1964). Since the objective of the current study is to examine how entrepreneurial
behavior is rationalized and maintained in the context of EFTs, rather than to explore internal
energizers of entrepreneurial action, we build on motivation process theories to develop our
hypotheses. More specifically, this study is rooted in the expectancy approach to motivation.
According to Vroom (1964), individuals’ work motivation is the product of three
catalysts: putting forth effort will lead to performance; performance will trigger rewards; and
these rewards are desirable. Similar to Vroom (1964), Porter and Lawler (1968) stipulate that
task completion is dependent on its anticipated rewards. Porter and Lawler (1968), however,
extended Vroom’s (1964) theory by categorizing rewards as intrinsic or extrinsic, thereby
proposing a model of intrinsic and extrinsic work motivation. ‘Intrinsic motivation’ involves
individuals engaging in a task because of the spontaneous satisfaction and sense of
achievement that arise upon successful task completion. ‘Extrinsic motivation’, in contrast,
emanates from outside the individual. In this case, motivation and satisfaction do not
originate from the task itself, but from the rewards to which the task is instrumental. Though
commonly accepted as an adequate theory for explaining task motivation, Porter and
Lawler’s (1968) Expectancy Theory is not without controversy. Especially its implicit
additivity assumption of intrinsic and extrinsic motivation, thus yielding total job satisfaction,
has been subject to widespread debate (Deci et al., 1999). Consequently, when positing their
SDT, Deci and Ryan (1985) no longer acknowledged motivation as a unitary concept.
4
Instead, they emphasized the relative strength of ‘autonomous motivation’ versus ‘controlled
motivation’, rather than sustaining the concept of ‘total motivation’.
Capitalizing on Porter and Lawler’s (1968) dichotomy between intrinsic and extrinsic
motivation, Deci and Ryan (1985) differentiate between ‘autonomous motivation’ and
‘controlled motivation’. Autonomous motivation involves doing a task because it is
inherently fun (subscale: ‘intrinsic motivation’) or because task accomplishment is
considered personally important (subscale: ‘identified regulation’). The latter designates an
internalized form of extrinsic motivation, whereby individuals have come to personally value
work-related behaviors and understand their importance for their own well-being (Ryan and
Deci, 2000). Nonetheless, identified regulation is still considered a type of extrinsic
motivation as it does not involve personal interest in the task at hand, yet requires an
instrumentality between the task and individually appreciated goals (Gagné and Deci, 2005).
Contrary to autonomous motivation, controlled motivation encompasses actions unilaterally
initiated and maintained by external contingencies (subscale: ‘external regulation’ or
‘extrinsic motivation’) as well as behaviors enacted with a feeling of pressure to avoid self-
inflicted punishment, such as shame or guilt (subscale: ‘introjected regulation’) (Deci and
Ryan, 1985). Though equally internalized, introjected regulation, as opposed to identified
regulation, skews more strongly towards pure extrinsic motivation. This type of motivation is
not perceived as congruent with personal objectives, which causes it to be considered less
valuable and important (Ryan and Deci, 2000). Hence, with controlled motivation, the cause
of the behavior is said to have an external ‘perceived locus of control’ (DeCharms, 1968).
Conversely, autonomous motivation originates from an internal ‘perceived locus of control’,
so that individuals feel (relatively) autonomous while performing the activities (DeCharms,
1968).
While several studies have supported the autonomous-controlled dichotomy as an
adequate approach to study work motivation, few have tested the theory within an
entrepreneurial setting. Instead, research on entrepreneurship motivation has largely
concentrated on push-pull factors as predictors of firm activities. Distinguishing between
autonomous and controlled motivation is, however, particularly insightful to entrepreneurship
literature. Several arguments support this view. First, the autonomous-controlled dichotomy
incorporates what is already known in terms of the push-pull strand in entrepreneurship
research (Segal et al., 2005). For instance, autonomous motivation implies a sense of volition
(pull), whereas controlled motivation implies a sense of pressure (push) (Gagné and Deci,
2005). As a result, the motives sustaining entrepreneurial behavior as recognized by the push-
pull perspective correspond to those accommodated by the controlled-autonomous
perspective. Second, contrary to push-pull motives, autonomous and controlled motivation
transcend pre-launch or launch activities of entrepreneurship. They thus can be employed to
ascertain post-entry alterations in motivations of practicing entrepreneurs. Third, the
autonomous-controlled distinction can be applied to a specific (individual or team-based)
activity, a project, or an entire profession. The push-pull debate, on the other hand, only
produces a classification at person-level, thereby disregarding any task-level motivational
differences. Finally, while autonomous and controlled motivation trigger and sustain
purposeful behavior in a diverse way, they are not considered mutually exclusive (Cameron,
2001). Ryan and Deci (2000) posit that individual actions may be simultaneously motivated
by a combination of autonomous and controlled factors operating in a parallel fashion.
Accordingly, autonomous motivation does not exclude an individual from seeking rewards,
and controlled motivation does not prohibit task enjoyment. This does, however, imply that
external contingencies (in case of autonomous motivation) or task interestingness (in the case
of controlled motivation) might be insufficient to maintain motivation.
5
When making predictions, much of the work in motivation literature unilaterally
considers the amount of ‘total motivation’ a person has for a task, thereby ignoring
subdimensional types of motivation. Gagné and Deci (2005) warn against treating motivation
as a unitary concept. They argue that determining motivation by various factors is of little use
if it is then operationalized into a single variable. Not only does this induce the reification of
the motivation construct, it also fails to capture any underlying variations between types of
motivation. That is why, consistent with Gagné and Deci (2005), the current study
differentiates between EFTs’ autonomous and controlled entrepreneurship motivation.
We know from the field of work motivation and regulated behavior that autonomous
and controlled motives aim at achieving separate goals (Deci and Ryan, 1985) that arise from
a different ‘perceived locus of causality’ (DeCharms, 1968). Though autonomous motivation
effects have emerged as the most stable, both types of motivation are said to elicit and sustain
purposeful behavior (Deci and Ryan, 2000). Within the entrepreneurship field, Herron and
Robinson (1993) identified entrepreneurial motivation as a key factor influencing venture
performance. Specifically, accumulating wealth, monetary compensation and building equity
in the firm (i.e., external contingencies) have long been recognized as important propellers of
entrepreneurial behavior (Langan-Fox and Roth, 1995; Shepherd and DeTienne, 2005).
According to Campbell (1992), individuals initiate, maintain and develop businesses if their
expected present value of entrepreneurship exceeds that of being an employee. Within SDT,
we know that when behavior is thus motivated, it suggests the financial performance of the
venture becomes instrumental to attain personally desired outcomes (e.g., wealth
accumulation, outside approval) or to avoid undesired ones (e.g., business failure, shame,
guilt). Therefore, we believe that with higher levels of controlled motivation, entrepreneurs,
either alone or as part of a team, will be increasingly energized into actions that foster new
ventures’ financial performance.
Yet, not all entrepreneurial behavior is coerced or seduced by (introjected) external
objectives. Kuratko et al. (1997: p.31) established that goals “of both an intrinsic and
extrinsic nature” are vital for sustaining entrepreneurship. Similarly, in their meta-analysis,
Carsrud and Brännback (2011) argue that while most researchers assume entrepreneurship to
be a pursuit of instrumental economic goals, substantial evidence exists of people engaging in
entrepreneurship without any apparent (dominant) reward other than task enjoyment. There
exists, however, relatively little research that has explored how autonomous entrepreneurship
motivation impacts new venture performance. Turning to educational psychology literature,
we learn that when activities are experienced as spontaneously satisfying and/or personally
important, people tend to persistently exert effort, be eager to acquire additional knowledge,
and show improved creativity (Deci et al., 2001; Wigfield et al., 2004). Since autonomous
motivated behaviors satisfy SDT’s basic needs of competence, autonomy and relatedness
(Deci and Ryan, 1985), engaging in such activities facilitates positive outcomes such as task
engagement, work performance, psychological well-being and behavioral persistence (Baard
et al., 2004). Extrapolating this research to an entrepreneurship context, we posit that
entrepreneurs with advanced levels of autonomous motivation will exert more time and effort
on business development, display more creativity and task engagement, and achieve higher
functional effectiveness, which ultimately should reflect in new ventures’ financial
performance. In sum, we hypothesize that while some entrepreneurs in EFTs may be (mainly)
driven by external contingencies (controlled motivation) and others by task enjoyment and
adopted job-attributes (autonomous motivation), they all share a commitment to pursue the
business opportunity, develop the new venture and sustain firm ownership. Hence, we believe
that both types of motivation will trigger EFTs to promote new ventures’ financial
performance.
6
Hypothesis 1: EFTs’ level of entrepreneurship motivation is positively associated with
new venture financial performance. This will be reflected by the positive effects of:
a) EFT autonomous entrepreneurship motivation, and
b) EFT controlled entrepreneurship motivation.
Despite the analogous anticipated association of EFTs’ autonomous and controlled
motivation with new venture financial performance, we know from lab experiments and
research in other domains that both types of motivation may differ in terms of their
magnitude and intensity. Gagné and Deci (2005) argue that work climates that encourage
intrinsic motivation or a profound internalization of extrinsic motivation yield improved
performance compared to climates that are control-oriented. Particularly for activities
demanding disciplined engagement, intellectual flexibility and complex problem solving
(e.g., activities related to new venture development), autonomous motivation has turned out
to be a better predictor of effective performance (Baard et al., 2004). Individuals appear to be
most creative when they find rewards in the task itself rather than when the task functions as
a means to an end (Amabile et al., 1990). In contrast, with regards to mundane tasks,
controlled motivation has been found to better facilitate short-term performance (Grolnick
and Ryan, 1987). Autonomous motivation produces better quality responses when
confronting multifaceted situations with high ambiguity. In such situations, which are
inherent to entrepreneurship, autonomy-oriented individuals appear to manage complex
problems better than their control-oriented colleagues (Erez et al., 1990). Not only do they
share a greater awareness of the environment, they also tackle unexpected events more
successfully. What is more, due to higher levels of excitement they are willing to dedicate
more personal resources to task fulfillment (e.g., effort and attention).
Building on the above arguments, we assume that EFTs’ autonomous
entrepreneurship motivation will exert a stronger influence on new venture financial
performance compared to their controlled entrepreneurship motivation. Thus, we
hypothesize:
Hypothesis 2: EFTs’ level of autonomous entrepreneurship motivation has a stronger
effect on new venture financial performance than EFTs’ level of controlled
entrepreneurship motivation.
Next to financial performance we also focus on new venture operational performance in the
form of innovation (bringing new goods and services to the market). Innovation is important
for firms, including new ventures (De Winne and Sels, 2010). It fuels organizations’
competitive advantage and stimulates growth and survival. Innovation is, therefore, to some
degree a goal for many new ventures. Yet, it also involves insecurity and risks (Smith et al.,
2005) and requires (strategic) perseverance. By looking at innovation as a goal, it is tied to
entrepreneurship motivation (Baum and Locke, 2004). More specifically, controlled
entrepreneurship motivation is expected to be negatively linked to new venture innovation.
After all, innovation is a goal rather distant in time, and such goals usually do not generate
the beneficial (control-oriented) motivational effects of short-term goals (Wood and Bandura,
1989). Past research, however, suggests that even controlled motivation could have a positive
effect on creativity and innovation (Eisenberger et al., 1999). More recent research has
refined that idea and made this effect contingent upon the rewards attached to the innovation
and/or other circumstantial conditions (Choi, 2004; Prabhu et al., 2006). In view of the
insecurity and risks attached to new venture innovation we expect that the possible rewards
attached to it are not strong enough to spur controlled motivation. Instead, stirred by the
prospect of accumulating wealth, control-motivated EFTs might shorten their time
7
perspectives and lucratively consume available means of production. Although such a modus
operandi does not particularly favor firm persistence, it may cause more control-oriented
EFTs to financially outperform primarily autonomous-oriented teams in the short term. The
level of controlled motivation, thus, is expected to reflect the EFTs’ averseness to any actions
that might endanger contingent monetary rewards (Deci et al., 2001; Ryan and Deci, 2000).
Autonomous motivation, on the other hand, should trigger new venture innovation.
After all, entrepreneurs can make up their own agenda (Baum and Locke, 2004). If they
consider innovation as something of genuine value, and anticipate that their efforts will lead
to innovation performance, then autonomous motivation will play a facilitating role in their
behavioral scheme. Aspects of autonomous motivation have been shown to correlate strongly
to expectancy beliefs (Gu et al., 2011). This line of reasoning is not entirely new; past
research has identified intrinsic or autonomous motivation elements as key ingredients to
individual innovation and creativity (Prabhu et al., 2006). Aimed at new venture continuity
and long-term fruition, rather than short-term profitability, autonomous motivation is more
likely to promote strategic investments, calculated risk-taking and the development of a
sound business foundation in order to achieve more challenging self-set goals (Watson et al.,
1993). While such actions foster long-term organizational responsiveness, they are
detrimental to new ventures’ short-term financial results. As such, we formulate the following
hypothesis:
Hypothesis 3: EFTs’ level of entrepreneurship motivation is associated with new
venture innovation performance. This will be reflected by:
a) The negative effect of EFT controlled entrepreneurship motivation, and
b) The positive effect of EFT autonomous entrepreneurship motivation.
Given that distinctive mechanisms govern autonomous and controlled motivation (Gagné and
Deci, 2005), and that a mixture of both types of motives is bound to occur within EFTs, we
anticipate that team motivation heterogeneity can affect the earlier hypothesized relationships
between autonomous/controlled entrepreneurship motivation and new venture
financial/innovative performance. Williams and O’Reilly (1998) distinguish two dominant
perspectives in the research on performance effects of team diversity: the ‘social
categorization’ perspective and the ‘information/decision-making’ perspective. The rationale
behind the social categorization perspective is that people use perceived similarities and
differences to categorize themselves and others into ‘in’ and ‘out’ social groups. As the
current study focuses on small EFTs, of which all members own and manage part of the new
venture, thinking along the lines of such categorizations is not very useful. The second view,
that is the information/decision-making perspective, is, however, more promising. This
tradition focuses on the task-related aspects of team processes. Its research interest lies at
team members’ exchange, discussion and integration of ideas, knowledge and insights
relevant to the tasks at hand (Van Knippenberg et al., 2004). Advancing idea and knowledge
diversity within teams as an informational resource, the information/decision-making
perspective argues that more diverse groups may outclass more homogeneous ones.
Contrary to the idea of social groups, it seems reasonable to assume that EFTs’
activities involve strong information/decision-making components. This setting, thus, creates
a rich soil for team diversity or heterogeneity to bear fruits. In the current study, team
motivation heterogeneity refers to differences among EFT members on deeper-level
controlled (e.g., differences in terms of external pressures and contingencies driving
behavior) and/or autonomous motives (e.g., differences in personal values, beliefs, opinions
and tasks regarded as ‘fun’ to do). Though some authors, while building on social identity
theory and the similarity-attraction proposition, assume that EFT members are similar to one
8
another (e.g., Cooper and Artz, 1995; Hmieleski et al., 2012), we maintain that they are, in
fact, less (motivational) homogeneous (and, therefore, more heterogeneous) than one would
initially assume. After all, motivational differences concern deep-level topics. As a result, it
may take a considerable amount of time and interaction for entrepreneurship motivational
diversity to emerge.
Key to yielding benefits from team motivation heterogeneity is not the mere presence
of diverse points of view, but the adequate use of the rich information embedded in these
differences (Van Knippenberg et al., 2004). While it is often proposed that conflicts resulting
from heterogeneity generate performance benefits (e.g., Jehn et al., 1999), we posit that it is
the creative processing of members’ diverse viewpoints and information that is essentially
beneficial (Van Knippenberg et al., 2004). In fact, to date, research findings on the conflict
perspective of team heterogeneity are inconclusive (Horwitz and Horwitz, 2007). So, if we
are to understand the influence of EFTs’ heterogeneity on performance, we need to perceive
EFTs as information processing entities (Hinsz et al., 1997).
Capitalizing on the information/decision-making perspective of team diversity, we
contend that EFTs will process information differently depending on (members’ motivation
for) the task at hand and its targeted outcome. In general terms, information processing
encompasses the learning and exchange of perspectives among group members, the
individual-level processing of this information, the feeding back of processing results to the
group, and the discussion and integration of member feedback (Van Knippenberg et al.,
2004). Though these information-processing requirements are highly demanding, they have
been argued to be conducive in the context of complex, cognitive and risky innovations.
Whenever learning processes resemble the information processing of the team, as is the case
with new venture innovations, diverse standpoints, orientations and beliefs are essential to
reach a greater understanding of strategic alternatives (Amason, 1996). In the current study,
we believe that the value, idea and knowledge diversity embedded in the autonomous and
controlled entrepreneurship motives of EFT members will challenge the status-quo, spur team
creativity in solving problems, and ultimately encourage new venture innovation. Hence, we
hypothesize:
Hypothesis 4: Motivation heterogeneity acts as a positive moderator of the
relationship between EFTs’ level of autonomous/controlled entrepreneurship
motivation and new venture innovation. This will be reflected by the positive
moderating effects of:
a) EFT autonomous entrepreneurship motivation heterogeneity, and
b) EFT controlled entrepreneurship motivation heterogeneity.
Given the requirements listed above, it is not surprising that team information processing is to
some extent an ambiguous and often time-consuming process (Nemeth and Staw, 1989). In
fact, these requirements may prevent a team from engaging in speedy decision-making and
achieving swift compromises. Hence, differences in individually valued goals, beliefs and
cognitive schemas that stem from motivation heterogeneity may disturb the setting for short-
term oriented tasks or objectives that capitalize on such efficient decision-making (e.g., short-
term financial performance). What is more, in order to secure team effectiveness,
heterogeneously motivated EFTs will have to devote some resources (e.g., time and effort) to
the reinstatement of group coherence and team consent. Equally motivated team members, in
contrast, because of their mutual beliefs, values and attributes, might find it much easier to
develop and sustain short-term communication patterns, group objectives and modi operandi.
While such communality or homogeneity among EFT members does not particularly favor
new venture innovation or long-term fruition, we believe it may enable less heterogeneously
9
motivated EFTs to facilitate new venture financial performance, at least in the short term
(with the latter being related to this study’s dependent). We, therefore, hypothesize:
Hypothesis 5: Motivation heterogeneity acts as a negative moderator of the
relationship between EFTs’ level of autonomous/controlled entrepreneurship
motivation and new venture financial performance. This will be reflected by the
negative moderating effects of:
a) EFT autonomous entrepreneurship motivation heterogeneity, and
b) EFT controlled entrepreneurship motivation heterogeneity.
Methodology
Sampling procedures
The sample for this research originates from START 2009, an extensive cross-sectional
survey on new ventures located in Flanders, Belgium. This is a biennial population survey of
Flemish incorporated companies that have been in business for one to three years, are active
within various economic sectors and, in 2009, had a minimum of one and a maximum of 49
employees. The primary source of data involved a structured interview with each of the new
ventures’ managers. Face-to-face interviewing allowed for the collection of comprehensive
information on their educational background, career trajectory and entrepreneurship
motivations. General information on the new ventures was collected using a questionnaire
that was mailed prior to the interviews. Finally, financial performance information was
captured using audited information from Bel-first, which denotes a financial database holding
information on the company accounts of all firms incorporated under Belgian law.
The total research population of new ventures consisted of 3183 firms in 2009. Due to
obsolete company data, 259 new ventures could not be reached. Out of the 2924
questionnaires mailed, 453 usable company responses (response rate of 15.5%) and 490
owner interviews were obtained. Within 42% of the responding ventures (190 companies),
daily management was shared by at least two individuals. To be included as an observation in
this study, data on both the venture and its founders were required. In order to be considered
a member of an EFT, respondents had to have an active hand in the founding of the venture,
own an equity stake of at least 10% and assume a key role in the venture’s current strategic
decision-making. Interview responses were required from all members of the EFT. Due to a
lack of data on the new venture and/or on one of its founders, 97 companies had to be
excluded from further analysis. Another 10 businesses were omitted because their current
managers did not meet the above criteria. The remaining 83 companies, representing 176
founders, were further reduced because of the use of listwise exclusion during statistical
procedures. This resulted in a final sample of 66 ventures, representing 142 founders. Teams
ranged in size from two to four members.
Tests between respondent and non-respondent ventures revealed no significant
differences regarding organization age and size. Using chi-square differences and t-tests, no
differences emerged regarding industry, size and organization age between the firms used in
the analyses and those that were eligible yet excluded because of missing values. Similarly,
no differences were detected pertaining to average age, industry experience and
entrepreneurship motivation of sampled EFTs and teams whose members did not all meet the
selection criteria. Finally, a means difference test showed no evidence of financial or
operational performance variations. While this evidence does not eliminate the concern of
possible non-response bias, it does indicate a certain level of representativeness of our
sample.
10
While sample size may be of concern in this study, we would like to point out that
similar sample sizes have been characterizing much of the team-based literature. For
example, Olsen et al. (2006) analyzed 66 teams in their study on the mediating role of
strategic choice between team heterogeneity and firm performance, as did Eisenhardt and
Schoonhoven (1990) when investigating the association of environment, technical innovation
and top management team characteristics on organization sales. West and Schwenk (1996)
studied 65 firms in search of moderating effects of environmental turbulence on the
relationship between top management team consensus and firm performance. Finally,
Talaulicar et al. (2005) called upon 56 teams to examine the influence of new venture team
organization and processes on the comprehensiveness and speed of strategic decision-
making.
Measures
New venture financial performance. Following Zahra et al. (2000), we addressed three issues
in measuring new venture performance. First, we decided to make use of a lagged
performance measure by averaging financial performance data over the company’s last two
years. Given this study’s focus on new ventures, a two-year time period should capture
unusual events in the market, while avoiding the introduction of noise (e.g., changing
industry structures and firm strategy variations) (Ramanujam and Varadarajan, 1989).
Second, we acquired audited financial information from a separate database more than one
year after the owner interviews and company questionnaires were concluded. To lessen the
possible effects of common method variance we decided not to make use of survey method
data. Instead, using the company’s unique identification number, we retrieved the annual
accounts of each company from the aforementioned Bel-first database. The final issue
concerned the selection of an adequate financial performance criterion. We adopted “return
on assets” (ROA) as an objective accounting measure of new venture performance. As an
indicator of the profitability of a company relative to its total assets, the ROA sheds light on
the efficiency of EFTs at using company assets to generate earnings. As a result, ROA
accurately reflects entrepreneurial effectiveness and the ability of EFTs to create wealth.
New venture innovation performance. Traditional conceptions of business
performance largely center on the use of financial accounting measures. These measures
retrospectively address ventures’ fulfilment of specific economic goals. They, unfortunately,
do not shed light on firms’ internal effectiveness nor do they reveal any mechanisms that
drive performance (Venkatraman and Ramanujam, 1986). We therefore aimed to complement
our financial dependent with an operational one to better capture the concept of new venture
performance.
One of a firm’s key parameters important for competitive success is its innovation
performance (Crossan and Apaydin, 2010). Being directly tied to the company’s value
creation, (product and process) innovations are likely to steer new ventures away from
potential familiarity, maturity and propinquity competence traps (Liao et al., 2003). They
elicit organizational learning and new knowledge accumulation (Maes and Sels,
forthcoming), which strengthens new ventures’ delicate competitive position.
Research has shown that various operationalizations of organizational innovation
exist. Following Crossan and Apaydin (2010), we selected three different measures to capture
three theoretically meaningful dimensions of new ventures’ innovation performance: product
portfolio, process technologies and served markets. First, we assessed the number of distinct
types of products or services the firm was offering at the time of the survey. This information
was transformed into a dummy variable differentiating single-product from multiple-product
firms. Second, we surveyed the number of process technologies used to create and deliver
these products or services. Possible values included ‘one dominant technology’ (46.97%),
11
‘two distinct technologies’ (12.12%), or ‘three or more distinct technologies’ (40.91%).
Third, concentrating on multi-market ventures, we examined the diversity of the markets to
which the firm targeted its products or services. Answers were scored on a five-point scale
ranging from ‘very equivalent’ (1) to ‘not equivalent at all’ (5). Together, the above measures
provide a comprehensive view of new ventures’ innovation performance. An aggregate ten-
point measure that was the sum of the above three measures was calculated for each firm
(e.g., 10 = multiple-product new venture, relying on minimum three distinct process
technologies, thereby serving disparate markets).
EFT level of autonomous and controlled entrepreneurship motivation. To assess
EFTs’ level of controlled and autonomous entrepreneurship motivation, we first determined
the entrepreneurship motivation of each of the venture’s founders. For this we adopted the
Gagné et al. (2010) and Vansteenkiste et al. (2009) inventory, which is grounded in SDT.
This 12-item inventory assesses four dimensions of work motivation, of which two pertain to
autonomous motivation (‘intrinsic motivation’ (Cronbach’s alpha = .764) and ‘identified
regulation’ (Cronbach’s alpha = .729)) and two to controlled motivation (‘external regulation’
(Cronbach’s alpha = .692) and ‘introjected regulation’ (Cronbach’s alpha = .851)). Prior
research has already demonstrated this inventory’s high levels of construct and concurrent
validity and internal consistency. Although initially aimed at measuring people’s motivation
for work-related behavior in the context of established organizations (Gagné et al., 2010), the
items are equally appropriate for measuring motivation of business founders in the context of
new ventures. Items were scored on a five-point Likert scale ranging from 1 (‘not important
at all’) to 5 (‘extremely important’). Using the formula suggested by Maes et al. (2005), we
determined for each founder two factors with scale ranges from 0 to 100: one representing
autonomous entrepreneurship motivation (Cronbach’s alpha = .827; 6 statements) and
another representing controlled entrepreneurship motivation (Cronbach’s alpha = .791; 6
statements). The adopted formula is displayed in the Appendix with all items and factor
loadings. To determine the EFTs’ level of autonomous (controlled) entrepreneurship
motivation, we calculated the average autonomous (controlled) motivation among its
members.
EFT entrepreneurship motivation heterogeneity. Motivation heterogeneity has seldom
been measured in the context of EFTs. To transform general EFT entrepreneurship
motivation information into team-based heterogeneity variables, we adopted Allison’s (1978)
coefficient of variation. This heterogeneity coefficient is constructed of the standard deviation
divided by the mean. A high score on this coefficient refers to high team heterogeneity,
whereas a low score denotes low heterogeneity. Using Allison’s (1978) formula, we
determined EFTs’ intrinsic and extrinsic entrepreneurship motivation heterogeneity.
Control variables. To isolate our hypotheses from possible rival explanations and to
minimize extraneous variation, we included the following organizational characteristics as
control variables: firm continuation, EFT size, industry, and environmental turbulence.
Furthermore, to explore the performance contribution specific to EFT motivations, we also
controlled for more traditional demographic diversity variables (e.g., gender, education,
industry experience).
Although all ventures in our sample are between one and three years of (legal) age,
not all of them are de novo firms (e.g., take-over of a bankrupt business). This implies that
the business activities could have been carried out before the current organizations were
legally established. We control for the possibility of firm continuation by including a dummy
variable indicating whether the business activities were already operational before the current
venture was founded. Because team heterogeneity and group size are positively associated
(Allison, 1978), it is imperative to control for EFT size. Prior work has indicated that larger
teams are linked to better performance, both at group level and firm level (Eisenhardt and
12
Schoonhoven, 1990). In order to prevent turbulence-originated bias in venture performance,
we introduced the dynamic nature of the venture’s environment as a control variable
(Ucbasaran et al., 2003). The scale items used to measure environmental turbulence are
adapted from Zahra (1993). Respondents were asked to rate six environmental statements on
a five-point Likert scale. Possible answers varied from ‘entirely disagree’ (1) to ‘entirely
agree’ (5). Again using the formula suggested by Maes et al. (2005), we created a factor with
scale ranges from 0 to 100 (Cronbach’s alpha = .855). The adopted formula is displayed in
the Appendix, along with all items and factor loadings. Because firms in our sample belong
to various industries, each with their own characteristics, we developed a series of dummy
variables to control for the different market conditions within each industry. Five dummies
were included in our analyses, using the manufacturing sector as a reference category.
Because Ucbasaran et al. (2008) have found a positive relationship between founder
experience and firm performance, we intended to control for EFTs’ entrepreneurship
experience. However, since at least one member of every EFT had already actively
participated in new venture creation, this dummy variable was excluded.
We again used Allison’s (1978) coefficient of variation to compute the age and
industry experience heterogeneity between the business founders. Due to collinearity issues
among the two independents, we were forced to exclude age and age heterogeneity from
further analysis. The remaining variables of heterogeneity (e.g., gender and educational
background) were developed using the Herfindal-Hirschman coefficient, also known as the
Blau categorical index (1977) (H = 1 - ?p
i
²). Before calculating the coefficient’s score, we
assigned the founders’ educational background to one of the following categories: arts,
sciences, engineering, business and economics, law and other. The above formula produced a
measure of heterogeneity with its complement being a measure of homogeneity. Though we
are unable to control for romantic couples, sample conditions did prevent the inclusion of
ventures in which a spouse was registered as a company partner without being a business
founder or assuming an active management role.
3.3 Statistical procedures
Hierarchical multiple regression was used as the statistical procedure to test our hypotheses.
As far as multivariate statistical tools are concerned, this technique has been subject to an
impressive set of tests of assumptions (Belsley et al., 1980). It allows investigating the
contribution above and beyond variables already entered into the regression equation. It also
enables us to examine the statistical influence of several variables at once. The variables were
mean-centered before any of the interaction terms were created. The highest VIF statistic
encountered in the models discussed below was 3.873, which is below the recommended
maximum value of 5 (Moreno and Casillas, 2008).
4. Results
Table 1 lists the means, standard deviations and bivariate correlations of this study’s
variables. All correlations are below .80 in absolute value, which is again an indication
against the possible presence of multicollinearity (Hair et al., 1998). The results of the
hierarchical regression analyses are displayed in Table 2 (financial performance) and Table 3
(innovation performance). Models 1, 2, 8 and 9 represent the ‘control models’, which only
include the aforementioned organizational control variables (Models 1 and 8), and the
traditional demographic diversity variables (Models 2 and 9). Models 3 and 10 relate to the
main effects, together with the control variables. We estimated each of the hypothesized
interaction terms in Models 4 to 7 (financial performance) and Models 11 to 14 (innovation
performance).
13
Table 1. Descriptive Statistics and Correlations
Variables Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
1. Return on assets 5.68 20.82 1
2. Innovation 4.98 3.13 -.14 1
3. Agriculture .08 .27 -.01 -.15 1
4. Construction .26 .44 .00 .20 -.17 1
5. Manufacturing .42 .50 .15 -.13 -.25
*
-.51
**
1
6. Transportation .03 .17 -.06 -.23 -.05 -.10 -.15 1
7. Banking and insurances .05 .21 -.10 .10 -.06 -.13 -.19 -.04 1
8. Professional services .17 .38 -.11 .09 -.13 -.26
*
-.38
**
-.08 -.10 1
9. Start-up continuation .00 .49 .02 -.10 .23 .05 -.06 .14 .03 -.22 1
10. EFT size 2.33 .44 .13 -.11 -.10 -.21 .19 -.06 -.08 .13 -.15 1
11. Environmental turbulence 46.97 22.10 -.04 .31
*
-.17 .18 -.19 -.23 -.07 .30
*
-.38
**
.01 1
12. EFT industry experience 13.90 9.31 .10 -.24 .14 -.10 .11 .13 .25
*
-.34
**
.35
**
-.09 -.30
*
1
13. EFT gender heterogeneity 20.94 24.63 -.09 .02 .22 .06 .00 -.15 -.04 -.14 .27
*
-.13 -.04 -.01 1
14.
EFT educational background
heterogeneity
20.96 24.62 -.17 .18 .11 -.01 -.24 -.15 -.04 .35
**
.14 -.11 .24 -.03 -.04 1
15.
EFT industry experience
heterogeneity
38.66 32.18 -.05 -.26
*
-.07 -.18 .36
**
-.20 -.23 .01 -.12 .40
**
.09 -.20 .07 -.02 1
16. EFT autonomous motivation 76.56 14.86 .20 -.16 -.13 -.12 .17 .01 .02 -.01 .03 .07 .06 .01 -.18 -.07 .29
*
1
17. EFT controlled motivation 35.82 17.11 .23 -.30
*
.17 .02 .11 .02 -.02 -.29
*
.07 -.03 .07 .04 .04 -.15 .17 .26
*
1
18.
Autonomous motivation
heterogeneity
23.87 26.53 -.11 .12 -.08 -.19 -.09 .14 .17 .23 -.09 .08 .00 -.10 -.35
**
.28
*
-.07 .33
**
-.23 1
19.
Controlled motivation
heterogeneity
112.36 60.44 -.06 .13 -.26
*
-.12 .03 .03 .00 .28
*
-.02 .09 -.05 -.04 -.17 .11 .08 .46
**
-.72
**
.48
**
Notes: **. Correlation is significant at the .01 level (2-tailed). *. Correlation is significant at the .05 level (2-tailed).
14
Table 2. Results of Hierarchical Regression Models of New Venture Financial Performance
Variables Model 1 Model 2
Model 3 Model 4 Model 5 Model 6 Model 7
Control variables: Organizational characteristics
Agriculture -.024 .008 -.034 -.039 -.064 -.019 -,047
Construction .045 .081 .044 .022 -.009 .058 ,023
Transportation -.062 -.132 -.153 -.144 -.156 -.125 -,123
Banking and insurances -.098 -.155 -.182 -.180 -.194 -.150 -,197
Professional services -.115 -.049 -.015 -.021 -.042 -.010 ,050
Start-up continuation .032 .085 .016 -.022 -.047 .002 ,046
EFT size .131 .164 .203 .215 .245 .218 ,178
Environmental turbulence -.012 .059 -.024 -.045 -.085 -.040 -,120
Control variables: Traditional diversity variables
EFT industry experience .095 .099 .105 .090 .070 ,104
EFT gender heterogeneity -.118 -.050 -.013 .026 -.078 -,097
EFT educational background heterogeneity -.167 -.108 -.088 -.039 -.056 -,135
EFT industry experience heterogeneity -.178 -.288* -.304* -.330* -.311* -,263
Main effects
EFT level of autonomous entrepreneurship motivation .196* .258* .260* .261** ,183
EFT level of controlled entrepreneurship motivation .206* .235* .266** .168 ,112
Moderators
Autonomous entrepreneurship motivation heterogeneity .121 -.044
Controlled entrepreneurship motivation heterogeneity -.153 -,322**
Two-way interactions
EFT level of autonomous entrepreneurship motivation x
autonomous entrepreneurship motivation heterogeneity
-.212
EFT level of controlled entrepreneurship motivation x
controlled entrepreneurship motivation heterogeneity
-.369**
F-Change .387 1.004 2.519** .602 .533 .801 4.630**
R²
.051 .118 .198 .207 .216 .210 .278
Notes: Standardized coefficients are shown (two-tailed, with directional hypothesis entries one-tailed); N = 66 start-ups, representing 142 founders;
***. Significant at the .01 level - **. Significant at the .05 level - *. Significant at the .10 level.
15
Table 3. Results of Hierarchical Regression Models of New Venture Innovation Performance
Variables Model 8 Model 9
Model 10 Model 11 Model 12 Model 13 Model 14
Control variables: Organizational characteristics
Agriculture -.177 -.154 -.094 -.090 -.044 -.102 -,081
Construction -.159 .000 .016 .034 .090 .009 ,035
Transportation -.220* -.208 -.191 -.198 -.177 -.206 -,208
Banking and insurances .047 .070 .081 .079 .103 .063 ,097
Professional services -.055 -.132 -.197 -.192 -.155 -.200 -,244
Start-up continuation .017 .031 .072 .101 .147 .080 ,046
EFT size -.100 .036 .011 .002 -.052 .003 ,033
Environmental turbulence .228 .213 .288** .304** .376** .296** ,356**
Control variables: Traditional diversity variables
EFT industry experience -.270** -.279** -.283** -.256* -.263* -,288**
EFT gender heterogeneity .045 .014 -.014 -.085 .030 ,045
EFT educational background heterogeneity .150 .104 .088 -.001 .075 ,134
EFT industry experience heterogeneity -.382*** -.318** -.306** -.260* -.305** -,342**
Main effects
EFT level of autonomous entrepreneurship motivation -.018 -.066 -.068 -.054 ,004
EFT level of controlled entrepreneurship motivation -.279** -.301** -.357*** -.258** -,216**
Moderators
Autonomous entrepreneurship motivation heterogeneity -.094 .205
Controlled entrepreneurship motivation heterogeneity .084 ,209
Two-way interactions
EFT level of autonomous entrepreneurship motivation x
autonomous entrepreneurship motivation heterogeneity
.382*
EFT level of controlled entrepreneurship motivation x
controlled entrepreneurship motivation heterogeneity
,273**
F-Change 1.524 2.703** 2.664* .466 2.318 .314 3.169*
R²
.176 .316 .381 .386 .414 .384 .422
Notes: Standardized coefficients are shown (two-tailed, with directional hypothesis entries one-tailed); N = 66 start-ups, representing 142 founders;
***. Significant at the .01 level - **. Significant at the .05 level - *. Significant at the .10 level.
16
Hypothesis 1a stated that the EFTs’ level of autonomous entrepreneurship motivation is
positively associated with new venture ROA. Hypothesis 1b suggested a similar relationship
for the EFTs’ level of controlled entrepreneurship motivation. Based on Model 3 of Table 2,
we can corroborate both hypotheses. While the regression coefficient of EFT autonomous
entrepreneurship motivation is significant and positive (? = .196), it appears not to be
significantly different from the one representing EFT controlled motivation (? = .206).
Consequently, we cannot conclude that an EFT’s level of autonomous entrepreneurship
motivation exerts a stronger influence on new venture ROA than its level of controlled
entrepreneurship motivation. Consequently, no support is found for Hypothesis 2.
Hypothesis 3a suggested that EFT autonomous entrepreneurship motivation is
positively associated with new ventures’ innovation performance. Model 10 of Table 3,
however, indicates that this relationship is insignificant (? = -.018). Hypothesis 3a receives,
therefore, no support. Hypothesis 3b suggested that the EFTs’ level of controlled
entrepreneurship motivation is negatively associated with new venture innovation. Model 10
of Table 3 empirically validates this hypothesis (? = -.279). We, thus, find support for
Hypothesis 3b.
Hypothesis 4a proposed that autonomous entrepreneurship motivation heterogeneity
positively moderates the relationship between EFT autonomous motivation and new venture
innovation. Hypothesis 4b stated the same for member controlled motivation heterogeneity
regarding the relationship between EFT controlled motivation and new venture innovation. In
other words, we expect teams with a higher level of entrepreneurship motivation to benefit
from higher motivation heterogeneity in terms of new venture innovation, whereas teams
with a lower level of entrepreneurship motivation should benefit from lower motivation
heterogeneity. We learn from Models 11 to 14 of Table 3 that both cross-products involving
motivation heterogeneity are positive and significant (? = .382; ? = .273). The graphical
representation of these interactions (Figures 1 and 2 below) confirms Hypotheses 4a and 4b.
Figure 1. Interaction Effect of EFT Autonomous Entrepreneurship Motivation and
Autonomous Entrepreneurship Motivation Heterogeneity on New Venture
Innovation
17
Figure 2. Interaction Effect of EFT Controlled Entrepreneurship Motivation and
Controlled Entrepreneurship Motivation Heterogeneity on New Venture Innovation
Hypothesis 5a suggested that autonomous motivation heterogeneity acts as a negative
moderator of the relationship between EFT autonomous entrepreneurship motivation and new
venture ROA. Hypothesis 5b proposed a similar influence for controlled motivation
heterogeneity on the relationship between EFT controlled motivation and new venture ROA.
In other words, we assumed that the anticipated positive main effects of team
entrepreneurship motivation on new venture ROA would be stronger if motivation
heterogeneity is low, and weaker if it is high. As shown by Models 4 to 7 of Table 2,
evidence only emerges for the cross-product involving controlled entrepreneurship
motivation heterogeneity (? = -.369). The graphical representation of this interaction (Figure
3 below) indicates that the relationship of EFT controlled entrepreneurship motivation with
new venture ROA is negative for EFTs with higher controlled motivation heterogeneity, and
positive for teams experiencing lower controlled motivation heterogeneity. Our results, thus,
corroborate Hypothesis 5b, whereas no support is found for Hypothesis 5a.
Figure 3. Interaction Effect of EFT Controlled Entrepreneurship Motivation and
Controlled Entrepreneurship Motivation Heterogeneity on New Venture Return on
Assets
18
Discussion
The research stream examining the complex relationship between team diversity and team
outcomes is impressive in volume (Horwitz and Horwitz, 2007). Since most new ventures are
founded by teams of entrepreneurs (McMullen et al., 2008), the importance of studying the
performance effects of EFTs’ characteristics increases. The present study is to be situated in
this domain. However, instead of focusing on bio-demographic EFT characteristics, we have
explored deeper-level motivation drivers of new venture performance, which have been
neglected in prior research (Van Knippenberg et al., 2004). Further, we introduced the aspect
of EFT motivational diversity into our arguments. In doing so, we believe we may also claim
to contribute to EFT diversity literature.
The specific objective of this study was to empirically examine the extent to which
entrepreneurship motivations of an entrepreneurial founding team affect new venture
financial and innovation performance. We first tested EFTs’ level of autonomous and
controlled entrepreneurship motivation as drivers of new venture return on assets and
innovation. We then introduced and empirically estimated the effect of two moderators
representing team motivation heterogeneity. Five main hypotheses were tested. In general,
support for these hypotheses was substantial. First and foremost, the study’s results
corroborate the distinction between autonomous and controlled entrepreneurship motivation.
We have demonstrated that both types can generate different effects. We can, therefore, rally
with Gagné and Deci (1995) against a unitary view on motivation. Our results also illustrate
the importance of EFTs’ motivational diversity or heterogeneity for new venture
performance. Deep-level diversity within EFTs is thus not to be neglected. We discuss our
findings in two subsequent sections: EFT autonomous motivation and EFT controlled
motivation.
EFT autonomous motivation
Within this study, a significant contribution of the EFTs’ level of autonomous
entrepreneurship motivation to new venture return on assets emerged. Contrariwise, no
significant link was found between the level of autonomous motivation and new venture
innovation. The latter finding comes somewhat unexpectedly. We discern the following
explanation for this absent link: Looking at Table 1, we observe that our sampled EFTs, on
average, display a high level of autonomous motivation (76.56). The level of controlled
motivation, on the other hand, is much lower (35.82), while its standard deviation is higher
(17.11 vs. 14.86). Our sample is thus characterized by a high level of autonomous motivation.
An explanation for the absent link lies in the close resemblance between becoming an
entrepreneur (e.g., starting up a new venture) and the innovation activities within young
organizations (e.g., bringing new goods and services to the market). By entering
entrepreneurship, EFT members’ general tasks resemble the tasks to be done when pursuing
innovation. Hence, it could be that the level of autonomous motivation within our sample is
too high for any additional autonomous motivation to trigger innovation effects. Securing
new venture return on assets, on the other hand, is far more of a managerial than a first-stage
(young) entrepreneurial activity. Resemblance between entrepreneurship and the latter is,
therefore, less outspoken, which allows for additional innovation effects to occur.
Consistent with our assumptions, our results revealed that autonomous motivation
heterogeneity acts as a positive moderator of the relationship between EFTs’ level of
autonomous motivation and new venture innovation. In other words, high motivation
heterogeneity was found to be instrumental to new venture innovation. Low heterogeneity, on
the contrary, appeared to be detrimental. Regarding new ventures’ financial performance, no
significant influence of autonomous motivation heterogeneity emerged. This finding seems to
19
indicate that only the level of EFTs’ autonomous entrepreneurship motivation affects new
ventures’ wealth creation, and not the distribution of that motivation within the founding
team. Again, this could be related to the aforementioned high level of EFTs’ autonomous
motivation within our sample, combined with a relatively low degree of inter-member
autonomous motivation heterogeneity (23.87; see Table 1).
EFT controlled motivation
Apart from internal or strongly internalized autonomous motives, our findings also shed light
on several interesting control-oriented motivation effects. In line with our expectations, the
level of controlled motivation was found to facilitate new ventures’ financial performance, as
reflected in their return on assets, whereas it hindered their operational performance, in terms
of firm innovations. As far as the interaction effects are concerned, both hypothesized
relationships were corroborated by our results. While low control-oriented motivation
heterogeneity facilitated new venture return on assets, it increasingly hindered new ventures’
ability to bringing new goods and services to the market. High motivation heterogeneity, on
the other hand, was found to foster new venture innovation, yet at the expense of short-term
firm financial performance.
We point out that our findings did not reveal a stronger effect of autonomous
entrepreneurship motivation on new venture financial performance compared to controlled
entrepreneurship motivation. In search of an explanation for this unexpected outcome, we
again turn to educational psychology literature. Within this strand of literature, it has been
suggested that autonomous motivated individuals have a higher probability of adopting a
long-term perspective (Watson et al., 1993). Control-oriented people, in contrast, tend to be
reluctant to take any actions that could endanger contingent external rewards. Instead, they
prefer to focus attention on the here and now (Erez et al., 1990). Although such a short-term
time perspective does not particularly encourage EFTs’ contribution to long-term venture
performance, it may trigger a contribution of controlled motivation to financial performance
similar to the one of autonomous motivation. While we believe this to be a mere short-term
outcome, additional research on this subject is imperative to empirically confirm this line of
thought.
All in all, we can say the levels of EFTs’ autonomous and controlled motivation
generate different effects when it comes to innovation performance, yet similar (positive)
effects regarding (short-term) financial performance. Furthermore, the (moderating) effects of
both types of motivational heterogeneity run parallel for innovation performance, whereas
they differ for financial performance. This implies that for studies focusing on understanding
deeper-level drivers of new venture performance within multi-founder firms it is essential to
consider not only the average level of the driver but also its heterogeneity within the
entrepreneurial founding team.
Limitations and Future Research
When interpreting our study findings, the following caveats should be recognized. First,
because this study makes use of self-describing interview data, it might suffer from social
desirability biases. However, we aimed to reduce possible biases by obtaining information
from multiple respondents (all of the founders) and by supplementing self-describing
information with more objective data (questionnaire and Bel-first data). Second, we only
assessed the entrepreneurship motivation of successful business founders. Yet, motivational
differences existing among nascent entrepreneurial team members could be very substantial,
encouraging some members to quit the team, or to freeze differences in order to not
jeopardize the emergence of the new firm. Future research could investigate these ideas
through a longitudinal study of nascent EFT dynamics. In this respect, the social
20
categorization theory of team diversity could be adopted as well, next to the
information/decision-making perspective. Hence, our sample could be biased with positive
selection because it includes only teams that were successful in creating a new venture. Third,
given the cross-sectional nature of our research design, we cannot prove the direction of the
cause-effect relationships in our model. However, by making use of a data structure in which
information on the dependent originates from a distinct database at a later point in time, we
established that founding team characteristics precede new venture performance. Fourth,
though the current study examines firms that are active within different industries, it does not
shed light on ventures in the retail industry. With START 2009 being its fourth wave, the
START research program has historically excluded the retail industry from its design, which
in most economies is one of the major industries for small and new businesses. However, as
the Belgian (and Flemish) economy is grounded in small and medium-sized businesses, with
micro-businesses (< 10 employees) being especially prevailing (94%), the retail industry is
typically not considered a dominant industry for new businesses in Belgium (European
Commission, 2012).
In conclusion, we distinguish the following recommendations for future research in
this area, on top of those already mentioned above. First, we urge researchers to repeat this
study’s moderator hypotheses for other new venture outcomes. Not only should this shed
light on possible contradictory mechanisms, it may equally extend our knowledge on the
contingencies that surround EFTs’ effectiveness. Further, following Horwitz and Horwitz
(2007), we identify the exploration of possible curvilinear relationships between motivation
level/heterogeneity and performance outcomes as an interesting and important line of
research. Another promising research direction is to adopt other psychological constructs
aimed at directly capturing unobservable team characteristics. For example, future
contributions might investigate how trust and friendship among the business founders
constitute important preconditions for team functioning. Alternatively, this research could
advance our understanding of the impact of founding team heterogeneity on the development
of network relationships and social ties, which already have been argued to benefit new
ventures (Shane and Cable, 2002). Future research should also look into the role of team
dynamics on new venture emergence and performance. Team composition is not always
fixed. While our sampling conditions required interview responses from all members of the
EFT, new members may join the founding team, while others might decide to leave it. In
turn, these flows are likely to affect social integration and team coherence. They also reflect a
transfer of knowledge, perspectives and resources, and the impact of these factors is not yet
clearly understood. Moreover, we also advise future research to take the intermember
relationships into account. Whereas motivation and motivational differences clearly are
strong forces affecting team cohesion, other elements are also important in this respect. We
could, for instance, think of family versus non-family linkages among EFT members and of
the distribution of firm ownership among founders. Finally, research on founding team
characteristics could include the moderating role of time on venture formation and
emergence, as established by Steffens et al. (forthcoming).
21
References
Adams JS (1963) Toward an understanding of inequity. Journal of Abnormal and Social
Psychology 67(5): 422–436.
Alderfer CP (1972) Existence, relatedness, and growth. New York: Free Press.
Allison P (1978) Measures of inequality. American Sociological Review 43(6): 865–880.
Amabile TM, Goldfarb P and Brackfield SC (1990) Social influences on creativity:
Evaluation, coaction, and surveillance. Creativity Research Journal 3(1): 6–21.
Amason AC (1996) Distinguishing the effects of functional and dysfunctional conflict on
strategic decision making: Resolving a paradox for top management teams. Academy
of Management Journal 39(1): 123–148.
Anderson JC and Gerbing DW (1988) Structural equation modeling in practice: A review and
recommended two-step approach. Psychological Bulletin 103(3): 411–423.
Baard PP, Deci EL and Ryan RM (2004) Intrinsic need satisfaction: A motivational basis of
performance and well-being in two work settings. Journal of Applied Social
Psychology 34(10): 2045–2068.
Barringer BR and Bluedorn AC (1999) The relationship between corporate entrepreneurship
and strategic management. Strategic Management Journal 20(5): 421–444.
Baum JR and Locke EA (2004) The relationships of entrepreneurial traits, skills and
motivation to subsequent venture growth. Journal of Applied Psychology 89(4): 587–
598.
Belsley DA, Kuh E and Welsch RE (1980) Regression diagnostics: Identifying influential
data and sources of collinearity. New York: Wiley.
Blau P (1977) Inequality and heterogeneity: A primitive theory of social structure. New
York: The Free Press.
Cameron J (2001) Negative effects of reward on intrinsic motivation - A limited
phenomenon: Comment on Deci, Koestner, and Ryan (2001). Review of Educational
Research 71(1): 29–42.
Campbell CA (1992) A decision theory model for entrepreneurial acts. Entrepreneurship:
Theory & Practice 17(1): 21–27.
Carsrud A and Brännback M (2011) Entrepreneurial motivations: What do we still need to
know? Journal of Small Business Management 49(1): 9–26.
Chandler GN, Honig B and Wiklund, J (2005) Antecedents, moderators and performance
consequences of membership change in new venture teams. Journal of Business
Venturing 20(5): 705–725.
Choi JN (2004) Individual and contextual predictors of creative performance: The mediating
role of psychological processes. Creativity Research Journal 16(2–3): 187–199.
Chowdhury S (2005) Demographic diversity for building an effective entrepreneurial team: Is
it important? Journal of Business Venturing 20(6): 727–746.
Cooper AC and Artz KW (1995) Determinants of satisfaction for entrepreneurs. Journal of
Business Venturing 10(6): 439–457.
Crossan MM and Apaydin M (2010) Multi-dimensional framework of organizational
innovation: A systematic review of the literature. Journal of Management Studies
47(6): 1154–1191.
DeCharms R (1968) Personal causation: The internal affective determinants of behavior.
New York: Academic Press.
Deci EL, Koestner R and Ryan RM (1999) A meta-analytic review of experiments examining
the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin 125(6):
627–668.
22
Deci EL and Ryan RM (1985) Intrinsic motivation and self-determination in human
behavior. New York: Plenum Publishing Co.
Deci EL, Ryan RM, Gagné M, Leone DR, Usunov J and Kornazheva BP (2001) Need
satisfaction, motivation, and well-being in the work organizations of a former Eastern
Bloc country. Personality and Social Psychology Bulletin 27(8): 930–942.
De Winne S and Sels L (2010) Interrelationships between human capital, HRM and
innovation in Belgian start-ups aiming at an innovation strategy. International
Journal of Human Resource Management 21(11): 1863–1883.
Eisenberger R, Rhoades L and Cameron J (1999) Does pay for performance increase or
decrease perceived self-determination and intrinsic motivation? Journal of
Personality and Social Psychology 77(5): 1026–1040.
Eisenhardt KM and Schoonhoven CB (1990) Organizational growth: Linking founding team,
strategy, and growth among U.S. semi-conductor ventures, 1978–1988.
Administrative Science Quarterly 35(3): 504–529.
Ensley MD and Hmieleski KM (2005) A comparative study of new venture top management
team composition, dynamics and performance between university-based and
independent startups. Research Policy 34(7): 1091–1105.
Ensley MD and Pearce CL (2001) Shared cognition in top management teams: Implications
for new venture performance. Journal of Organizational Behavior 22(2): 145–160.
Erez M, Gopher D and Arzi N (1990) Effects of goal difficulty, self-set goals, and monetary
rewards on dual task performance. Organizational Behavior and Human Decision
Processes 47(2): 247–269.
European Commission (2012) SBA Fact Sheet 2012 – Belgium.
http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/performance-
review/files/countries-sheets/2012/belgium_en.pdf
Gagné M and Deci EL (2005) Self-determination theory and work motivation. Journal of
Organizational Behavior 26(4): 331–362.
Gagné M, Forest J, Gilbert MH, Aubé C, Morin E and Malorni A (2010) The motivation at
work scale: Validation and evidence in two languages. Educational and
Psychological Measurement 70(4): 628–646.
Gartner WB, Shaver KG, Gatewood E and Katz JA (1994) Finding the entrepreneur in
entrepreneurship. Entrepreneurship: Theory & Practice 18(3): 5–9.
Grolnick WS and Ryan RM (1987) Autonomy in children’s learning: An experimental and
individual difference investigation. Journal of Personality and Social Psychology
52(5): 890–898.
Gu X, Solmon MA, Zhang T and Xiang P (2011) Group cohesion, achievement motivation
and motivational outcomes among female college students. Journal of Applied Sport
Psychology 23(2): 175–188.
Hair JF, Anderson RE, Tatham RL and Black WC (1998) Multivariate data analysis (5th ed).
Upper Saddle River: Prentice Hall.
Harrison DA, Price KH and Bell MP (1998) Time and the effects of surface- and deep-level
diversity on work group cohesion. Academy of Management Journal 41(1): 96–107.
Herron L and Robinson RB Jr (1993) A structural model of the effects of entrepreneurial
characteristics on venture performance. Journal of Business Venturing 8(3): 281–294.
Herzberg F (1966) Work and the nature of man. Cleveland, OH: World.
Hinsz VB, Tindale RS and Vollrath DA (1997) The emerging conceptualization of groups as
information processes. Psychological Bulletin 121(1): 43–64.
Hmieleski KM, Cole MS and Baron RA (2012) Shared authentic leadership and new venture
performance. Journal of Management 38(5): 1476–1499.
23
Horwitz SK and Horwitz IB (2007) The effects of team diversity on team outcomes: A meta-
analytic review of team demography. Journal of Management 33(6): 987–1015.
Jehn KA, Northcraft GB and Neale MA (1999) Why differences make a difference: A field
study of diversity, conflict, and performance in workgroups. Administrative Science
Quarterly 44(4): 741–763.
Kamm JB, Shuman JC, Seeger JA and Nurick AJ (1990) Entrepreneurial teams in new
venture creation: A research agenda. Entrepreneurship: Theory & Practice 14(4): 7–
17.
Kuratko DF, Hornsby JS and Naffziger DW (1997) An examination of owner’s goals in
sustaining entrepreneurship. Journal of Small Business Management 35(1): 24–33.
Langan-Fox J and Roth S (1995) Achievement motivation and female entrepreneurs. Journal
of Occupational & Organizational Psychology 68(3): 209–218.
Liao J, Welsch H and Stoica M (2003) Organizational absorptive capacity and
responsiveness: An empirical investigation of growth-oriented SMEs.
Entrepreneurship: Theory & Practice 28(1): 63–85.
McClelland DC (1961) The achieving society. New York: D. Van Nostrand Company, Inc.
McMullen JS, Bagby DR and Palich LE (2008) Economics freedom and the motivation to
engage in entrepreneurial action. Entrepreneurship: Theory & Practice 32(5): 875–
895.
Maes J and Sels L (forthcoming) SMEs’ radical product innovation: The role of internally
and externally oriented knowledge capabilities. Journal of Small Business
Management.
Maes J, Sels L and Roodhooft F (2005) Modeling the link between management practices
and financial performance. Evidence from small construction companies. Small
Business Economics 25(1): 17–34.
Maslow AH (1954) Motivation and personality. New York: Harper & Row.
Milliken FJ and Martins LL (1996) Searching for common threads: understanding the
multiple effects of diversity in work groups. Academy of Management Review 21(2):
402–433.
Moreno AM and Casillas JC (2008) Entrepreneurial orientation and SMEs: A causal model.
Entrepreneurship: Theory & Practice 32(3): 507–528.
Nemeth CJ and Staw BM (1989) The tradeoffs of social control and innovation within groups
and organizations. In: Berkowitz L (ed) Advances in experimental social psychology,
Vol. 22. New York: Academic Press, pp. 175–210.
Olsen BJ, Parayitam S and Twigg NW (2006) Mediating role of strategic choice between top
management team diversity and firm performance: Upper echelons theory revisited.
Journal of Business and Management 12(2): 111–126.
Podsakoff PM, MacKenzie SB, Lee JY and Podsakoff NP (2003) Common method biases in
behavioral research: A critical review of the literature and recommended remedies.
Journal of Applied Psychology 88(5): 879–903.
Porter LW and Lawler EE III (1968) Managerial attitudes and performance. Homewood, IL:
Irwin-Dorsey.
Prabhu V, Sutton C and Sauser W (2006) Creativity and certain personality traits:
Understanding the mediating effect of intrinsic motivation. Creativity Research
Journal 20(1): 53–66.
Ramanujam V and Varadarajan P (1989) Research on corporate diversification: A synthesis.
Strategic Management Journal 10(6): 523–552.
Ryan RM and Deci EL (2000) Self-determination theory and the facilitation of intrinsic
motivation, social development, and well-being. American Psychologist 55(1): 68–78.
24
Segal G, Borgia D and Schoenfeld J (2005) The motivation to become an entrepreneur.
International Journal of Entrepreneurial Behaviour & Research 11(1): 42–57.
Shane S and Cable D (2002) Network ties, reputation, and the financing of new ventures.
Management Science 48(3): 364–381.
Shepherd DA and DeTienne DR (2005) Prior knowledge, potential financial reward, and
opportunity identification. Entrepreneurship: Theory & Practice 29(1): 91–112.
Smith KG, Collins CG and Clark KD (2005) Existing knowledge, knowledge creation
capability, and the rate of new product introduction in new high-technology firms.
Academy of Management Journal 48(2): 346–357.
Steffens P, Terjesen S and Davidsson P (forthcoming). Birds of a feather get lost together:
New venture team composition and performance. Small Business Economics DOI:
10.1007/s11187-011-9358-z.
Talaulicar T, Grundei J and von Werder A (2005) Strategic decision-making in start-ups, the
effect of top management team organization and processes on speed and
comprehensiveness. Journal of Business Venturing 20(4): 519–541.
Ucbasaran D, Westhead P and Wright M (2008) Opportunity identification and pursuit: Does
an entrepreneur’s human capital matter? Small Business Economics 30(2): 153–173.
Ucbasaran D, Lockett A, Wright M and Westhead P (2003) Entrepreneurial founder teams:
Factors associated with member entry and exit. Entrepreneurship: Theory & Practice
28(2): 107–128.
Van Knippenberg D, De Dreu CKW and Homan AC (2004) Work group diversity and group
performance: An integrative model and research agenda. Journal of Applied
Psychology 89(6): 1008–1022.
Vansteenkiste M, Soenens B, Sierens E, Luyckx K and Lens W (2009) Motivational profiles
from a self-determination perspective: The quality of motivation matters. Journal of
Educational Psychology 101(3): 671–688.
Venkatraman N and Ramanujam V (1986) Measurement of business performance in strategy
research: A comparison of approaches. The Academy of Management Review 11(4):
801–814.
Vroom VH (1964) Work and motivation. New York: Wiley.
Watson WE, Kumar K and Michaelsen LK (1993) Cultural diversity’s impact on interaction
process and performance: Comparing homogenous and diverse task groups. Academy
of Management Journal 36(3): 590–602.
West CT and Schwenk CR (1996) Top management team strategic consensus, demographic
homogeneity and firm performance: A report of resounding nonfindings. Strategic
Management Journal 17(7): 571–576.
Wigfield A, Guthrie JT, Tonks S and Perencevich KC (2004) Children’s motivation for
reading: Domain specificity and instructional influences. Journal of Educational
Research 97(6): 299–309.
Williams KY and O’Reilly CA (1998) Demography and diversity in organizations: A review
of 40 years of research. Research in Organizational Behavior 20 (1998): 77–140.
Wood R and Bandura A (1989) Social cognitive theory of organizational management.
Academy of Management Review 14(3): 361–384.
Zahra SA (1993) Environment, corporate entrepreneurship, and financial performance: A
taxonomic approach. Journal of Business Venturing 8(4): 319–340.
Zahra SA, Ireland RD and Hitt MA (2000) International expansion by new venture firms:
International diversity, model of market entry, technological learning, and
performance. Academy of Management Journal 43(5): 925–950.
25
Appendix
Appendix Factor Loadings and Cronbach’s Alphas.
Item
External
regulation
Introjected
regulation
Controlled
motivation
Identified
regulation
Intrinsic
motivation
Autonomous
motivation
Environmental
turbulence
I put in effort because this allows me to make more money. .869 .237 .513 .229 .004 .201 /
I put in effort to obtain more work security. .830 .346 .578 .398 .138 .355 /
I put in effort because I would lose financial rewards otherwise. .606 .633 .741 .089 -.314 -.135 /
I put in effort because I would feel ashamed otherwise. .309 .879 .780 .074 -.110 -.054 /
I put in effort to avoid disappointment from others. .364 .845 .784 .203 -.067 .040 /
I put in effort so that I would not have to feel guilty. .280 .837 .741 .128 -.082 -.012 /
I put in effort because this entrepreneurial work matches with my personal values. .272 .170 .142 .348 .771 .707 /
I put in effort because I find my entrepreneurial work very significant. .227 .075 .154 .856 .559 .778 /
I put in effort because my entrepreneurial work allows me to reach my life goals. .351 .220 .345 .911 .354 .668 /
I put in effort because I have fun doing this type of work. -.188 -.346 -.403 .358 .802 .702 /
I put in effort because I enjoy this work very much. .110 -.033 -.017 .620 .695 .761 /
I put in effort because I find entrepreneurial work extremely interesting. -.090 -.148 -.218 .451 .843 .772 /
Within our industry the possibilities for technological innovations are considerable. / / / / / / .672
Within our industry a lot of opportunities for new products and/or new services exist. / / / / / / .766
Within our industry customer demand for new products and/or new services is increasing. / / / / / / .814
Within our industry the need for a new technology is growing. / / / / / / .852
Within our industry the market for new products and/or new services is expanding. / / / / / / .714
Our industry requires technological innovations in order to continue to grow. / / / / / / .759
N 142 142 142 142 142 142 66
Cronbach’s alpha .692 .851 .791 .729 .764 .827 .855
Notes: Extraction Method: Principal Components Analysis; Promax rotation; To compute both factors we made use of the following formula: F = ((S - V) / ((V x W) - V)) x
100 with S equal to the sum of all initial values (before transformation), V referring to the number of variables and W representing the number of scale points (Maes et al.,
2005); The use of two distinct methods to collect the data (questionnaires and interviews) minimizes possible common-method variance effects. Open-ended questions were
interspersed with other types of questions, which prevented respondents from adopting a scale-based pattern linked to Likert or semantic differential scales (Podsakoff et al.,
2003); Harman’s single factor test was used to examine concerns of possible common-method variance (interview information). Multiple factors with eigenvalues greater
than one emerged. The first factor only explained 30.4% of the variance; Construct validity was established by developing measures from well-grounded theory (Barringer
and Bluedorn, 1999); Confirmatory factor analysis was performed to assess the convergent and discriminant validity. The fit of the unconstrained two-factor model (including
both constructs in a way that each item loaded solely on the factor for which it was an intended indicator) was reasonably good (e.g., GFI = .81) and better than the fit of the
four-factor model (convergent validity). The pair-wise difference between the chi-squared value of the unconstrained model and that of the constrained model largely
exceeded 3.84 (5% critical value) (discriminant validity) (Anderson and Gerbing, 1988).
doc_377398424.pdf
On this brief description concerning smells like team spirit how entrepreneurial founding team motivations affect new venture.
STEUNPUNT ONDERNEMEN EN REGIONALE ECONOMIE
NAAMSESTRAAT 61 – BUS 3550
BE-3000 LEUVEN
TEL +32 16 32 66 61 | FAX +32 16 37 35 11
[email protected]
www.steunpuntore.be
Beleidsrapport STORE-B-14-011
Smells like team spirit? How entrepreneurial
founding team motivations affect new venture
financial and innovation performance
J onas Debrulle
1,a
en J ohan Maes
a
a
IESEG School of Management
23 december 2014
1
De resultaten in dit rapport geven de mening van de auteurs weer en niet deze van de Vlaamse overheid: de
Vlaamse Gemeenschap/het Vlaams Gewest is niet aansprakelijk voor het gebruik dat kan worden gemaakt van de
in deze mededeling of bekendmaking opgenomen gegevens.
1
Smells like team spirit? How entrepreneurial founding team
motivations affect new venture financial and
innovation performance
Jonas Debrulle
Johan Maes
Abstract
This study investigates the relationship between entrepreneurial founding teams’ (EFTs)
entrepreneurship motivations and new venture financial and operational performance,
measured as ‘return on assets’ and ‘firm innovation’. We consecutively introduce the level
and heterogeneity of EFTs’ autonomous and controlled entrepreneurship motivation as
potential drivers of new venture performance. Our analyses are based on a sample of 66
teams representing 142 founders. We observe that EFTs’ level of autonomous and controlled
entrepreneurship motivation contribute to new venture return on assets. We also find support
for the negative impact of EFTs’ controlled entrepreneurship motivation on new venture
innovation. Introduction of the moderating motivation heterogeneity variables uncovers
complex interactions between EFTs’ levels of entrepreneurship motivation, team member
motivation heterogeneity, and this study’s dependents. Our findings reveal that motivation
heterogeneity acts as a positive moderator of the relationship between EFTs’ level of
autonomous/controlled entrepreneurship motivation and new venture innovation, while it
emerges as a negative moderator of the relationship between EFTs’ controlled motivation and
new venture financial performance. Implications for further research are highlighted.
Introduction
New venture creation and development rarely involves a “lonely hero” exploiting a lucrative
opportunity. Rather, evidence supports the idea that entrepreneurship commonly represents a
collective activity, whereby teams of entrepreneurs are responsible for creating and managing
new ventures (McMullen et al., 2008). Ever since Gartner et al. (1994) emphasized that
organizations are in fact social entities, and that the “entrepreneur in entrepreneurship” is
more likely to be plural than singular, an increasing number of researchers have shifted their
attention to the entrepreneurial team phenomenon. Building on prior contributions, Kamm et
al. (1990) defined an entrepreneurial founding team as “two or more individuals who jointly
establish a business in which they have an equity interest” (p. 7). Ucbasaran et al. (2003)
extended this definition to include individuals who “have a key role in the strategic decision-
making of the venture at the time of founding” (p. 109). This study focuses on entrepreneurial
founding teams (EFTs) of which all members meet these conditions and continue to impact
the strategic development of the new venture.
So far, two types of team heterogeneity have dominated research on EFTs. On the one
hand, prior work has emphasized demographic differences between entrepreneurs. These
differences are conceptualized as “surface-level” diversity or heterogeneity (Harrison et al.,
1998). Particular interest has been directed to age and/or gender diversity between business
founders (e.g., Chandler et al., 2005; Chowdhury, 2005). On the other hand, task- or skill-
related heterogeneity (Milliken and Martins, 1996) has equally emerged as a chief topic.
Though a regularly debated topic, extant work suggests that diversity of skills within EFTs
increases the pool of cognitive resources, which enables the team to better handle the
2
complexities inherent to new venture emergence and development, thus stimulating new
venture performance. Widely held indicators of task and skill heterogeneity include
differences in educational specialization, functional expertise, and industry experience (e.g.,
Chandler et al., 2005).
In sharp contrast to these demographic diversity and skill differences, for which a link
between team heterogeneity and firm performance has been established in the context of
EFTs and top management teams (TMTs) in more established organizations, the issue of
more psychological characteristics has received far less research attention. Recent
contributions have, nevertheless, argued that such characteristics are just as likely to affect
team effectiveness and organizational outcomes (Chowdhury, 2005; Van Knippenberg et al.,
2004). In an empirical study of new venture TMT composition, Ensley and Pearce (2001) and
Ensley and Hmieleski (2005) shed light on the implications of team cohesion for new venture
performance. They found that contrary to affective conflicts, cognitive conflicts between
entrepreneurial team members contribute directly to new venture success. Likewise, when
evaluating the theoretical basis for demographic diversity in EFTs, Chowdhury (2005)
established that entrepreneurial team effectiveness builds on team member commitment and
comprehensiveness in strategic decision-making. This study extends the field of deep-level
(directly unobservable) team diversity by exploring the effects of individual motivation
heterogeneity between entrepreneurial team members on new venture performance. All in all,
however, motivation remains a highly neglected variable in team heterogeneity research (Van
Knippenberg et al., 2004).
According to Self-Determination Theory (SDT) (Gagné and Deci, 2005), people can
adhere to a task with eagerness and volition because they genuinely enjoy doing it, or because
the task at hand feels personally important (autonomous motivation). In contrast, they can
engage in an activity with a sense of pressure, because they wish to attain certain rewards that
are dependent upon task completion, or because they strive for outside approval (controlled
motivation). Hence, in an entrepreneurial setting, ‘autonomous entrepreneurship motivation’
causes individuals to engage in new firm formation and development simply because they
find this activity enjoyable and derive satisfaction from it, or because they are deeply
committed to becoming an entrepreneur and wish to maintain their business. ‘Controlled
entrepreneurship motivation’, on the contrary, provokes individuals to create and develop
new ventures because of the anticipated rewards (including social rewards), or because they
appreciate the status associated with being an entrepreneur while evading the stigma of
business failure. Though autonomous and controlled motivation both elicit and sustain
purposeful behavior, they are governed by distinctive mechanisms (Gagné and Deci, 2005)
and are aimed at different objectives (Deci et al., 1999) that arise from a particular ‘locus of
causality’ (DeCharms, 1968). Despite these inconsistencies, in the research specific to
entrepreneurial team heterogeneity little is known about the effect of motivational team
composition on new venture performance.
This study aims to make the following two contributions to existing literature. First, it
encompasses a theoretical perspective. It draws on motivation theory, specifically SDT
(Gagné and Deci, 2005), to introduce autonomous and controlled motivation as EFT
psychological driving forces of new venture performance. By exploring motivational
differences in the context of EFTs, we wish to corroborate the idea that factors other than
demographic diversity and skill differences affect team effectiveness and organizational
outcomes (Chowdhury, 2005). This research gap not only concerns research on private firms
owned by EFTs, but also that addressing TMTs in general. While we acknowledge that the
relationships found in this study, due to differences in individual risk bearing, ownership and
decision-making power (Ucbasaran et al., 2003), may not necessarily hold for TMTs in
established firms, they could inspire researchers and policy-makers to broaden their view of
3
team-based venture determinants to include diversity in psychological and personality
characteristics. A second contribution of the current study encompasses an empirical
perspective. How to define and measure new venture performance has long been debated.
Both financial and operational conceptualizations have been advocated. It is therefore
appropriate to be skeptical about how to best capture new venture performance. We address
this issue by investigating two performance outcomes: one grounded in audited financial
information, and another in operational data. From a financial standpoint, we define new
venture performance in terms of the company’s profitability relative to its total assets (return
on assets), whereas from an operational standpoint we focus on new ventures’ innovation
performance.
This article proceeds as follows: First, we capture the notion of autonomous and
controlled entrepreneurship motivation and discuss its anticipated model impacts. We
identify gaps in existing research and formulate five hypotheses to be tested. Next, we
describe our research methodology, with special emphasis on our sampling procedures,
measures and measurement validity tests. Hypotheses developed in this study were tested
using data from 142 individual entrepreneurs representing 66 EFTs. Finally, we present and
discuss our findings, after which we conclude with some caveats and opportunities for future
research.
Theoretical background and hypotheses
Motivation is the internal disposition that initiates, reinforces and maintains goal-oriented
behaviors. It is the psychological drive that encourages individuals to take action, whether it
is to eat when we feel hungry, or to enroll in a university to obtain a degree. A considerable
part of motivation research interest has been aimed at understanding work-related behavior.
As classics in organizational behavior literature, the theories of Maslow (1954), Herzberg
(1966), Alderfer (1972), and McClelland (1961) are generally referred to as ‘content theories
of motivation’ as they attempt to explain ‘what’ motivates behavior, yet fail to shed light on
‘how’ motivation occurs. ‘Process theories of motivation’, on the other hand, do not primarily
deal with the energizers of motivation (i.e., needs-based approach), but pay attention to ‘how’
individuals direct work-related behavior (e.g., Adams, 1963; Porter and Lawler, 1968;
Vroom, 1964). Since the objective of the current study is to examine how entrepreneurial
behavior is rationalized and maintained in the context of EFTs, rather than to explore internal
energizers of entrepreneurial action, we build on motivation process theories to develop our
hypotheses. More specifically, this study is rooted in the expectancy approach to motivation.
According to Vroom (1964), individuals’ work motivation is the product of three
catalysts: putting forth effort will lead to performance; performance will trigger rewards; and
these rewards are desirable. Similar to Vroom (1964), Porter and Lawler (1968) stipulate that
task completion is dependent on its anticipated rewards. Porter and Lawler (1968), however,
extended Vroom’s (1964) theory by categorizing rewards as intrinsic or extrinsic, thereby
proposing a model of intrinsic and extrinsic work motivation. ‘Intrinsic motivation’ involves
individuals engaging in a task because of the spontaneous satisfaction and sense of
achievement that arise upon successful task completion. ‘Extrinsic motivation’, in contrast,
emanates from outside the individual. In this case, motivation and satisfaction do not
originate from the task itself, but from the rewards to which the task is instrumental. Though
commonly accepted as an adequate theory for explaining task motivation, Porter and
Lawler’s (1968) Expectancy Theory is not without controversy. Especially its implicit
additivity assumption of intrinsic and extrinsic motivation, thus yielding total job satisfaction,
has been subject to widespread debate (Deci et al., 1999). Consequently, when positing their
SDT, Deci and Ryan (1985) no longer acknowledged motivation as a unitary concept.
4
Instead, they emphasized the relative strength of ‘autonomous motivation’ versus ‘controlled
motivation’, rather than sustaining the concept of ‘total motivation’.
Capitalizing on Porter and Lawler’s (1968) dichotomy between intrinsic and extrinsic
motivation, Deci and Ryan (1985) differentiate between ‘autonomous motivation’ and
‘controlled motivation’. Autonomous motivation involves doing a task because it is
inherently fun (subscale: ‘intrinsic motivation’) or because task accomplishment is
considered personally important (subscale: ‘identified regulation’). The latter designates an
internalized form of extrinsic motivation, whereby individuals have come to personally value
work-related behaviors and understand their importance for their own well-being (Ryan and
Deci, 2000). Nonetheless, identified regulation is still considered a type of extrinsic
motivation as it does not involve personal interest in the task at hand, yet requires an
instrumentality between the task and individually appreciated goals (Gagné and Deci, 2005).
Contrary to autonomous motivation, controlled motivation encompasses actions unilaterally
initiated and maintained by external contingencies (subscale: ‘external regulation’ or
‘extrinsic motivation’) as well as behaviors enacted with a feeling of pressure to avoid self-
inflicted punishment, such as shame or guilt (subscale: ‘introjected regulation’) (Deci and
Ryan, 1985). Though equally internalized, introjected regulation, as opposed to identified
regulation, skews more strongly towards pure extrinsic motivation. This type of motivation is
not perceived as congruent with personal objectives, which causes it to be considered less
valuable and important (Ryan and Deci, 2000). Hence, with controlled motivation, the cause
of the behavior is said to have an external ‘perceived locus of control’ (DeCharms, 1968).
Conversely, autonomous motivation originates from an internal ‘perceived locus of control’,
so that individuals feel (relatively) autonomous while performing the activities (DeCharms,
1968).
While several studies have supported the autonomous-controlled dichotomy as an
adequate approach to study work motivation, few have tested the theory within an
entrepreneurial setting. Instead, research on entrepreneurship motivation has largely
concentrated on push-pull factors as predictors of firm activities. Distinguishing between
autonomous and controlled motivation is, however, particularly insightful to entrepreneurship
literature. Several arguments support this view. First, the autonomous-controlled dichotomy
incorporates what is already known in terms of the push-pull strand in entrepreneurship
research (Segal et al., 2005). For instance, autonomous motivation implies a sense of volition
(pull), whereas controlled motivation implies a sense of pressure (push) (Gagné and Deci,
2005). As a result, the motives sustaining entrepreneurial behavior as recognized by the push-
pull perspective correspond to those accommodated by the controlled-autonomous
perspective. Second, contrary to push-pull motives, autonomous and controlled motivation
transcend pre-launch or launch activities of entrepreneurship. They thus can be employed to
ascertain post-entry alterations in motivations of practicing entrepreneurs. Third, the
autonomous-controlled distinction can be applied to a specific (individual or team-based)
activity, a project, or an entire profession. The push-pull debate, on the other hand, only
produces a classification at person-level, thereby disregarding any task-level motivational
differences. Finally, while autonomous and controlled motivation trigger and sustain
purposeful behavior in a diverse way, they are not considered mutually exclusive (Cameron,
2001). Ryan and Deci (2000) posit that individual actions may be simultaneously motivated
by a combination of autonomous and controlled factors operating in a parallel fashion.
Accordingly, autonomous motivation does not exclude an individual from seeking rewards,
and controlled motivation does not prohibit task enjoyment. This does, however, imply that
external contingencies (in case of autonomous motivation) or task interestingness (in the case
of controlled motivation) might be insufficient to maintain motivation.
5
When making predictions, much of the work in motivation literature unilaterally
considers the amount of ‘total motivation’ a person has for a task, thereby ignoring
subdimensional types of motivation. Gagné and Deci (2005) warn against treating motivation
as a unitary concept. They argue that determining motivation by various factors is of little use
if it is then operationalized into a single variable. Not only does this induce the reification of
the motivation construct, it also fails to capture any underlying variations between types of
motivation. That is why, consistent with Gagné and Deci (2005), the current study
differentiates between EFTs’ autonomous and controlled entrepreneurship motivation.
We know from the field of work motivation and regulated behavior that autonomous
and controlled motives aim at achieving separate goals (Deci and Ryan, 1985) that arise from
a different ‘perceived locus of causality’ (DeCharms, 1968). Though autonomous motivation
effects have emerged as the most stable, both types of motivation are said to elicit and sustain
purposeful behavior (Deci and Ryan, 2000). Within the entrepreneurship field, Herron and
Robinson (1993) identified entrepreneurial motivation as a key factor influencing venture
performance. Specifically, accumulating wealth, monetary compensation and building equity
in the firm (i.e., external contingencies) have long been recognized as important propellers of
entrepreneurial behavior (Langan-Fox and Roth, 1995; Shepherd and DeTienne, 2005).
According to Campbell (1992), individuals initiate, maintain and develop businesses if their
expected present value of entrepreneurship exceeds that of being an employee. Within SDT,
we know that when behavior is thus motivated, it suggests the financial performance of the
venture becomes instrumental to attain personally desired outcomes (e.g., wealth
accumulation, outside approval) or to avoid undesired ones (e.g., business failure, shame,
guilt). Therefore, we believe that with higher levels of controlled motivation, entrepreneurs,
either alone or as part of a team, will be increasingly energized into actions that foster new
ventures’ financial performance.
Yet, not all entrepreneurial behavior is coerced or seduced by (introjected) external
objectives. Kuratko et al. (1997: p.31) established that goals “of both an intrinsic and
extrinsic nature” are vital for sustaining entrepreneurship. Similarly, in their meta-analysis,
Carsrud and Brännback (2011) argue that while most researchers assume entrepreneurship to
be a pursuit of instrumental economic goals, substantial evidence exists of people engaging in
entrepreneurship without any apparent (dominant) reward other than task enjoyment. There
exists, however, relatively little research that has explored how autonomous entrepreneurship
motivation impacts new venture performance. Turning to educational psychology literature,
we learn that when activities are experienced as spontaneously satisfying and/or personally
important, people tend to persistently exert effort, be eager to acquire additional knowledge,
and show improved creativity (Deci et al., 2001; Wigfield et al., 2004). Since autonomous
motivated behaviors satisfy SDT’s basic needs of competence, autonomy and relatedness
(Deci and Ryan, 1985), engaging in such activities facilitates positive outcomes such as task
engagement, work performance, psychological well-being and behavioral persistence (Baard
et al., 2004). Extrapolating this research to an entrepreneurship context, we posit that
entrepreneurs with advanced levels of autonomous motivation will exert more time and effort
on business development, display more creativity and task engagement, and achieve higher
functional effectiveness, which ultimately should reflect in new ventures’ financial
performance. In sum, we hypothesize that while some entrepreneurs in EFTs may be (mainly)
driven by external contingencies (controlled motivation) and others by task enjoyment and
adopted job-attributes (autonomous motivation), they all share a commitment to pursue the
business opportunity, develop the new venture and sustain firm ownership. Hence, we believe
that both types of motivation will trigger EFTs to promote new ventures’ financial
performance.
6
Hypothesis 1: EFTs’ level of entrepreneurship motivation is positively associated with
new venture financial performance. This will be reflected by the positive effects of:
a) EFT autonomous entrepreneurship motivation, and
b) EFT controlled entrepreneurship motivation.
Despite the analogous anticipated association of EFTs’ autonomous and controlled
motivation with new venture financial performance, we know from lab experiments and
research in other domains that both types of motivation may differ in terms of their
magnitude and intensity. Gagné and Deci (2005) argue that work climates that encourage
intrinsic motivation or a profound internalization of extrinsic motivation yield improved
performance compared to climates that are control-oriented. Particularly for activities
demanding disciplined engagement, intellectual flexibility and complex problem solving
(e.g., activities related to new venture development), autonomous motivation has turned out
to be a better predictor of effective performance (Baard et al., 2004). Individuals appear to be
most creative when they find rewards in the task itself rather than when the task functions as
a means to an end (Amabile et al., 1990). In contrast, with regards to mundane tasks,
controlled motivation has been found to better facilitate short-term performance (Grolnick
and Ryan, 1987). Autonomous motivation produces better quality responses when
confronting multifaceted situations with high ambiguity. In such situations, which are
inherent to entrepreneurship, autonomy-oriented individuals appear to manage complex
problems better than their control-oriented colleagues (Erez et al., 1990). Not only do they
share a greater awareness of the environment, they also tackle unexpected events more
successfully. What is more, due to higher levels of excitement they are willing to dedicate
more personal resources to task fulfillment (e.g., effort and attention).
Building on the above arguments, we assume that EFTs’ autonomous
entrepreneurship motivation will exert a stronger influence on new venture financial
performance compared to their controlled entrepreneurship motivation. Thus, we
hypothesize:
Hypothesis 2: EFTs’ level of autonomous entrepreneurship motivation has a stronger
effect on new venture financial performance than EFTs’ level of controlled
entrepreneurship motivation.
Next to financial performance we also focus on new venture operational performance in the
form of innovation (bringing new goods and services to the market). Innovation is important
for firms, including new ventures (De Winne and Sels, 2010). It fuels organizations’
competitive advantage and stimulates growth and survival. Innovation is, therefore, to some
degree a goal for many new ventures. Yet, it also involves insecurity and risks (Smith et al.,
2005) and requires (strategic) perseverance. By looking at innovation as a goal, it is tied to
entrepreneurship motivation (Baum and Locke, 2004). More specifically, controlled
entrepreneurship motivation is expected to be negatively linked to new venture innovation.
After all, innovation is a goal rather distant in time, and such goals usually do not generate
the beneficial (control-oriented) motivational effects of short-term goals (Wood and Bandura,
1989). Past research, however, suggests that even controlled motivation could have a positive
effect on creativity and innovation (Eisenberger et al., 1999). More recent research has
refined that idea and made this effect contingent upon the rewards attached to the innovation
and/or other circumstantial conditions (Choi, 2004; Prabhu et al., 2006). In view of the
insecurity and risks attached to new venture innovation we expect that the possible rewards
attached to it are not strong enough to spur controlled motivation. Instead, stirred by the
prospect of accumulating wealth, control-motivated EFTs might shorten their time
7
perspectives and lucratively consume available means of production. Although such a modus
operandi does not particularly favor firm persistence, it may cause more control-oriented
EFTs to financially outperform primarily autonomous-oriented teams in the short term. The
level of controlled motivation, thus, is expected to reflect the EFTs’ averseness to any actions
that might endanger contingent monetary rewards (Deci et al., 2001; Ryan and Deci, 2000).
Autonomous motivation, on the other hand, should trigger new venture innovation.
After all, entrepreneurs can make up their own agenda (Baum and Locke, 2004). If they
consider innovation as something of genuine value, and anticipate that their efforts will lead
to innovation performance, then autonomous motivation will play a facilitating role in their
behavioral scheme. Aspects of autonomous motivation have been shown to correlate strongly
to expectancy beliefs (Gu et al., 2011). This line of reasoning is not entirely new; past
research has identified intrinsic or autonomous motivation elements as key ingredients to
individual innovation and creativity (Prabhu et al., 2006). Aimed at new venture continuity
and long-term fruition, rather than short-term profitability, autonomous motivation is more
likely to promote strategic investments, calculated risk-taking and the development of a
sound business foundation in order to achieve more challenging self-set goals (Watson et al.,
1993). While such actions foster long-term organizational responsiveness, they are
detrimental to new ventures’ short-term financial results. As such, we formulate the following
hypothesis:
Hypothesis 3: EFTs’ level of entrepreneurship motivation is associated with new
venture innovation performance. This will be reflected by:
a) The negative effect of EFT controlled entrepreneurship motivation, and
b) The positive effect of EFT autonomous entrepreneurship motivation.
Given that distinctive mechanisms govern autonomous and controlled motivation (Gagné and
Deci, 2005), and that a mixture of both types of motives is bound to occur within EFTs, we
anticipate that team motivation heterogeneity can affect the earlier hypothesized relationships
between autonomous/controlled entrepreneurship motivation and new venture
financial/innovative performance. Williams and O’Reilly (1998) distinguish two dominant
perspectives in the research on performance effects of team diversity: the ‘social
categorization’ perspective and the ‘information/decision-making’ perspective. The rationale
behind the social categorization perspective is that people use perceived similarities and
differences to categorize themselves and others into ‘in’ and ‘out’ social groups. As the
current study focuses on small EFTs, of which all members own and manage part of the new
venture, thinking along the lines of such categorizations is not very useful. The second view,
that is the information/decision-making perspective, is, however, more promising. This
tradition focuses on the task-related aspects of team processes. Its research interest lies at
team members’ exchange, discussion and integration of ideas, knowledge and insights
relevant to the tasks at hand (Van Knippenberg et al., 2004). Advancing idea and knowledge
diversity within teams as an informational resource, the information/decision-making
perspective argues that more diverse groups may outclass more homogeneous ones.
Contrary to the idea of social groups, it seems reasonable to assume that EFTs’
activities involve strong information/decision-making components. This setting, thus, creates
a rich soil for team diversity or heterogeneity to bear fruits. In the current study, team
motivation heterogeneity refers to differences among EFT members on deeper-level
controlled (e.g., differences in terms of external pressures and contingencies driving
behavior) and/or autonomous motives (e.g., differences in personal values, beliefs, opinions
and tasks regarded as ‘fun’ to do). Though some authors, while building on social identity
theory and the similarity-attraction proposition, assume that EFT members are similar to one
8
another (e.g., Cooper and Artz, 1995; Hmieleski et al., 2012), we maintain that they are, in
fact, less (motivational) homogeneous (and, therefore, more heterogeneous) than one would
initially assume. After all, motivational differences concern deep-level topics. As a result, it
may take a considerable amount of time and interaction for entrepreneurship motivational
diversity to emerge.
Key to yielding benefits from team motivation heterogeneity is not the mere presence
of diverse points of view, but the adequate use of the rich information embedded in these
differences (Van Knippenberg et al., 2004). While it is often proposed that conflicts resulting
from heterogeneity generate performance benefits (e.g., Jehn et al., 1999), we posit that it is
the creative processing of members’ diverse viewpoints and information that is essentially
beneficial (Van Knippenberg et al., 2004). In fact, to date, research findings on the conflict
perspective of team heterogeneity are inconclusive (Horwitz and Horwitz, 2007). So, if we
are to understand the influence of EFTs’ heterogeneity on performance, we need to perceive
EFTs as information processing entities (Hinsz et al., 1997).
Capitalizing on the information/decision-making perspective of team diversity, we
contend that EFTs will process information differently depending on (members’ motivation
for) the task at hand and its targeted outcome. In general terms, information processing
encompasses the learning and exchange of perspectives among group members, the
individual-level processing of this information, the feeding back of processing results to the
group, and the discussion and integration of member feedback (Van Knippenberg et al.,
2004). Though these information-processing requirements are highly demanding, they have
been argued to be conducive in the context of complex, cognitive and risky innovations.
Whenever learning processes resemble the information processing of the team, as is the case
with new venture innovations, diverse standpoints, orientations and beliefs are essential to
reach a greater understanding of strategic alternatives (Amason, 1996). In the current study,
we believe that the value, idea and knowledge diversity embedded in the autonomous and
controlled entrepreneurship motives of EFT members will challenge the status-quo, spur team
creativity in solving problems, and ultimately encourage new venture innovation. Hence, we
hypothesize:
Hypothesis 4: Motivation heterogeneity acts as a positive moderator of the
relationship between EFTs’ level of autonomous/controlled entrepreneurship
motivation and new venture innovation. This will be reflected by the positive
moderating effects of:
a) EFT autonomous entrepreneurship motivation heterogeneity, and
b) EFT controlled entrepreneurship motivation heterogeneity.
Given the requirements listed above, it is not surprising that team information processing is to
some extent an ambiguous and often time-consuming process (Nemeth and Staw, 1989). In
fact, these requirements may prevent a team from engaging in speedy decision-making and
achieving swift compromises. Hence, differences in individually valued goals, beliefs and
cognitive schemas that stem from motivation heterogeneity may disturb the setting for short-
term oriented tasks or objectives that capitalize on such efficient decision-making (e.g., short-
term financial performance). What is more, in order to secure team effectiveness,
heterogeneously motivated EFTs will have to devote some resources (e.g., time and effort) to
the reinstatement of group coherence and team consent. Equally motivated team members, in
contrast, because of their mutual beliefs, values and attributes, might find it much easier to
develop and sustain short-term communication patterns, group objectives and modi operandi.
While such communality or homogeneity among EFT members does not particularly favor
new venture innovation or long-term fruition, we believe it may enable less heterogeneously
9
motivated EFTs to facilitate new venture financial performance, at least in the short term
(with the latter being related to this study’s dependent). We, therefore, hypothesize:
Hypothesis 5: Motivation heterogeneity acts as a negative moderator of the
relationship between EFTs’ level of autonomous/controlled entrepreneurship
motivation and new venture financial performance. This will be reflected by the
negative moderating effects of:
a) EFT autonomous entrepreneurship motivation heterogeneity, and
b) EFT controlled entrepreneurship motivation heterogeneity.
Methodology
Sampling procedures
The sample for this research originates from START 2009, an extensive cross-sectional
survey on new ventures located in Flanders, Belgium. This is a biennial population survey of
Flemish incorporated companies that have been in business for one to three years, are active
within various economic sectors and, in 2009, had a minimum of one and a maximum of 49
employees. The primary source of data involved a structured interview with each of the new
ventures’ managers. Face-to-face interviewing allowed for the collection of comprehensive
information on their educational background, career trajectory and entrepreneurship
motivations. General information on the new ventures was collected using a questionnaire
that was mailed prior to the interviews. Finally, financial performance information was
captured using audited information from Bel-first, which denotes a financial database holding
information on the company accounts of all firms incorporated under Belgian law.
The total research population of new ventures consisted of 3183 firms in 2009. Due to
obsolete company data, 259 new ventures could not be reached. Out of the 2924
questionnaires mailed, 453 usable company responses (response rate of 15.5%) and 490
owner interviews were obtained. Within 42% of the responding ventures (190 companies),
daily management was shared by at least two individuals. To be included as an observation in
this study, data on both the venture and its founders were required. In order to be considered
a member of an EFT, respondents had to have an active hand in the founding of the venture,
own an equity stake of at least 10% and assume a key role in the venture’s current strategic
decision-making. Interview responses were required from all members of the EFT. Due to a
lack of data on the new venture and/or on one of its founders, 97 companies had to be
excluded from further analysis. Another 10 businesses were omitted because their current
managers did not meet the above criteria. The remaining 83 companies, representing 176
founders, were further reduced because of the use of listwise exclusion during statistical
procedures. This resulted in a final sample of 66 ventures, representing 142 founders. Teams
ranged in size from two to four members.
Tests between respondent and non-respondent ventures revealed no significant
differences regarding organization age and size. Using chi-square differences and t-tests, no
differences emerged regarding industry, size and organization age between the firms used in
the analyses and those that were eligible yet excluded because of missing values. Similarly,
no differences were detected pertaining to average age, industry experience and
entrepreneurship motivation of sampled EFTs and teams whose members did not all meet the
selection criteria. Finally, a means difference test showed no evidence of financial or
operational performance variations. While this evidence does not eliminate the concern of
possible non-response bias, it does indicate a certain level of representativeness of our
sample.
10
While sample size may be of concern in this study, we would like to point out that
similar sample sizes have been characterizing much of the team-based literature. For
example, Olsen et al. (2006) analyzed 66 teams in their study on the mediating role of
strategic choice between team heterogeneity and firm performance, as did Eisenhardt and
Schoonhoven (1990) when investigating the association of environment, technical innovation
and top management team characteristics on organization sales. West and Schwenk (1996)
studied 65 firms in search of moderating effects of environmental turbulence on the
relationship between top management team consensus and firm performance. Finally,
Talaulicar et al. (2005) called upon 56 teams to examine the influence of new venture team
organization and processes on the comprehensiveness and speed of strategic decision-
making.
Measures
New venture financial performance. Following Zahra et al. (2000), we addressed three issues
in measuring new venture performance. First, we decided to make use of a lagged
performance measure by averaging financial performance data over the company’s last two
years. Given this study’s focus on new ventures, a two-year time period should capture
unusual events in the market, while avoiding the introduction of noise (e.g., changing
industry structures and firm strategy variations) (Ramanujam and Varadarajan, 1989).
Second, we acquired audited financial information from a separate database more than one
year after the owner interviews and company questionnaires were concluded. To lessen the
possible effects of common method variance we decided not to make use of survey method
data. Instead, using the company’s unique identification number, we retrieved the annual
accounts of each company from the aforementioned Bel-first database. The final issue
concerned the selection of an adequate financial performance criterion. We adopted “return
on assets” (ROA) as an objective accounting measure of new venture performance. As an
indicator of the profitability of a company relative to its total assets, the ROA sheds light on
the efficiency of EFTs at using company assets to generate earnings. As a result, ROA
accurately reflects entrepreneurial effectiveness and the ability of EFTs to create wealth.
New venture innovation performance. Traditional conceptions of business
performance largely center on the use of financial accounting measures. These measures
retrospectively address ventures’ fulfilment of specific economic goals. They, unfortunately,
do not shed light on firms’ internal effectiveness nor do they reveal any mechanisms that
drive performance (Venkatraman and Ramanujam, 1986). We therefore aimed to complement
our financial dependent with an operational one to better capture the concept of new venture
performance.
One of a firm’s key parameters important for competitive success is its innovation
performance (Crossan and Apaydin, 2010). Being directly tied to the company’s value
creation, (product and process) innovations are likely to steer new ventures away from
potential familiarity, maturity and propinquity competence traps (Liao et al., 2003). They
elicit organizational learning and new knowledge accumulation (Maes and Sels,
forthcoming), which strengthens new ventures’ delicate competitive position.
Research has shown that various operationalizations of organizational innovation
exist. Following Crossan and Apaydin (2010), we selected three different measures to capture
three theoretically meaningful dimensions of new ventures’ innovation performance: product
portfolio, process technologies and served markets. First, we assessed the number of distinct
types of products or services the firm was offering at the time of the survey. This information
was transformed into a dummy variable differentiating single-product from multiple-product
firms. Second, we surveyed the number of process technologies used to create and deliver
these products or services. Possible values included ‘one dominant technology’ (46.97%),
11
‘two distinct technologies’ (12.12%), or ‘three or more distinct technologies’ (40.91%).
Third, concentrating on multi-market ventures, we examined the diversity of the markets to
which the firm targeted its products or services. Answers were scored on a five-point scale
ranging from ‘very equivalent’ (1) to ‘not equivalent at all’ (5). Together, the above measures
provide a comprehensive view of new ventures’ innovation performance. An aggregate ten-
point measure that was the sum of the above three measures was calculated for each firm
(e.g., 10 = multiple-product new venture, relying on minimum three distinct process
technologies, thereby serving disparate markets).
EFT level of autonomous and controlled entrepreneurship motivation. To assess
EFTs’ level of controlled and autonomous entrepreneurship motivation, we first determined
the entrepreneurship motivation of each of the venture’s founders. For this we adopted the
Gagné et al. (2010) and Vansteenkiste et al. (2009) inventory, which is grounded in SDT.
This 12-item inventory assesses four dimensions of work motivation, of which two pertain to
autonomous motivation (‘intrinsic motivation’ (Cronbach’s alpha = .764) and ‘identified
regulation’ (Cronbach’s alpha = .729)) and two to controlled motivation (‘external regulation’
(Cronbach’s alpha = .692) and ‘introjected regulation’ (Cronbach’s alpha = .851)). Prior
research has already demonstrated this inventory’s high levels of construct and concurrent
validity and internal consistency. Although initially aimed at measuring people’s motivation
for work-related behavior in the context of established organizations (Gagné et al., 2010), the
items are equally appropriate for measuring motivation of business founders in the context of
new ventures. Items were scored on a five-point Likert scale ranging from 1 (‘not important
at all’) to 5 (‘extremely important’). Using the formula suggested by Maes et al. (2005), we
determined for each founder two factors with scale ranges from 0 to 100: one representing
autonomous entrepreneurship motivation (Cronbach’s alpha = .827; 6 statements) and
another representing controlled entrepreneurship motivation (Cronbach’s alpha = .791; 6
statements). The adopted formula is displayed in the Appendix with all items and factor
loadings. To determine the EFTs’ level of autonomous (controlled) entrepreneurship
motivation, we calculated the average autonomous (controlled) motivation among its
members.
EFT entrepreneurship motivation heterogeneity. Motivation heterogeneity has seldom
been measured in the context of EFTs. To transform general EFT entrepreneurship
motivation information into team-based heterogeneity variables, we adopted Allison’s (1978)
coefficient of variation. This heterogeneity coefficient is constructed of the standard deviation
divided by the mean. A high score on this coefficient refers to high team heterogeneity,
whereas a low score denotes low heterogeneity. Using Allison’s (1978) formula, we
determined EFTs’ intrinsic and extrinsic entrepreneurship motivation heterogeneity.
Control variables. To isolate our hypotheses from possible rival explanations and to
minimize extraneous variation, we included the following organizational characteristics as
control variables: firm continuation, EFT size, industry, and environmental turbulence.
Furthermore, to explore the performance contribution specific to EFT motivations, we also
controlled for more traditional demographic diversity variables (e.g., gender, education,
industry experience).
Although all ventures in our sample are between one and three years of (legal) age,
not all of them are de novo firms (e.g., take-over of a bankrupt business). This implies that
the business activities could have been carried out before the current organizations were
legally established. We control for the possibility of firm continuation by including a dummy
variable indicating whether the business activities were already operational before the current
venture was founded. Because team heterogeneity and group size are positively associated
(Allison, 1978), it is imperative to control for EFT size. Prior work has indicated that larger
teams are linked to better performance, both at group level and firm level (Eisenhardt and
12
Schoonhoven, 1990). In order to prevent turbulence-originated bias in venture performance,
we introduced the dynamic nature of the venture’s environment as a control variable
(Ucbasaran et al., 2003). The scale items used to measure environmental turbulence are
adapted from Zahra (1993). Respondents were asked to rate six environmental statements on
a five-point Likert scale. Possible answers varied from ‘entirely disagree’ (1) to ‘entirely
agree’ (5). Again using the formula suggested by Maes et al. (2005), we created a factor with
scale ranges from 0 to 100 (Cronbach’s alpha = .855). The adopted formula is displayed in
the Appendix, along with all items and factor loadings. Because firms in our sample belong
to various industries, each with their own characteristics, we developed a series of dummy
variables to control for the different market conditions within each industry. Five dummies
were included in our analyses, using the manufacturing sector as a reference category.
Because Ucbasaran et al. (2008) have found a positive relationship between founder
experience and firm performance, we intended to control for EFTs’ entrepreneurship
experience. However, since at least one member of every EFT had already actively
participated in new venture creation, this dummy variable was excluded.
We again used Allison’s (1978) coefficient of variation to compute the age and
industry experience heterogeneity between the business founders. Due to collinearity issues
among the two independents, we were forced to exclude age and age heterogeneity from
further analysis. The remaining variables of heterogeneity (e.g., gender and educational
background) were developed using the Herfindal-Hirschman coefficient, also known as the
Blau categorical index (1977) (H = 1 - ?p
i
²). Before calculating the coefficient’s score, we
assigned the founders’ educational background to one of the following categories: arts,
sciences, engineering, business and economics, law and other. The above formula produced a
measure of heterogeneity with its complement being a measure of homogeneity. Though we
are unable to control for romantic couples, sample conditions did prevent the inclusion of
ventures in which a spouse was registered as a company partner without being a business
founder or assuming an active management role.
3.3 Statistical procedures
Hierarchical multiple regression was used as the statistical procedure to test our hypotheses.
As far as multivariate statistical tools are concerned, this technique has been subject to an
impressive set of tests of assumptions (Belsley et al., 1980). It allows investigating the
contribution above and beyond variables already entered into the regression equation. It also
enables us to examine the statistical influence of several variables at once. The variables were
mean-centered before any of the interaction terms were created. The highest VIF statistic
encountered in the models discussed below was 3.873, which is below the recommended
maximum value of 5 (Moreno and Casillas, 2008).
4. Results
Table 1 lists the means, standard deviations and bivariate correlations of this study’s
variables. All correlations are below .80 in absolute value, which is again an indication
against the possible presence of multicollinearity (Hair et al., 1998). The results of the
hierarchical regression analyses are displayed in Table 2 (financial performance) and Table 3
(innovation performance). Models 1, 2, 8 and 9 represent the ‘control models’, which only
include the aforementioned organizational control variables (Models 1 and 8), and the
traditional demographic diversity variables (Models 2 and 9). Models 3 and 10 relate to the
main effects, together with the control variables. We estimated each of the hypothesized
interaction terms in Models 4 to 7 (financial performance) and Models 11 to 14 (innovation
performance).
13
Table 1. Descriptive Statistics and Correlations
Variables Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
1. Return on assets 5.68 20.82 1
2. Innovation 4.98 3.13 -.14 1
3. Agriculture .08 .27 -.01 -.15 1
4. Construction .26 .44 .00 .20 -.17 1
5. Manufacturing .42 .50 .15 -.13 -.25
*
-.51
**
1
6. Transportation .03 .17 -.06 -.23 -.05 -.10 -.15 1
7. Banking and insurances .05 .21 -.10 .10 -.06 -.13 -.19 -.04 1
8. Professional services .17 .38 -.11 .09 -.13 -.26
*
-.38
**
-.08 -.10 1
9. Start-up continuation .00 .49 .02 -.10 .23 .05 -.06 .14 .03 -.22 1
10. EFT size 2.33 .44 .13 -.11 -.10 -.21 .19 -.06 -.08 .13 -.15 1
11. Environmental turbulence 46.97 22.10 -.04 .31
*
-.17 .18 -.19 -.23 -.07 .30
*
-.38
**
.01 1
12. EFT industry experience 13.90 9.31 .10 -.24 .14 -.10 .11 .13 .25
*
-.34
**
.35
**
-.09 -.30
*
1
13. EFT gender heterogeneity 20.94 24.63 -.09 .02 .22 .06 .00 -.15 -.04 -.14 .27
*
-.13 -.04 -.01 1
14.
EFT educational background
heterogeneity
20.96 24.62 -.17 .18 .11 -.01 -.24 -.15 -.04 .35
**
.14 -.11 .24 -.03 -.04 1
15.
EFT industry experience
heterogeneity
38.66 32.18 -.05 -.26
*
-.07 -.18 .36
**
-.20 -.23 .01 -.12 .40
**
.09 -.20 .07 -.02 1
16. EFT autonomous motivation 76.56 14.86 .20 -.16 -.13 -.12 .17 .01 .02 -.01 .03 .07 .06 .01 -.18 -.07 .29
*
1
17. EFT controlled motivation 35.82 17.11 .23 -.30
*
.17 .02 .11 .02 -.02 -.29
*
.07 -.03 .07 .04 .04 -.15 .17 .26
*
1
18.
Autonomous motivation
heterogeneity
23.87 26.53 -.11 .12 -.08 -.19 -.09 .14 .17 .23 -.09 .08 .00 -.10 -.35
**
.28
*
-.07 .33
**
-.23 1
19.
Controlled motivation
heterogeneity
112.36 60.44 -.06 .13 -.26
*
-.12 .03 .03 .00 .28
*
-.02 .09 -.05 -.04 -.17 .11 .08 .46
**
-.72
**
.48
**
Notes: **. Correlation is significant at the .01 level (2-tailed). *. Correlation is significant at the .05 level (2-tailed).
14
Table 2. Results of Hierarchical Regression Models of New Venture Financial Performance
Variables Model 1 Model 2
Model 3 Model 4 Model 5 Model 6 Model 7
Control variables: Organizational characteristics
Agriculture -.024 .008 -.034 -.039 -.064 -.019 -,047
Construction .045 .081 .044 .022 -.009 .058 ,023
Transportation -.062 -.132 -.153 -.144 -.156 -.125 -,123
Banking and insurances -.098 -.155 -.182 -.180 -.194 -.150 -,197
Professional services -.115 -.049 -.015 -.021 -.042 -.010 ,050
Start-up continuation .032 .085 .016 -.022 -.047 .002 ,046
EFT size .131 .164 .203 .215 .245 .218 ,178
Environmental turbulence -.012 .059 -.024 -.045 -.085 -.040 -,120
Control variables: Traditional diversity variables
EFT industry experience .095 .099 .105 .090 .070 ,104
EFT gender heterogeneity -.118 -.050 -.013 .026 -.078 -,097
EFT educational background heterogeneity -.167 -.108 -.088 -.039 -.056 -,135
EFT industry experience heterogeneity -.178 -.288* -.304* -.330* -.311* -,263
Main effects
EFT level of autonomous entrepreneurship motivation .196* .258* .260* .261** ,183
EFT level of controlled entrepreneurship motivation .206* .235* .266** .168 ,112
Moderators
Autonomous entrepreneurship motivation heterogeneity .121 -.044
Controlled entrepreneurship motivation heterogeneity -.153 -,322**
Two-way interactions
EFT level of autonomous entrepreneurship motivation x
autonomous entrepreneurship motivation heterogeneity
-.212
EFT level of controlled entrepreneurship motivation x
controlled entrepreneurship motivation heterogeneity
-.369**
F-Change .387 1.004 2.519** .602 .533 .801 4.630**
R²
.051 .118 .198 .207 .216 .210 .278
Notes: Standardized coefficients are shown (two-tailed, with directional hypothesis entries one-tailed); N = 66 start-ups, representing 142 founders;
***. Significant at the .01 level - **. Significant at the .05 level - *. Significant at the .10 level.
15
Table 3. Results of Hierarchical Regression Models of New Venture Innovation Performance
Variables Model 8 Model 9
Model 10 Model 11 Model 12 Model 13 Model 14
Control variables: Organizational characteristics
Agriculture -.177 -.154 -.094 -.090 -.044 -.102 -,081
Construction -.159 .000 .016 .034 .090 .009 ,035
Transportation -.220* -.208 -.191 -.198 -.177 -.206 -,208
Banking and insurances .047 .070 .081 .079 .103 .063 ,097
Professional services -.055 -.132 -.197 -.192 -.155 -.200 -,244
Start-up continuation .017 .031 .072 .101 .147 .080 ,046
EFT size -.100 .036 .011 .002 -.052 .003 ,033
Environmental turbulence .228 .213 .288** .304** .376** .296** ,356**
Control variables: Traditional diversity variables
EFT industry experience -.270** -.279** -.283** -.256* -.263* -,288**
EFT gender heterogeneity .045 .014 -.014 -.085 .030 ,045
EFT educational background heterogeneity .150 .104 .088 -.001 .075 ,134
EFT industry experience heterogeneity -.382*** -.318** -.306** -.260* -.305** -,342**
Main effects
EFT level of autonomous entrepreneurship motivation -.018 -.066 -.068 -.054 ,004
EFT level of controlled entrepreneurship motivation -.279** -.301** -.357*** -.258** -,216**
Moderators
Autonomous entrepreneurship motivation heterogeneity -.094 .205
Controlled entrepreneurship motivation heterogeneity .084 ,209
Two-way interactions
EFT level of autonomous entrepreneurship motivation x
autonomous entrepreneurship motivation heterogeneity
.382*
EFT level of controlled entrepreneurship motivation x
controlled entrepreneurship motivation heterogeneity
,273**
F-Change 1.524 2.703** 2.664* .466 2.318 .314 3.169*
R²
.176 .316 .381 .386 .414 .384 .422
Notes: Standardized coefficients are shown (two-tailed, with directional hypothesis entries one-tailed); N = 66 start-ups, representing 142 founders;
***. Significant at the .01 level - **. Significant at the .05 level - *. Significant at the .10 level.
16
Hypothesis 1a stated that the EFTs’ level of autonomous entrepreneurship motivation is
positively associated with new venture ROA. Hypothesis 1b suggested a similar relationship
for the EFTs’ level of controlled entrepreneurship motivation. Based on Model 3 of Table 2,
we can corroborate both hypotheses. While the regression coefficient of EFT autonomous
entrepreneurship motivation is significant and positive (? = .196), it appears not to be
significantly different from the one representing EFT controlled motivation (? = .206).
Consequently, we cannot conclude that an EFT’s level of autonomous entrepreneurship
motivation exerts a stronger influence on new venture ROA than its level of controlled
entrepreneurship motivation. Consequently, no support is found for Hypothesis 2.
Hypothesis 3a suggested that EFT autonomous entrepreneurship motivation is
positively associated with new ventures’ innovation performance. Model 10 of Table 3,
however, indicates that this relationship is insignificant (? = -.018). Hypothesis 3a receives,
therefore, no support. Hypothesis 3b suggested that the EFTs’ level of controlled
entrepreneurship motivation is negatively associated with new venture innovation. Model 10
of Table 3 empirically validates this hypothesis (? = -.279). We, thus, find support for
Hypothesis 3b.
Hypothesis 4a proposed that autonomous entrepreneurship motivation heterogeneity
positively moderates the relationship between EFT autonomous motivation and new venture
innovation. Hypothesis 4b stated the same for member controlled motivation heterogeneity
regarding the relationship between EFT controlled motivation and new venture innovation. In
other words, we expect teams with a higher level of entrepreneurship motivation to benefit
from higher motivation heterogeneity in terms of new venture innovation, whereas teams
with a lower level of entrepreneurship motivation should benefit from lower motivation
heterogeneity. We learn from Models 11 to 14 of Table 3 that both cross-products involving
motivation heterogeneity are positive and significant (? = .382; ? = .273). The graphical
representation of these interactions (Figures 1 and 2 below) confirms Hypotheses 4a and 4b.
Figure 1. Interaction Effect of EFT Autonomous Entrepreneurship Motivation and
Autonomous Entrepreneurship Motivation Heterogeneity on New Venture
Innovation
17
Figure 2. Interaction Effect of EFT Controlled Entrepreneurship Motivation and
Controlled Entrepreneurship Motivation Heterogeneity on New Venture Innovation
Hypothesis 5a suggested that autonomous motivation heterogeneity acts as a negative
moderator of the relationship between EFT autonomous entrepreneurship motivation and new
venture ROA. Hypothesis 5b proposed a similar influence for controlled motivation
heterogeneity on the relationship between EFT controlled motivation and new venture ROA.
In other words, we assumed that the anticipated positive main effects of team
entrepreneurship motivation on new venture ROA would be stronger if motivation
heterogeneity is low, and weaker if it is high. As shown by Models 4 to 7 of Table 2,
evidence only emerges for the cross-product involving controlled entrepreneurship
motivation heterogeneity (? = -.369). The graphical representation of this interaction (Figure
3 below) indicates that the relationship of EFT controlled entrepreneurship motivation with
new venture ROA is negative for EFTs with higher controlled motivation heterogeneity, and
positive for teams experiencing lower controlled motivation heterogeneity. Our results, thus,
corroborate Hypothesis 5b, whereas no support is found for Hypothesis 5a.
Figure 3. Interaction Effect of EFT Controlled Entrepreneurship Motivation and
Controlled Entrepreneurship Motivation Heterogeneity on New Venture Return on
Assets
18
Discussion
The research stream examining the complex relationship between team diversity and team
outcomes is impressive in volume (Horwitz and Horwitz, 2007). Since most new ventures are
founded by teams of entrepreneurs (McMullen et al., 2008), the importance of studying the
performance effects of EFTs’ characteristics increases. The present study is to be situated in
this domain. However, instead of focusing on bio-demographic EFT characteristics, we have
explored deeper-level motivation drivers of new venture performance, which have been
neglected in prior research (Van Knippenberg et al., 2004). Further, we introduced the aspect
of EFT motivational diversity into our arguments. In doing so, we believe we may also claim
to contribute to EFT diversity literature.
The specific objective of this study was to empirically examine the extent to which
entrepreneurship motivations of an entrepreneurial founding team affect new venture
financial and innovation performance. We first tested EFTs’ level of autonomous and
controlled entrepreneurship motivation as drivers of new venture return on assets and
innovation. We then introduced and empirically estimated the effect of two moderators
representing team motivation heterogeneity. Five main hypotheses were tested. In general,
support for these hypotheses was substantial. First and foremost, the study’s results
corroborate the distinction between autonomous and controlled entrepreneurship motivation.
We have demonstrated that both types can generate different effects. We can, therefore, rally
with Gagné and Deci (1995) against a unitary view on motivation. Our results also illustrate
the importance of EFTs’ motivational diversity or heterogeneity for new venture
performance. Deep-level diversity within EFTs is thus not to be neglected. We discuss our
findings in two subsequent sections: EFT autonomous motivation and EFT controlled
motivation.
EFT autonomous motivation
Within this study, a significant contribution of the EFTs’ level of autonomous
entrepreneurship motivation to new venture return on assets emerged. Contrariwise, no
significant link was found between the level of autonomous motivation and new venture
innovation. The latter finding comes somewhat unexpectedly. We discern the following
explanation for this absent link: Looking at Table 1, we observe that our sampled EFTs, on
average, display a high level of autonomous motivation (76.56). The level of controlled
motivation, on the other hand, is much lower (35.82), while its standard deviation is higher
(17.11 vs. 14.86). Our sample is thus characterized by a high level of autonomous motivation.
An explanation for the absent link lies in the close resemblance between becoming an
entrepreneur (e.g., starting up a new venture) and the innovation activities within young
organizations (e.g., bringing new goods and services to the market). By entering
entrepreneurship, EFT members’ general tasks resemble the tasks to be done when pursuing
innovation. Hence, it could be that the level of autonomous motivation within our sample is
too high for any additional autonomous motivation to trigger innovation effects. Securing
new venture return on assets, on the other hand, is far more of a managerial than a first-stage
(young) entrepreneurial activity. Resemblance between entrepreneurship and the latter is,
therefore, less outspoken, which allows for additional innovation effects to occur.
Consistent with our assumptions, our results revealed that autonomous motivation
heterogeneity acts as a positive moderator of the relationship between EFTs’ level of
autonomous motivation and new venture innovation. In other words, high motivation
heterogeneity was found to be instrumental to new venture innovation. Low heterogeneity, on
the contrary, appeared to be detrimental. Regarding new ventures’ financial performance, no
significant influence of autonomous motivation heterogeneity emerged. This finding seems to
19
indicate that only the level of EFTs’ autonomous entrepreneurship motivation affects new
ventures’ wealth creation, and not the distribution of that motivation within the founding
team. Again, this could be related to the aforementioned high level of EFTs’ autonomous
motivation within our sample, combined with a relatively low degree of inter-member
autonomous motivation heterogeneity (23.87; see Table 1).
EFT controlled motivation
Apart from internal or strongly internalized autonomous motives, our findings also shed light
on several interesting control-oriented motivation effects. In line with our expectations, the
level of controlled motivation was found to facilitate new ventures’ financial performance, as
reflected in their return on assets, whereas it hindered their operational performance, in terms
of firm innovations. As far as the interaction effects are concerned, both hypothesized
relationships were corroborated by our results. While low control-oriented motivation
heterogeneity facilitated new venture return on assets, it increasingly hindered new ventures’
ability to bringing new goods and services to the market. High motivation heterogeneity, on
the other hand, was found to foster new venture innovation, yet at the expense of short-term
firm financial performance.
We point out that our findings did not reveal a stronger effect of autonomous
entrepreneurship motivation on new venture financial performance compared to controlled
entrepreneurship motivation. In search of an explanation for this unexpected outcome, we
again turn to educational psychology literature. Within this strand of literature, it has been
suggested that autonomous motivated individuals have a higher probability of adopting a
long-term perspective (Watson et al., 1993). Control-oriented people, in contrast, tend to be
reluctant to take any actions that could endanger contingent external rewards. Instead, they
prefer to focus attention on the here and now (Erez et al., 1990). Although such a short-term
time perspective does not particularly encourage EFTs’ contribution to long-term venture
performance, it may trigger a contribution of controlled motivation to financial performance
similar to the one of autonomous motivation. While we believe this to be a mere short-term
outcome, additional research on this subject is imperative to empirically confirm this line of
thought.
All in all, we can say the levels of EFTs’ autonomous and controlled motivation
generate different effects when it comes to innovation performance, yet similar (positive)
effects regarding (short-term) financial performance. Furthermore, the (moderating) effects of
both types of motivational heterogeneity run parallel for innovation performance, whereas
they differ for financial performance. This implies that for studies focusing on understanding
deeper-level drivers of new venture performance within multi-founder firms it is essential to
consider not only the average level of the driver but also its heterogeneity within the
entrepreneurial founding team.
Limitations and Future Research
When interpreting our study findings, the following caveats should be recognized. First,
because this study makes use of self-describing interview data, it might suffer from social
desirability biases. However, we aimed to reduce possible biases by obtaining information
from multiple respondents (all of the founders) and by supplementing self-describing
information with more objective data (questionnaire and Bel-first data). Second, we only
assessed the entrepreneurship motivation of successful business founders. Yet, motivational
differences existing among nascent entrepreneurial team members could be very substantial,
encouraging some members to quit the team, or to freeze differences in order to not
jeopardize the emergence of the new firm. Future research could investigate these ideas
through a longitudinal study of nascent EFT dynamics. In this respect, the social
20
categorization theory of team diversity could be adopted as well, next to the
information/decision-making perspective. Hence, our sample could be biased with positive
selection because it includes only teams that were successful in creating a new venture. Third,
given the cross-sectional nature of our research design, we cannot prove the direction of the
cause-effect relationships in our model. However, by making use of a data structure in which
information on the dependent originates from a distinct database at a later point in time, we
established that founding team characteristics precede new venture performance. Fourth,
though the current study examines firms that are active within different industries, it does not
shed light on ventures in the retail industry. With START 2009 being its fourth wave, the
START research program has historically excluded the retail industry from its design, which
in most economies is one of the major industries for small and new businesses. However, as
the Belgian (and Flemish) economy is grounded in small and medium-sized businesses, with
micro-businesses (< 10 employees) being especially prevailing (94%), the retail industry is
typically not considered a dominant industry for new businesses in Belgium (European
Commission, 2012).
In conclusion, we distinguish the following recommendations for future research in
this area, on top of those already mentioned above. First, we urge researchers to repeat this
study’s moderator hypotheses for other new venture outcomes. Not only should this shed
light on possible contradictory mechanisms, it may equally extend our knowledge on the
contingencies that surround EFTs’ effectiveness. Further, following Horwitz and Horwitz
(2007), we identify the exploration of possible curvilinear relationships between motivation
level/heterogeneity and performance outcomes as an interesting and important line of
research. Another promising research direction is to adopt other psychological constructs
aimed at directly capturing unobservable team characteristics. For example, future
contributions might investigate how trust and friendship among the business founders
constitute important preconditions for team functioning. Alternatively, this research could
advance our understanding of the impact of founding team heterogeneity on the development
of network relationships and social ties, which already have been argued to benefit new
ventures (Shane and Cable, 2002). Future research should also look into the role of team
dynamics on new venture emergence and performance. Team composition is not always
fixed. While our sampling conditions required interview responses from all members of the
EFT, new members may join the founding team, while others might decide to leave it. In
turn, these flows are likely to affect social integration and team coherence. They also reflect a
transfer of knowledge, perspectives and resources, and the impact of these factors is not yet
clearly understood. Moreover, we also advise future research to take the intermember
relationships into account. Whereas motivation and motivational differences clearly are
strong forces affecting team cohesion, other elements are also important in this respect. We
could, for instance, think of family versus non-family linkages among EFT members and of
the distribution of firm ownership among founders. Finally, research on founding team
characteristics could include the moderating role of time on venture formation and
emergence, as established by Steffens et al. (forthcoming).
21
References
Adams JS (1963) Toward an understanding of inequity. Journal of Abnormal and Social
Psychology 67(5): 422–436.
Alderfer CP (1972) Existence, relatedness, and growth. New York: Free Press.
Allison P (1978) Measures of inequality. American Sociological Review 43(6): 865–880.
Amabile TM, Goldfarb P and Brackfield SC (1990) Social influences on creativity:
Evaluation, coaction, and surveillance. Creativity Research Journal 3(1): 6–21.
Amason AC (1996) Distinguishing the effects of functional and dysfunctional conflict on
strategic decision making: Resolving a paradox for top management teams. Academy
of Management Journal 39(1): 123–148.
Anderson JC and Gerbing DW (1988) Structural equation modeling in practice: A review and
recommended two-step approach. Psychological Bulletin 103(3): 411–423.
Baard PP, Deci EL and Ryan RM (2004) Intrinsic need satisfaction: A motivational basis of
performance and well-being in two work settings. Journal of Applied Social
Psychology 34(10): 2045–2068.
Barringer BR and Bluedorn AC (1999) The relationship between corporate entrepreneurship
and strategic management. Strategic Management Journal 20(5): 421–444.
Baum JR and Locke EA (2004) The relationships of entrepreneurial traits, skills and
motivation to subsequent venture growth. Journal of Applied Psychology 89(4): 587–
598.
Belsley DA, Kuh E and Welsch RE (1980) Regression diagnostics: Identifying influential
data and sources of collinearity. New York: Wiley.
Blau P (1977) Inequality and heterogeneity: A primitive theory of social structure. New
York: The Free Press.
Cameron J (2001) Negative effects of reward on intrinsic motivation - A limited
phenomenon: Comment on Deci, Koestner, and Ryan (2001). Review of Educational
Research 71(1): 29–42.
Campbell CA (1992) A decision theory model for entrepreneurial acts. Entrepreneurship:
Theory & Practice 17(1): 21–27.
Carsrud A and Brännback M (2011) Entrepreneurial motivations: What do we still need to
know? Journal of Small Business Management 49(1): 9–26.
Chandler GN, Honig B and Wiklund, J (2005) Antecedents, moderators and performance
consequences of membership change in new venture teams. Journal of Business
Venturing 20(5): 705–725.
Choi JN (2004) Individual and contextual predictors of creative performance: The mediating
role of psychological processes. Creativity Research Journal 16(2–3): 187–199.
Chowdhury S (2005) Demographic diversity for building an effective entrepreneurial team: Is
it important? Journal of Business Venturing 20(6): 727–746.
Cooper AC and Artz KW (1995) Determinants of satisfaction for entrepreneurs. Journal of
Business Venturing 10(6): 439–457.
Crossan MM and Apaydin M (2010) Multi-dimensional framework of organizational
innovation: A systematic review of the literature. Journal of Management Studies
47(6): 1154–1191.
DeCharms R (1968) Personal causation: The internal affective determinants of behavior.
New York: Academic Press.
Deci EL, Koestner R and Ryan RM (1999) A meta-analytic review of experiments examining
the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin 125(6):
627–668.
22
Deci EL and Ryan RM (1985) Intrinsic motivation and self-determination in human
behavior. New York: Plenum Publishing Co.
Deci EL, Ryan RM, Gagné M, Leone DR, Usunov J and Kornazheva BP (2001) Need
satisfaction, motivation, and well-being in the work organizations of a former Eastern
Bloc country. Personality and Social Psychology Bulletin 27(8): 930–942.
De Winne S and Sels L (2010) Interrelationships between human capital, HRM and
innovation in Belgian start-ups aiming at an innovation strategy. International
Journal of Human Resource Management 21(11): 1863–1883.
Eisenberger R, Rhoades L and Cameron J (1999) Does pay for performance increase or
decrease perceived self-determination and intrinsic motivation? Journal of
Personality and Social Psychology 77(5): 1026–1040.
Eisenhardt KM and Schoonhoven CB (1990) Organizational growth: Linking founding team,
strategy, and growth among U.S. semi-conductor ventures, 1978–1988.
Administrative Science Quarterly 35(3): 504–529.
Ensley MD and Hmieleski KM (2005) A comparative study of new venture top management
team composition, dynamics and performance between university-based and
independent startups. Research Policy 34(7): 1091–1105.
Ensley MD and Pearce CL (2001) Shared cognition in top management teams: Implications
for new venture performance. Journal of Organizational Behavior 22(2): 145–160.
Erez M, Gopher D and Arzi N (1990) Effects of goal difficulty, self-set goals, and monetary
rewards on dual task performance. Organizational Behavior and Human Decision
Processes 47(2): 247–269.
European Commission (2012) SBA Fact Sheet 2012 – Belgium.
http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/performance-
review/files/countries-sheets/2012/belgium_en.pdf
Gagné M and Deci EL (2005) Self-determination theory and work motivation. Journal of
Organizational Behavior 26(4): 331–362.
Gagné M, Forest J, Gilbert MH, Aubé C, Morin E and Malorni A (2010) The motivation at
work scale: Validation and evidence in two languages. Educational and
Psychological Measurement 70(4): 628–646.
Gartner WB, Shaver KG, Gatewood E and Katz JA (1994) Finding the entrepreneur in
entrepreneurship. Entrepreneurship: Theory & Practice 18(3): 5–9.
Grolnick WS and Ryan RM (1987) Autonomy in children’s learning: An experimental and
individual difference investigation. Journal of Personality and Social Psychology
52(5): 890–898.
Gu X, Solmon MA, Zhang T and Xiang P (2011) Group cohesion, achievement motivation
and motivational outcomes among female college students. Journal of Applied Sport
Psychology 23(2): 175–188.
Hair JF, Anderson RE, Tatham RL and Black WC (1998) Multivariate data analysis (5th ed).
Upper Saddle River: Prentice Hall.
Harrison DA, Price KH and Bell MP (1998) Time and the effects of surface- and deep-level
diversity on work group cohesion. Academy of Management Journal 41(1): 96–107.
Herron L and Robinson RB Jr (1993) A structural model of the effects of entrepreneurial
characteristics on venture performance. Journal of Business Venturing 8(3): 281–294.
Herzberg F (1966) Work and the nature of man. Cleveland, OH: World.
Hinsz VB, Tindale RS and Vollrath DA (1997) The emerging conceptualization of groups as
information processes. Psychological Bulletin 121(1): 43–64.
Hmieleski KM, Cole MS and Baron RA (2012) Shared authentic leadership and new venture
performance. Journal of Management 38(5): 1476–1499.
23
Horwitz SK and Horwitz IB (2007) The effects of team diversity on team outcomes: A meta-
analytic review of team demography. Journal of Management 33(6): 987–1015.
Jehn KA, Northcraft GB and Neale MA (1999) Why differences make a difference: A field
study of diversity, conflict, and performance in workgroups. Administrative Science
Quarterly 44(4): 741–763.
Kamm JB, Shuman JC, Seeger JA and Nurick AJ (1990) Entrepreneurial teams in new
venture creation: A research agenda. Entrepreneurship: Theory & Practice 14(4): 7–
17.
Kuratko DF, Hornsby JS and Naffziger DW (1997) An examination of owner’s goals in
sustaining entrepreneurship. Journal of Small Business Management 35(1): 24–33.
Langan-Fox J and Roth S (1995) Achievement motivation and female entrepreneurs. Journal
of Occupational & Organizational Psychology 68(3): 209–218.
Liao J, Welsch H and Stoica M (2003) Organizational absorptive capacity and
responsiveness: An empirical investigation of growth-oriented SMEs.
Entrepreneurship: Theory & Practice 28(1): 63–85.
McClelland DC (1961) The achieving society. New York: D. Van Nostrand Company, Inc.
McMullen JS, Bagby DR and Palich LE (2008) Economics freedom and the motivation to
engage in entrepreneurial action. Entrepreneurship: Theory & Practice 32(5): 875–
895.
Maes J and Sels L (forthcoming) SMEs’ radical product innovation: The role of internally
and externally oriented knowledge capabilities. Journal of Small Business
Management.
Maes J, Sels L and Roodhooft F (2005) Modeling the link between management practices
and financial performance. Evidence from small construction companies. Small
Business Economics 25(1): 17–34.
Maslow AH (1954) Motivation and personality. New York: Harper & Row.
Milliken FJ and Martins LL (1996) Searching for common threads: understanding the
multiple effects of diversity in work groups. Academy of Management Review 21(2):
402–433.
Moreno AM and Casillas JC (2008) Entrepreneurial orientation and SMEs: A causal model.
Entrepreneurship: Theory & Practice 32(3): 507–528.
Nemeth CJ and Staw BM (1989) The tradeoffs of social control and innovation within groups
and organizations. In: Berkowitz L (ed) Advances in experimental social psychology,
Vol. 22. New York: Academic Press, pp. 175–210.
Olsen BJ, Parayitam S and Twigg NW (2006) Mediating role of strategic choice between top
management team diversity and firm performance: Upper echelons theory revisited.
Journal of Business and Management 12(2): 111–126.
Podsakoff PM, MacKenzie SB, Lee JY and Podsakoff NP (2003) Common method biases in
behavioral research: A critical review of the literature and recommended remedies.
Journal of Applied Psychology 88(5): 879–903.
Porter LW and Lawler EE III (1968) Managerial attitudes and performance. Homewood, IL:
Irwin-Dorsey.
Prabhu V, Sutton C and Sauser W (2006) Creativity and certain personality traits:
Understanding the mediating effect of intrinsic motivation. Creativity Research
Journal 20(1): 53–66.
Ramanujam V and Varadarajan P (1989) Research on corporate diversification: A synthesis.
Strategic Management Journal 10(6): 523–552.
Ryan RM and Deci EL (2000) Self-determination theory and the facilitation of intrinsic
motivation, social development, and well-being. American Psychologist 55(1): 68–78.
24
Segal G, Borgia D and Schoenfeld J (2005) The motivation to become an entrepreneur.
International Journal of Entrepreneurial Behaviour & Research 11(1): 42–57.
Shane S and Cable D (2002) Network ties, reputation, and the financing of new ventures.
Management Science 48(3): 364–381.
Shepherd DA and DeTienne DR (2005) Prior knowledge, potential financial reward, and
opportunity identification. Entrepreneurship: Theory & Practice 29(1): 91–112.
Smith KG, Collins CG and Clark KD (2005) Existing knowledge, knowledge creation
capability, and the rate of new product introduction in new high-technology firms.
Academy of Management Journal 48(2): 346–357.
Steffens P, Terjesen S and Davidsson P (forthcoming). Birds of a feather get lost together:
New venture team composition and performance. Small Business Economics DOI:
10.1007/s11187-011-9358-z.
Talaulicar T, Grundei J and von Werder A (2005) Strategic decision-making in start-ups, the
effect of top management team organization and processes on speed and
comprehensiveness. Journal of Business Venturing 20(4): 519–541.
Ucbasaran D, Westhead P and Wright M (2008) Opportunity identification and pursuit: Does
an entrepreneur’s human capital matter? Small Business Economics 30(2): 153–173.
Ucbasaran D, Lockett A, Wright M and Westhead P (2003) Entrepreneurial founder teams:
Factors associated with member entry and exit. Entrepreneurship: Theory & Practice
28(2): 107–128.
Van Knippenberg D, De Dreu CKW and Homan AC (2004) Work group diversity and group
performance: An integrative model and research agenda. Journal of Applied
Psychology 89(6): 1008–1022.
Vansteenkiste M, Soenens B, Sierens E, Luyckx K and Lens W (2009) Motivational profiles
from a self-determination perspective: The quality of motivation matters. Journal of
Educational Psychology 101(3): 671–688.
Venkatraman N and Ramanujam V (1986) Measurement of business performance in strategy
research: A comparison of approaches. The Academy of Management Review 11(4):
801–814.
Vroom VH (1964) Work and motivation. New York: Wiley.
Watson WE, Kumar K and Michaelsen LK (1993) Cultural diversity’s impact on interaction
process and performance: Comparing homogenous and diverse task groups. Academy
of Management Journal 36(3): 590–602.
West CT and Schwenk CR (1996) Top management team strategic consensus, demographic
homogeneity and firm performance: A report of resounding nonfindings. Strategic
Management Journal 17(7): 571–576.
Wigfield A, Guthrie JT, Tonks S and Perencevich KC (2004) Children’s motivation for
reading: Domain specificity and instructional influences. Journal of Educational
Research 97(6): 299–309.
Williams KY and O’Reilly CA (1998) Demography and diversity in organizations: A review
of 40 years of research. Research in Organizational Behavior 20 (1998): 77–140.
Wood R and Bandura A (1989) Social cognitive theory of organizational management.
Academy of Management Review 14(3): 361–384.
Zahra SA (1993) Environment, corporate entrepreneurship, and financial performance: A
taxonomic approach. Journal of Business Venturing 8(4): 319–340.
Zahra SA, Ireland RD and Hitt MA (2000) International expansion by new venture firms:
International diversity, model of market entry, technological learning, and
performance. Academy of Management Journal 43(5): 925–950.
25
Appendix
Appendix Factor Loadings and Cronbach’s Alphas.
Item
External
regulation
Introjected
regulation
Controlled
motivation
Identified
regulation
Intrinsic
motivation
Autonomous
motivation
Environmental
turbulence
I put in effort because this allows me to make more money. .869 .237 .513 .229 .004 .201 /
I put in effort to obtain more work security. .830 .346 .578 .398 .138 .355 /
I put in effort because I would lose financial rewards otherwise. .606 .633 .741 .089 -.314 -.135 /
I put in effort because I would feel ashamed otherwise. .309 .879 .780 .074 -.110 -.054 /
I put in effort to avoid disappointment from others. .364 .845 .784 .203 -.067 .040 /
I put in effort so that I would not have to feel guilty. .280 .837 .741 .128 -.082 -.012 /
I put in effort because this entrepreneurial work matches with my personal values. .272 .170 .142 .348 .771 .707 /
I put in effort because I find my entrepreneurial work very significant. .227 .075 .154 .856 .559 .778 /
I put in effort because my entrepreneurial work allows me to reach my life goals. .351 .220 .345 .911 .354 .668 /
I put in effort because I have fun doing this type of work. -.188 -.346 -.403 .358 .802 .702 /
I put in effort because I enjoy this work very much. .110 -.033 -.017 .620 .695 .761 /
I put in effort because I find entrepreneurial work extremely interesting. -.090 -.148 -.218 .451 .843 .772 /
Within our industry the possibilities for technological innovations are considerable. / / / / / / .672
Within our industry a lot of opportunities for new products and/or new services exist. / / / / / / .766
Within our industry customer demand for new products and/or new services is increasing. / / / / / / .814
Within our industry the need for a new technology is growing. / / / / / / .852
Within our industry the market for new products and/or new services is expanding. / / / / / / .714
Our industry requires technological innovations in order to continue to grow. / / / / / / .759
N 142 142 142 142 142 142 66
Cronbach’s alpha .692 .851 .791 .729 .764 .827 .855
Notes: Extraction Method: Principal Components Analysis; Promax rotation; To compute both factors we made use of the following formula: F = ((S - V) / ((V x W) - V)) x
100 with S equal to the sum of all initial values (before transformation), V referring to the number of variables and W representing the number of scale points (Maes et al.,
2005); The use of two distinct methods to collect the data (questionnaires and interviews) minimizes possible common-method variance effects. Open-ended questions were
interspersed with other types of questions, which prevented respondents from adopting a scale-based pattern linked to Likert or semantic differential scales (Podsakoff et al.,
2003); Harman’s single factor test was used to examine concerns of possible common-method variance (interview information). Multiple factors with eigenvalues greater
than one emerged. The first factor only explained 30.4% of the variance; Construct validity was established by developing measures from well-grounded theory (Barringer
and Bluedorn, 1999); Confirmatory factor analysis was performed to assess the convergent and discriminant validity. The fit of the unconstrained two-factor model (including
both constructs in a way that each item loaded solely on the factor for which it was an intended indicator) was reasonably good (e.g., GFI = .81) and better than the fit of the
four-factor model (convergent validity). The pair-wise difference between the chi-squared value of the unconstrained model and that of the constrained model largely
exceeded 3.84 (5% critical value) (discriminant validity) (Anderson and Gerbing, 1988).
doc_377398424.pdf