Opportunity tension: bridge between entrepreneurship and complexity science Towards further integration of process and emergent theories
Wouter Johan van Monsjou – 5732964 Submitted: 17/04/2015 – final version
MSc. in Business Administration – Entrepreneurship & Innovation Amsterdam Business School
Tsvi Vinig (1st) and Wietze van der Aa (2nd) Number of words: 9.206
Table&of&Contents&
Statement of originality ... 3!
Abstract ... 4!
Introduction ... 6!
Theory ... 9!
From opportunity to tension ... 9!
Opportunity tension ... 12!
Human capital and cognition ... 13!
Perceived opportunity and commitment ... 16!
Commitment, enactment and organizing dynamics ... 17!
Count of activity ... 19!
Rate of activity ... 20!
Timing of activity ... 22!
Conceptual model ... 23!
Data and method ... 23!
Sample ... 23! Variables ... 25! Dependent variable ... 25! Independent variables ... 25! Control variables ... 26! Method ... 27! Results ... 29! Discussion ... 36! Conclusions ... 39! References ... 42! Appendices ... 46!
Statement&of&originality&
This document is written by Wouter van Monsjou, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract&&
This article argues opportunity tension to be an essential concept in bringing entrepreneurship and complexity science closer together. First, it further embeds the complexity science concept of opportunity tension within the entrepreneurship literature by integrating it with bricolage, effectuation, structuration and creation theory. Second, a logistic regression is used to study the effect of opportunity tension on emergence and the mediating effects of count and rate of activity on that relationship. Data comes from the widely accepted and often-used longitudinal PSED II database. The findings prove opportunity tension to be a robust predictor of emergence.
“No problem can be solved from the same level of consciousness that created it.”
Introduction&
There is little doubt among scholars that the creation of a new venture can be seen as a journey, quest or more general: a process. It is a process in which an opportunity emerges from the interactions between the entrepreneur and its environment. Subsequent successful exploitation of the opportunity is achieved through interdependent and iterative organizing activities conducted by the entrepreneur, over time establishing legitimacy with its audience so as to enable exchange between the firm and its environment. (Gartner 1985; Katz and Gartner, 1988; Delmar and Shane, 2004; Dimov, 2010)
Central to this process is the concept of opportunity. For long (Shane and Venkataraman, 2000; Sarasvathy, et al., 2012) the concept of opportunity has become increasingly important to the field of entrepreneurship, some stating that without opportunity there is no entrepreneurship (Short et al., 2010); others acknowledging the origins of an opportunity can be seen as a defining puzzle to entrepreneurship (Suddaby, Bruton and Si, 2014).
Several process theories have been proposed and developed to explain what is the nature of opportunities, how is the interaction with the entrepreneur, and most importantly how do they help us explain and understand the creation of a new firm: bricolage (Baker and Nelson, 2005), effectuation (Sarasvathy, 2001), discovery versus creation (Alvarez and Barney, 2007) and structuration theory (Sarason, Dean and Dillard, 2006).
Common topics within these theories are whether opportunities exist as singular phenomena external to or created by and thus idiosyncratic to the entrepreneur; whether
entrepreneurs are different (i.e. entrepreneurial orientation) from non-entrepreneurs making them more successful in recognizing and exploiting opportunities; and whether agency is important and how is it distributed (Suddaby, Bruton and Si, 2014).
Whereas some scholars use terminology that suggest mutually exclusive mechanisms – discovery versus creation (Alvarez and Barney, 2007); causation versus effectuation (Sarasvathy, 2001); others try to bring seemingly distinct mechanisms together by explaining how they interact and co-evolve: imprinting and reflexivity (Suddaby, Bruton and Si, 2014) and structuration theory (Sean, Dean and Dillard, 2006).
However, these process theories are often studied using stage model designs - assuming a linear path of venture development along a number of ex-ante identifiable stages and predictable transitions - or through static frameworks that map out the process’ main concepts and its causal links without accounting for temporal dynamics (Moroz and Hindle, 2012). These models are useful for exploring what concepts are contributing to the creation of the firm, yet fail to answer how these constructs interact (especially over time), nor do they answer how these constructs and interactions lead to the eventual creation of a new firm.
Selden and Fletcher (2014, p. 12) question whether these related yet different process constructs and theories can be brought together and shed light on the overall emergent dynamics of the venture creating process. Others find most process models to be quite artificial and unable to explain the constructs, relationships and drivers that describe the creation of a new firm (Levie and Lichtenstein, 2010). Instead, they propose to approach the phenomenon with theory and methods better equipped for studying the non-linear, co-evolving, temporal and emergent dynamics that are at play. Efforts to
further develop this stream of theory range from harmonizing the existing entrepreneurship process theory (Moroz and Hindle, 2012); educating scholars in using longitudinal panel methodology (Daviddson and Gordon, 2012); shifting from Gaussian towards Paretian distributions (Crawford and McKelvey, 2012); variance to process perspectives (McMullen and Dimov, 2013) to introducing concepts from emergence and complexity science (Lichtenstein and McKelvey, 2011; Lichtenstein, 2014). Indeed – crossing disciplinary boundaries – increasingly scholars are focusing their efforts on getting a better understanding on the properties, process and outcome of new venture creation (Katz and Gartner, 1988; Brush et al., 2008; Levie and Lichtenstein, 2010; Goldstein, Hazy and Silbertang, 2010).
It becomes clear the path between the fields of entrepreneurship and those of emergence and complexity is an important one and is gradually being paved out before us. This article builds on those efforts, focusing on the integration of process, emergence and complexity theories, using opportunity tension as a bridge. Lichtenstein (2009) describes opportunity tension as the key driver of entrepreneurial order creation, or more specific: “Opportunity tension is initiated when an entrepreneur identifies and begins to develop a business opportunity, i.e. an energy differential which defines a (niche) market, and simultaneous constructs a way to capitalize on that economic potential through a unique and sustainable business model”. It is a promising concept inhibiting the capacity to bring together entrepreneurship process theories (Levie and Lichtenstein, 2010, p. 34) and expand the toolbox of entrepreneurship researchers with concepts and methods from complexity science.
This article is structured as follows. First, process and emergent theories are discussed. Second, opportunity tension is introduced, bringing these theories together and highlighting how perceived opportunities give lead to tension and subsequent organizing. Third, the PSED II dataset and outcome of the logistic regression with serial mediation are presented. Fourth, these results will then be discussed as to how they contribute to the integration and understanding of opportunity tension.
Theory&
From&opportunity&to&tension&
The opportunity, the entrepreneur, and the mechanisms connecting these two distinct concepts, have been a centerpiece of entrepreneurial advancement. The individual-opportunity nexus as introduced by Shane (2001) has proven to be fertile ground for further entrepreneurship theorizing. Increasingly, theories try to not look at the individual and the opportunity as separate concepts but try to integrate them into one. They acknowledge the lack of power of the opportunity alone to predict new venture creation (Dimov 2007) and recognize interdependencies between the two concepts (Sarason, Dean and Dillard 2006).
Having presented a thorough overview of three distinct typologies of opportunities – recognition, discovery and creation – Sarasvathy et al. (2010) conclude their paper with a most inclusive definition of the venture creation process:
“Ergo, the lags (temporal and otherwise) between any invention and the creation of new economic welfare enabled by it, require not only the ability and alertness to recognize, and the perception and perseverance to discover opportunities for the achievement of pre-determined goals such
as increasing profits and larger market shares, but also necessitate decisions and actions based often only on human imagination and human aspirations, that may or may not in time lead to new products, firms and markets.”
While this definition gives a clear indication on what elements (see emphasis added) contribute to the creation of a new firm, it reveals little of how a new venture comes into existence. Note that opportunity theories do not necessarily strive to explain how opportunity leads to emergence; they focus on recognition, discovery and creation mechanisms of the opportunity. As such, here it is not the intention to criticize these theories, rather to make the distinction that highlights how these ‘what’-theories can be used as valuable stepping stones to inform ‘how’-theories of venture creation.
To construct such ‘how’ -theories, the dynamics between these interdependent elements and how these may or may not in time lead to new products, firms and markets, should be analyzed. Indeed, Dimov (2007) states that the pursuit of an opportunity is not powerful enough to predict whether a venture will emerge, rather the gradual, iterative and reflexive process is of interest. However, even these words fail to describe the emergence of a new venture. The processes that lead to emergence may be gradual and iterative, yet the event itself is the birth of something new (Bhave, 1994); is preceded by significant, punctuated shifts (Lichtenstein et al., 2006); resulting in order creation (Chiles, 2010) or genuine novelty (Foster and Metcalfe, 2012).
It is the field of complexity that introduces a richer vocabulary enabling us to describe the emergence of a new firm: ranging from threshold and critical values to phase transitions, bifurcation points, criticalization, and tipping points. These are all concepts not describing additive linear processes, rather multiplicative non-linear points of change, leading to a qualitative new state of the venture (Crawford and McKelvey, 2012, p. 8). It
is those dynamics that many stress to be crucial to understanding the venture creation process.
Adding to that, in emergence – the philosophical construct, not the simple noun – we find a useful framework providing the necessary dimensions needed to understand and conceptualize the interdependent, co-evolving dynamics that are at play. Emergence is conceived as a process whereby larger entities, patterns, and regularities arise through interactions among smaller or simpler entities that themselves do not exhibit such properties. Goldstein (1999) provides the basic properties of emergent phenomena. 1) They exhibit radical novelty and thus cannot be fully anticipated in their full richness ex-ante. 2) They are coherent, being an integrated whole. 3) They are observed at the macro level, that is the level higher than its lower-level components. 4) They are dynamical, arising as a complex system over time. 5) They are ostensive, or in other words can be recognized.
Increasingly these properties are used to construct a much-needed framework for entrepreneurship researchers interested in the interdependent elements and interactions that lead to venture creation. Katz and Gartner (1988) and Brush, Manolova and Edelman (2008) essentially define what properties are essential for emerging ventures, namely intentions, resources, boundary and exchange. In addition, with the use of their emergent hierarchy framework Fletcher and Selden (2014) help us understand how a business idea evolves from one level to the next (e.g. the entrepreneur sense-making, entrepreneur-stakeholder, and entrepreneurial firm level) and how different endogenous, self-causing, context-creating, self-organizing and path-dependent dynamics are present in each stage.
Complementary to Fletcher and Selden (2014), Lichtenstein (2015, in press) introduces three different degrees of emergence, going from order, to systemic, to radical emergence. In the first, agents (entrepreneurs) through their interactions create order or rules internal to the system (venture). Still, such rules are often not being observed or recognized as something new or emerged. In the case of second-degree emergence it is about the creation of a new entity that operates at a higher level than its components, exhibiting genuine novelty. In entrepreneurship this would be the successful creation of a new venture. The third, radical emergence happens when these higher-level entities exert influence on the lower level components, also known as supervenience. Such effects and emergence are present in the creation of new industries.
Where process theories try to explain the interactions between entrepreneur and opportunity, emergent theories try to explain how these far-from-equilibrium interactions can lead to nonlinearities, adaptive tension and the creation of novelty that in turn result in the emergence of new order. (Lichtenstein, 2009)
Above most basic building blocks of complexity science and the concept of emergence are necessary to frame and conceptualize the dynamic venture creation process. It should be clear the line between entrepreneurship and complexity science is thin and synergies can be found in further integration of the two fields. The next section thus explores the origins of opportunity tension and how well it blends in with the entrepreneurship literature.
Opportunity&tension&&
The workings of opportunity tension can be traced back to the complexity science concept of adaptive tension. Adaptive tension is an energy-differential or state of
disequilibrium that has a tendency to be dissipated or resolved (Nicolis and Prigogine, 1977). It can be seen as the perceived difference between the system’s current and desired states (Lichtenstein et al., 2007), a driver of order-creation processes through phase transitions (McKelvey, 2004).
Applied to entrepreneurship this difference can be described as a perceived opportunity or a personal aspiration to start a business (Lichtenstein, 2007). In the context of entrepreneurship McKelvey (2004) describes adaptive tension as a difference between resources within the entrepreneur and those he wants to access, that needs to be resolved. It is this perceived disequilibrium that can be seen as the driver for entrepreneurial action. Opportunity tension, as defined by Levie and Lichtenstein (2010) is “the perception (co-creation) of an untapped market potential, and the commitment to act on that potential by creating value.” The interdependent and co-evolving nature of the opportunity tension concept clearly stems from this definition: the perception of the opportunity and the enactment upon it are intrinsically linked and only together – like two sides of the same coin - bring about opportunity tension. At the same time it becomes clear opportunity tension is a dynamic process unfolding over time – the untapped opportunity being converted into value - and dependent on the agency of one or more entrepreneurial agents as they are committed to act.
Human&capital&and&cognition&
The first half of the coin – the perceiving, imagining or recognizing of an opportunity – stems from the characteristics and cognitive capabilities of the entrepreneur. For example, much research has focused on whether entrepreneurs differ from non-entrepreneurs or whether entrepreneurs exist before their entrepreneurial
journey or are created during that process. Alvarez and Barney (2007) credit the earlier mixed results on both issues to the inability of previous work to distinguish between discovery and creation theory. They explain that the first assumes entrepreneurs are different from non-entrepreneurs but says nothing about how the entrepreneurial journey affects them; while the latter is agnostic about ex-ante differences, but assumes significant differences are created during the exploitation of opportunities.
Still - in order to imagine or perceive an untapped market potential - an entrepreneur needs to possess some kind of knowledge about that market gained through previous education, startup experience or working experience (Shane, 2000; Mathias, Williams and Smith, 2015). Indeed human capital variables and their impact on successful venture creation have been of much interest in entrepreneurship literature. While many have theorized that for example education or previous startup experience can have a positive impact on entrepreneurial success, little empirical support has been found. Liao and Welsch (2008) found no effect of previous education on the gestation period of a venture-in-creation. Tornikoski and Newbert (2007) surprisingly found negative relationships between entrepreneurial experience and several factors of emergence.
Instead, job, market or industry experience seems to be a more valuable predictor. Liao and Welsch (2008) find a positive effect of professional experience on the outcome of the gestation period. Dimov (2010) distinguishes between entrepreneurial experience and industry experience and found that only industry experience was able to predict successful venture emergence. On the other hand Torniksoki and Newbert (2007) find no relationship of industry experience or college education on new venture creation. From a qualitative perspective Mathias, Williams and Smith (2015) find prior work experience to
be a powerful source of imprint, suggesting the experience these entrepreneurs have in the field, allows them to exploit opportunities within known knowledge fields.
The entrepreneur combines these pieces of knowledge in useful ways in order to create novel means-ends relationships. These combinations can be seen as fact sets from the social and economic milieu (Sarason Dean and Dillard, 2006) or entrepreneurial stories shaped by the entrepreneur (Lounsbury and Glynn, 2001). In either case - much like creationist theory (Alvarez and Barney, 2007) - opportunities are socially constructed by and idiosyncratic to the entrepreneur and his interpretation of the social context. More recently, Suddaby et al. (2015) add to this notion and explain how their concept of reflexivity holds that opportunities “are generated by reflection on the possibility of new and creative social realities.” In fact, the interpretations, reflections and judgments of the entrepreneur are about future social realities (Dimov, 2010, p. 1127) - or more specific about future customer needs (Chiles et al., 2010) – and give direction or purpose to the entrepreneurial organizing. According to Foster and Metcalfe (2012) it is the mental activity of human beings that gives rise to emergent capital patterns.
Table I: Comparison of key definitions between complexity science and entrepreneurship Complexity science Entrepreneurship Reference
Emergence Venture creation
Goldstein (1999), Katz and Gartner (1988), Bhave (1994), Dimov (2010), Lichtenstein (2015, in press)
Adaptive / opportunity tension
A perceived opportunity that drives enactment
Nicolis and Prigogine (1977), McKelvey (2004), Levie and Lichtenstein (2010), Dimov (2010), Fletcher and Selden (2014)
Energy-differential
Untapped market potential / Perceived opportunity
McKelvey (2004), Levie and Lichtenstein (2010)
Dissipate Enact / Organize
Nicolis and Prigogine (1977), Edelman and Yli-Renko (2010), Chiles et al. (2010)
Perceived&opportunity&and&commitment&
Whenever the entrepreneur imagines a novel product (Chiles et al., 2010, p. 29), endows resources with new purposes (Baker and Nelson, 2005) or realizes a certain problem has not been solved in the current equilibrium state (Goldstein, Hazy and Silberstang, 2010, p. 105) he or she in fact creates a perceived disequilibrium: a gap to be filled. This perceived opportunity – a difference between the current and a desired state – gives rise to opportunity tension. Opportunity tension can be seen as the entrepreneurs’ internal drive or intention, a great personal passion or ‘creative tension’ to start exploiting the opportunity (Lichtenstein 2015, in press).1 McKelvey (2001) refers to this drive as motivational valence: the felt belief the opportunity is viable and worth pursuing.
Not surprisingly then, tension is also described by drive theories in psychology, where the need for tension reduction leads to motivation: setting a goal deliberately creates tension and reaching a goal releases this tension (Carsrud et al., 2009). Foster and Metcalfe (2012) state that humans ‘form aspirational goals of some kind and seek to achieve them in states of uncertainty’. These goals - similar to perceived opportunities - “are mental representations of what the future could be similar to, enabling individuals to persist” (Carsrud and Brännback, 2011, p. 17).
Shane, Locke and Collins (2003, p. 268) like to capture all these highly interrelated concepts of ambition, goals, energy and stamina, and persistence under the umbrella term drive: the willingness to put effort. Instead here - following the example of Hop and Sonderegger (2014) – this willingness to put effort is referred to as commitment. In general the more opportunity tension there is, or in other words the stronger the entrepreneurs’ startup commitment is, the more likely a new venture will emerge (Dimov, 2010).
H1 There is a positive relationship between opportunity tension and emergence.
Commitment,&enactment&and&organizing&dynamics&
A distinction can be made between final and instrumental motivation (Carsrud et al., 2009) where the first one is aimed at reaching a certain goal – the creation of the new venture – and the latter at doing something that leads to the final goal – organizing. In the words of Fletcher and Selden (2014, p. 7): “a commitment to a particular end involves both an intended outcome and ancillary expectations and criteria for evaluating short-run events and interim goals.” Thus, the drive, intention or passion to enact the opportunity
also incites the entrepreneur to make things do (Baker and Nelson, 2005) effectuate (Sarasvathy, 2001) or simply: start organizing. Indeed the moment the entrepreneur realizes exploitation of the opportunity cannot be done alone he or she starts to organize: ‘setting up networks of connections between elements, such as machines and people; using sets of organizational rules; and accessing appropriate and affordable energy sources’ (Foster and Metcalfe, 2012). Higher levels of opportunity tension justify the organizing effort needed to address an opportunity (Goldstein et al., 2010 p. 106). Organizing - or (re)combining resource modules – is needed to bring the opportunity to material fruition (Chiles et al., 2010). In bricolage – making do with what is at hand - this recombining of resources leads to processes of testing and counteracting limitations (Baker and Nelson, 2005).
This brings us to the second half of the opportunity tension coin: the commitment to act on the potential by creating value results in a recursive process of continuous testing of the opportunity (Levie and Lichtenstein, 2010). Both process researchers and complexity scientists highlight this recursive nature of the entrepreneurial process. Opportunities are explored and exploited through the iterative process of interpretation and influence (Sarason, Dean and Dillard, 2006); the perceiving, imagining and acting on (Chiles et al., 2010) or adaptation through self-organization (Lichtenstein, 2007). In their pursuit of the opportunity entrepreneurs take action in order to test, and retest - when needed redefine and re-contextualize - their pre-held assumptions (Fletcher and Selden, 2014). Because of this trial and error basis these recursive processes give rise to creativity, improvisation, and various social and network skills of the entrepreneur (Baker and Nelson, 2005, p. 354). Renko, Kroeck and Bullough (2012) find that personal
learning and personal growth opportunities are strong motivations to spend more time organizing. During the process, increasingly entrepreneurs are able to make more informed judgments about their assumptions and feasibility of the opportunity (Dimov, 2007).
Creating a new venture is a true test of the entrepreneur’s commitment. While the entrepreneurs stated motivations, beliefs, aspirations or intentions could lead to new venture creation, a stronger predictor might be their behavior, or actions. After all, actions are manifestations of commitment (Mowday, Steers and Porter, 1979). Empirical findings in earlier studies show how different dynamics of organizing can impact the venture creation success rate. Dimensions like duration, type, sequence, rate, concentration and timing of activity can all influence new venture creation (Carter, Gartner and Reynolds 1996; Lichtenstein et al. 2007; Liao and Welsch 2008).
Count&of&activity&
While not all results are consistent, in general a larger count of startup activities seems to lead to higher probabilities of new venture creation. Renko, Kroeck and Bullough (2012) and Honig and Samuelsson (2012) find that the number - or accumulation - of startup activities affects operational business status significantly in some of their models, but not in others. Where Gartner and Reynolds (1996) find a positive effect of average number of activities on new venture creation, Lichtenstein et al. (2007) – in a study concerning exactly these organizing dynamics - do not even include the number of activities. Furthermore, Menzies et al. (2006) found the number of gestation activities to be significantly predicting the creation of an operation business. Moreover – while predicting the non-occurrence of a quit is not the same as predicting the successful
creation of a firm - Shane and Delmar (2004) find the number of activities undertaken to be strongly reducing the hazard of terminating the startup process. Last, Edelman and Yli-Renko (2010) find efforts to create a venture – operationalized as the sum of activities conducted – to be a consistent predictor of venture start-up, even when perceived market opportunity is included as a mediating factor.
H2a: The count of activity mediates the relationship between opportunity tension and emergence.
H2b: There is a positive relationship between the count of activity and emergence. Rate&of&activity&
When it comes to rate of activity – that is the number of activities taking place within a given period of time – findings predominantly support that those processes characterized by higher rates are more likely to lead to new venture creation. Lichtenstein et al. (2007) find support for higher rates leading to new venture creation. Carter, Gartner and Reynolds (1996) find the rate of activity after the first year positively affecting new venture creation. Still some negative or non-significant results are found. Brush, Manolova and Edelman (2008) find – contrary to their expectations - that slower organizing is more effective. whereas Hopp and Sonderegger (2014) found no significant effect of rate on successful venture creation.
The relationship between the count and rate of activity - most notably highlighted by Hopp and Sonderegger (2014) – is ambiguous, as can be seen from different significance levels of both variables under different circumstances. This might indicate a mediating effect of the rate of activity on the relationship between count of activity and emergence is present. Also looking at how the rate of activity is calculated – with count in the denominator of the rate function – it can be expected the number of activities influences the rate of activity.
H3b: The rate of activity mediates the relationship between the count of activity and emergence.
Carsrud et al. (2009) and Fletcher and Selden (2014) distinguish between the commitment of the entrepreneur that can have a final goal – successful emergence of the venture – and an instrumental goal – the organizing of activities in order to create the new venture. Given this line of reasoning, opportunity tension might positively affect the startup process outcome both directly, and indirectly through the organizing of activities, where both the amount of as well as the rate at which these activities are executed matter. Indeed Storm (2012) finds opportunity tension to be a strong influencer of the rate of activity and concentration.
H3c: The count of activity and the rate of activity sequentially mediate the relationship between opportunity tension and emergence.
Timing&of&activity&
Still, higher counts and rates of activity alone might not lead to new venture emergence. As discussed before, emergence of a new venture implies genuine novelty as a result of some punctuated shifts, forces that can overcome a certain threshold. Such a transition is characterized by a state change. From adaptive tension we learn that tension leads to a phase transition if it exceeds a certain critical value (McKelvey 2001). Lichtenstein (2007) calls this momentum, as operationalized as later timing of activity.
He finds later timing of activity to be a good predictor of new venture creation (Lichtenstein 2007). In another (single case) study Lichtenstein (2006) finds the entrepreneur experiencing higher levels of frustration just before the moment her venture changes from one state to another. Carter, Gartner and Reynolds (1996) find the rate of activity after the first year positively affecting new venture creation. This implies that entrepreneurs that only actively organize activities the first year are more probably giving up after this early organizing, being consistent with the findings of Lichtenstein et al. (2007) that later timing of activity is able to predict new venture creation.
As momentum, or later timing of activity is an important part of opportunity tension and expected to positively impact the probability of emergence, it is included in the model. Yet, due to the focus of this article on the effect of opportunity tension on emergence and the mediating effects of count and rate of activity, the effect of later timing of activity is not explicitly hypothesized.
Conceptual&model&
Figure I: Conceptual model of hypothesized relationships
Data&and&method&
Sample&The data comes from the Panel Study of Entrepreneurial Dynamics (PSED) II, which followed nascent entrepreneurs in their venture creation efforts over a period of six years. Yearly interviews were held, from 2005 to 2011, specifically designed to collect relevant information needed to shed light on the following central question: where do startups come from? It is the only database that includes data covering the entire venture creation process from conception to eventual new firm birth, allowing researchers to explore the dynamics that lead to the eventual emergence of a new venture. (Reynolds and Curtin 2008, p. 161)
Both the original data as well as the harmonized data2 on start-up transitions and outcomes have been combined. The original consists of a huge collection of variables covering the entrepreneurs’ background and motivation, the start-up concept, information about the entrepreneurs’ financial conditions and social environment among others. The harmonized data is derived from the original and consists of variables dealing with startup activities and transitions. Two transitional moments are measured, namely going from still active startup, to new firm or quit; and going from new firm to still profitable or quit. Also included are timestamps for all activities and transitions. From these timestamps several variables are derived, like time between the first and eight startup activity and time from active startup to new firm. (Reynolds and Curtin, 2008)
PSED II consists of 1.214 cases of nascent entrepreneurs with no positive cash flow at the time of the first point of contact. They were selected from a representative sample of 31.845 American respondents. Adjusting for several other selection criteria. 965 cases are suitable for start-up outcome analysis, of which 221 were able to emerge as a successful new venture. (Reynolds and Curtin, 2011)
An additional two criteria are established. First, because of ease of comparability only single startup founders are selected. Second, those startups that did not experience a transition to a profitable new venture or quit – identified as still active – are dropped from the analysis. Still active startups are those that have not shown a transition to another state because 1) the entrepreneurs could not be reached anymore or 2) they were simply still active while the interviews came to an end. The main reason to exclude these still active startups is because they blur the organizing dynamics measures.
Some independent variables such as count and rate of activity – count the number of activities undertaken until the moment of transition (new firm or quit), where this moment of transition is expressed in the form of a timestamp. While the same calculations apply for still active startups, these startups have not experienced a transition, therefore making measures as count, rate and timing of activity to transition meaningless and unreliable. Eliminating those cases results in a database comprised of 642 startups, of which 200 reached a state of profitability and 442 saw themselves forced to quit.
Variables&
Dependent&variable&&
Emergence is a binary measure with a value of 0 for startups that eventually quit and a value of 1 for startups that were successfully created. A quit is based on the respondents’ claim that they exert little work on, expect no future work for and no future career plans involve this startup. The criteria for a startup to be counted as a successfully created new venture is that it has to be profitable. Still active firms are excluded from the analysis as explained above. (Reynolds and Curtin, 2011)
Independent&variables&
Opportunity tension is measured using a 5-item scale asking the entrepreneur to what extent he or she agrees to several statements. These are about the capacity for maximum personal effort, the willingness to do whatever it takes; desirability of entrepreneurship as a career option and relevance of the new firm to personal goals. Taking these items together results in one scale of opportunity tension (! = 0.718), resembling the Intensity
construct as described by Reynolds and Curtin (2008). The difference being that opportunity tension includes one extra item (I am confident I can put in the effort needed to start this new business), resulting in a higher internal validity. See for the individual items Appendix I.
Count of activity is measured as the number of activities that have been undertaken between the moment of the first activity and that of the transition. Activities range from initial serious thoughts and business plan initiated tos hired an accountant and joined a trade association.3 Similar to the criteria by Reynolds and Curtin (2011) cases with less than 3 start-up activities based on this measure were excluded. As a total of 33 distinct startup activities have been reported, this measure can vary between 4 and 33.
Rate of activity is calculated as the number of activities divided by the time from the first activity to the moment of transition. To account for the non-normal distribution of organizing rate, the log of the rate of activity is calculated.
Control&variables&
To control for entrepreneurial background age, gender, income and education are included. Being an important part of opportunity tension previous industry experience is included and measured in years of experience. Industry experience refers to the entrepreneur having experience in the same industry the startup will compete in. Next, an entrepreneurial motivational measure – achievement - is included, and is captured in a 4-item scale (! = 0.716). Achievement is argued to impact new venture success rates significantly (Carsrud et al., 2009). See for the individual items Appendix I. To control
for regional and industry effects, dummies on urban density and startup industry are included.
Last to control for organizing dynamics, spread and timing of activity are included. Spread is measured as the standard deviation of activities between first activity and moment of transition. Thus a lower spread means activities are clustered closer together. For the same reason as with rate of activity, spread of activity is logged. Timing is measured as the average time to activity before the moment of transition, divided by the time to transition, resulting in a ratio of 0 to .99. For ease of interpretation these values were then multiplied by 100, resulting in an average timing of activity lying between 0% and 99% of the process. Cases with timing of 0 were then excluded.
Method&
Due to the dichotomous nature of the dependent variable a binominal logistic regression analysis is performed. The logistic regression enables us to predict the probability of a certain binary outcome by taking the log of the odds of the event of interest, where the odds is the probability of positive outcomes, divided by the probability of negative outcomes. !""# = !!(!!!)!(!!!). The log translates the odds values so that a linear regression can be conducted on Y. Instead of minimizing the sum of squared errors, the logistic regression maximizes the likelihood of observing the sample values.
The ! -coefficients can easily be translated into odds-ratios through exponentiation: !""# = ! !!. These odds-ratios can then be interpreted as a change in the
odds-ratio of Y for a one-unit change in X, with odds-ratios higher than 1 increasing the probability of a positive outcome and odds-ratios lower than 1 decreasing this probability.
It should be noted that calculating in- and decreases of odds-ratios is done through multiplication, not addition.
In addition, to analyze the mediating effects of the count and rate of activity on the relationship between opportunity tension and emergence a serial mediation is included.
&Results&&
Table II: Descriptive statistics
Variable N Min Max Mean S.D.
Emergence 698 0 1 0.31 0.46 Entrepreneur Gender 698 0 1 0.59 0.49 Age 691 18 83 43.63 12.92 Income 659 1 10 6.24 2.90 Education 697 1 6 3.27 1.13 Industry experience 696 0 54 8.73 10.29 Achievement 698 1 5 2.59 0.98 Opportunity tension 698 2 5 4.19 0.61 Organizing dynamics Count 688 4 33 15.27 6.57 Spread (log) 694 -1.49 5.07 1.81 0.98 Rate (log) 698 -4.50 2.25 -0.91 0.92 Timing 698 1 99 38.61 22.20 Urban area
Metro: outside center 698 0 1 0.20 0.40
Metro: suburban 698 0 1 0.19 0.39
Metro: no city center 698 0 1 0.03 0.17
Non-metropolitan 698 0 1 0.28 0.45 Industry Restaurant 697 0 1 0.04 0.19 Consumer Service 697 0 1 0.35 0.48 Health Service 697 0 1 0.07 0.26 Manufacturing 697 0 1 0.05 0.21 Construction 697 0 1 0.07 0.25
Agriculture & Mining 697 0 1 0.04 0.21
Wholesale 697 0 1 0.05 0.21
Transportation 697 0 1 0.01 0.11
Utilities & Communications 697 0 1 0.03 0.18
Finance & Insurance 697 0 1 0.02 0.15
Real Estate 697 0 1 0.06 0.24
Table III: Correlations Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 Emergence 2 Gender 0.06 3 Age 0.01 -0.02 4 Income 0.01 0.06 0.14 5 Education 0.08 -0.02 0.22 0.29 6 Industry experience 0.11 0.16 0.33 0.04 0.08 7 Achievement 0.02 0.01 -0.08 -0.15 -0.06 0.06 8 Opportunity tension 0.11 0.02 -0.17 -0.08 -0.17 0.06 0.29 9 Count 0.30 0.02 0.04 0.27 0.19 0.10 0.03 0.06 10 Spread -0.03 0.08 0.06 -0.06 0.03 0.17 0.19 0.07 0.08 11 Rate 0.23 -0.05 -0.06 0.15 0.04 -0.10 -0.15 -0.07 0.31 -0.71 12 Timing 0.22 0.05 -0.09 -0.02 -0.01 0.03 0.11 0.10 -0.04 0.25 -0.32
13 Metro: outside center 0.03 -0.02 0.02 0.08 0.05 0.03 -0.06 -0.03 0.05 0.02 0.01 0.01
14 Metro: suburban -0.02 -0.01 0.01 0.13 0.02 -0.06 0.02 0.07 0.02 -0.03 0.02 -0.04 -0.24
15 Metro: no city center -0.10 0.00 -0.03 0.02 -0.01 -0.06 0.02 -0.04 -0.05 -0.02 -0.01 -0.04 -0.09 -0.09
16 Non-metropolitan 0.00 0.00 0.02 -0.17 -0.14 0.05 -0.05 0.01 -0.01 0.01 0.00 0.02 -0.31 -0.30 -0.11 17 Restaurant -0.06 0.02 -0.03 -0.12 -0.07 0.01 0.06 0.04 -0.03 0.03 -0.06 0.04 0.00 -0.03 -0.03 18 Consumer Service -0.01 -0.03 -0.03 -0.06 -0.06 -0.13 -0.04 -0.04 -0.09 0.01 0.01 -0.01 -0.02 0.02 -0.03 19 Health Service -0.01 -0.13 0.03 -0.05 0.10 0.01 0.01 -0.01 -0.04 -0.01 -0.03 0.07 0.03 -0.02 0.00 20 Manufacturing 0.00 0.05 0.10 0.05 -0.01 0.09 0.04 0.00 0.04 0.05 -0.02 -0.05 -0.02 0.02 0.09 21 Construction 0.03 0.12 -0.07 -0.08 -0.07 0.07 -0.10 -0.02 -0.01 -0.01 0.01 0.03 0.01 -0.05 -0.05 22 Agriculture & Mining 0.05 0.05 0.03 0.02 -0.02 0.06 -0.06 -0.01 0.05 0.07 -0.08 -0.01 0.03 -0.03 -0.04 23 Wholesale -0.01 0.04 -0.03 0.07 -0.05 -0.06 0.02 -0.03 0.05 -0.03 0.02 0.01 -0.05 0.07 0.10 24 Transportation -0.02 0.05 0.03 -0.01 -0.09 0.07 -0.01 0.00 0.00 0.04 -0.03 0.04 -0.03 -0.06 -0.02 25 Utilities & Communications 0.01 -0.02 0.05 0.00 0.05 0.04 0.06 -0.04 -0.05 0.05 -0.07 0.00 0.06 -0.01 -0.03 26 Finance & Insurance 0.06 0.02 -0.03 0.02 0.03 -0.01 -0.02 0.05 0.05 -0.07 0.08 -0.03 -0.01 0.06 0.01
Table III: Correlations (continued) 1 2 3 4 5 6 7 8 9 10 11 12 16 17 -0.03 18 0.03 -0.13 19 -0.03 -0.05 -0.20 20 0.01 -0.05 -0.18 -0.07 21 -0.01 -0.05 -0.18 -0.07 -0.06 22 0.15 -0.04 -0.17 -0.07 -0.06 -0.06 23 -0.02 -0.04 -0.16 -0.06 -0.05 -0.06 -0.05 24 0.01 -0.02 -0.08 -0.03 -0.03 -0.03 -0.03 -0.03 25 -0.06 -0.03 -0.13 -0.05 -0.04 -0.05 -0.04 -0.04 -0.02 26 -0.04 -0.03 -0.11 -0.04 -0.04 -0.04 -0.04 -0.04 -0.02 -0.03 27 -0.02 -0.05 -0.18 -0.07 -0.06 -0.06 -0.06 -0.06 -0.03 -0.05 -0.04 28 -0.10 -0.06 -0.22 -0.08 -0.07 -0.07 -0.07 -0.07 -0.03 -0.05 -0.05 0.08
Notes: Correlations with absolute values 0.084 or higher are significant for p > 0.01.
The descriptive statistics for all variables used in the logistic regression can be seen in Table II. Most notable is the lower N of income, as 66 respondents have refused to state their income. The bivariate correlations are then presented in Table III. Other than the high correlations between count and rate (0.31) and spread and rate (-0.71), there are no signs of possible multicollinearity. Further testing proved no multicollinearity exists as VIF values no larger than 5 are found, with spread showing only a VIF of 1.189.4
It can be observed that count and rate have a correlation of 0.31, suggesting that duration – the nominator when calculating rate of activity – introduces enough variability so that both predictors show no signs of multicollinearity. The positive sign is to be expected, as a higher count in the denominator of the rate of activity will lead to a higher
4 Additional testing revealed duration – the time between the first activity and the moment of
rate of activity. Next, the strong correlation between spread and rate of -0.71 implies that processes characterized by high rates of organizing have activities clustered more closely together.
Next, the regression outputs are presented. First, a two-step logistic regression is presented in Table IV, distinguishing between Model 1 – including only the covariates - and Model 2 – adding both the main effect and the mediators. Here the answers to H1a, H2a, and H3a are covered. Overall Model 2 is significantly better at predicting the outcome of the venture creation process, showing a LL of 631.01, down with 102.08 from 733.81, significant at the 0.001 level. Model 2 has been able to correctly predict 77.4% of the startup process outcomes as opposed to the base model with an overall percentage of 66.8%. In terms of sensitivity and specificity Model 2 produced values of 51.0% and 89.4% respectively. Hypothesis 1a predicted a positive relationship between opportunity tension and emergence. From Model 2 in Table IV we find the coefficient of opportunity tension to be positive and significant (! = 0.475. ! < 0.01). These results suggest opportunity tension leads to higher probabilities of emergence. Hypothesis 2a predicted a positive relationship between count of activity and emergence. Here the coefficient of count of activity appears to be positive and weakly significant (! = 0.038!! < 0.10). These figures imply that startup processes characterized by a higher number of activities in the period preceding the venture outcome are more likely to result in successful emergence. Hypothesis 3a expected a positive effect of rate of activity on emergence. Looking at the rate of activity coefficient we find it to be positive and strongly significant (! = 1.432. ! < 0.001). These results suggest that indeed startup processes where activities are undertaken at a faster rate are more likely to result in successful emergence.
Table IV: Logistic estimation on venture emergence
Model 1 Model 2
Variable Coefficient S.E. Coefficient S.E.
Opportunity tension 0.475 ** 0.18 Count 0.038 ° 0.02 Rate 1.432 *** 0.24 Gender 0.107 0.20 0.070 0.21 Age 0.002 0.01 0.016 ° 0.01 Income -0.039 ** 0.04 -0.103 ** 0.04 Education 0.157 0.09 0.114 0.10 Industry experience 0.019 0.01 0.011 0.01 Achievement 0.007 0.10 -0.140 0.11 Spread -0.267 ** 0.10 0.608 ** 0.20 Timing 0.024 *** 0.01 0.038 *** 0.01
Metro: outside center 0.108 0.25 0.004 0.28
Metro: suburban -0.228 0.27 -0.382 0.30
Metro: no city center -2.114 ° 1.09 -1.388 1.13
Non-metropolitan -0.097 0.24 -0.191 0.26 Restaurant -1.028 0.70 -1.168 0.79 Consumer Service 0.371 0.31 0.365 0.34 Health Service 0.093 0.43 0.182 0.47 Manufacturing 0.262 0.53 0.117 0.58 Construction 0.431 0.44 0.332 0.50
Agriculture & Mining 0.802 ° 0.48 1.132 * 0.54
Wholesale 0.343 0.50 0.170 0.56
Transportation -0.158 0.87 -0.435 1.09
Utilities & Communications 0.469 0.55 1.110 ° 0.59
Finance & Insurance 1.339 * 0.62 0.971 0.69
Real Estate 0.512 0.44 0.471 0.48 Business Consulting 0.491 0.43 0.285 0.47 Constant -2.148 0.62 -5.247 1.06 LL 733.81 631.01 !! 62.68 165.47 Δ!! 102.80 *** Nagelkerke !! 0.131 0.320 ° p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001
From the control variables we observe income and spread of activity to be significant predictors emergence. Timing is strongly significant, with timing later in the process leading to a higher probability of emergence. Spread has a positive effect on emergence, indicating that when activities are spread out over a longer period of time this will increase the probability of successful new venture creation. It is also interesting to note how the sign of spread of activity changes from negative to positive in Model 1 and 2 respectively. Urban density seems to have no effect. Last, startups that operate in the agricultural & mining or utilities & communications industry seem to be more likely of emerging than other startups.
Second, the path coefficients and indirect effects on the mediators are shown in Table V.5 Here, answers to H1b, H2b and H3b can be found. For a mediating effect to be significant, four criteria have to be met: 1) the total effect of X on Y should be significant; 2) the effect of X on M should be significant; 3) the effect of M on Y controlling for X should be significant and 4) the resulting direct effect of X on Y should be weaker than the original total effect of X on Y (MacKinnon, 2008). These effects are called paths c, a, b and c’ respectively and can be found looking at the coefficients presented in their respective regressions. After 1000 rounds of bootstrapping in PROCESS, the boot confidence intervals are produced, indicating the effect is significant whenever the interval does not include zero.
Figure II: Path analysis
Table V: Path Coefficients and Indirect Effects
Path Coefficients Indirect Effects
to Count to Rate to Emergence Boot 95% Confidence Interval Estimate Opp. tension 1.15*** (.42) -0.03 (.03) 0.47** (.18) Count 0.06*** (.00) 0.04° (.02) Rate 1.43*** (.23) Total .10 (.08) -.06; .23 Ind1 (a1a2) .04 (.03) -.00; .13 Ind2 (a2b2) -0.4 (.05) -.16; .04 Ind3 (a1d12b2) .10** (.05) .01; .19 Notes: ° p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001
Hypothesis 1b predicted the positive effect of opportunity tension on emergence to be mediated by the count of activity. While path c, a1 and b1 are positive and significant; the indirect effect fails to produce significant results, finding no support for H1b.
Hypothesis 2b predicted rate of activity to mediate the effect of count of activity on emergence. First, running a separate regression of count of activity on emergence,
(! = 0.0595, ! < 0.001) . Next, both d12 and b2 are positive and significant. Multiplying these specific indirect effects results in a positive and significant indirect effect of count of activity on emergence (! = 0.0852, ! < 0.001). The resulting direct effect of count of activity on emergence is not significant at the 0.05 level. These results suggest the effect of count of activity on emergence is completely mediated by rate of activity.
Hypothesis 3b predicted the positive effect of opportunity tension on emergence to be serially mediated through both count of activity and rate. As c’, a1, d12 and b2 are all significant, mediation can be present. Calculating the indirect effect of path a1d12b2 a positive and significant estimate is produced (! = 0.0979, !!"# = !0.0052; !"#$ = 0.1853). While the mediating effect is found to be significant the resulting direct effect of opportunity tension on emergence c’ is still significant ! = 0.4745, ! < 0.05 , indicating only partial mediation. These results provide support for hypothesis 3b, implying that the effect of opportunity tension on emergence is partially mediated through count and rate of activity.
Discussion((
Current process theories have shaped the field of entrepreneurship and developed our understanding on what contributes to the creation of a new venture. To solve some of the still unanswered questions, such as how ‘something’ (i.e. new venture) is created out of ‘nothing’ (i.e. emerges), a different level of analysis or consciousness is needed. However, before complexity science theories and methodology can be applied to the field of entrepreneurship, a credible bridge between the two fields needs to be built.
The main contribution of this article is the integration of both entrepreneurship concepts with complexity dynamics. As from the complexity science perspective it is repeatedly observed that patterns of organizing are strong predictors of emergence, it is interesting to see these dynamics do not fully cancel out the entrepreneurship variables like age and income. Indeed opportunity tension - in the form of intentions, commitment or persistence of the entrepreneur - is a reliable factor in explaining the successful start of a new venture. Even after finding evidence that opportunity tension predicts emergence through a higher count and faster rate of activity, the direct effect of opportunity tension is not fully mediated away.
These results show how opportunity tension in the form of both willingness to put effort and actions increases the probability of successful venture emergence. This touches on the discussion on whether intentionality precedes other startup activities. While it can be said intentionality is needed to bring about - and thus precede – a higher count and rate of activity, intentionality in itself contributes to emergence during the entire startup process. This is an indication that intentions do not only precede other elements of the process but work alongside them as well. (Brush, Manolova and Edelman, 2008)
Another finding is that while the count of activity in itself can be an important factor in explaining emergence, it is actually through co-determination of the rate of activity that count of activity indeed leads to a higher probability of emergence. This shows that researchers should be careful when referring to count and rates of activity interchangeably (Edelman and Yli-Renko, 2010, p. 840). Moreover, excluding rate of activity might give an overestimation of the positive effect count of activity has on emergence. Whenever possible is would be advisable to include rate of activity as explanatory variable in favor of simple count of activity.
In addition, it is interesting to find that opportunity tension and organizational dynamics are found to be consistent for all levels of urban density and almost all types of industry.
In line with the many attempts to further bridge the gap between entrepreneurship and other disciplines – most notably that of complexity science and emergence - this article adds to these efforts as it further integrates entrepreneurship and complexity science through the concept of opportunity tension. While the use of a logistic regression is still suboptimal when it comes to studying such emergent phenomena it contributes by modeling the paths through which opportunity tension leads to emergence. The model also links elements of classic entrepreneurial theories such as bricolage, opportunity creation and effectuation with complexity science dynamics, enabling researchers to find common grounds.
Off course, these findings are not without limitations. First of all, in fact only one side of the coin that is called opportunity tension has been properly researched. This is mainly due to the selection criteria of the PSED II, as only serious entrepreneurs that had already started organizing have been selected. The perceiving and shaping of, and the decision to start to enact on the opportunity have - for the most of it - already taken place and are therefore relatively constant over all cases. This might explain the non-significant effect of industry experience on opportunity tension. While indeed industry experience is often found to be capable of predicting new venture creation, it could in fact be most critical in determining who decides to commit to serious venture creation efforts in the first place. In order to study the dynamics of opportunity tension in full a comparable number of individuals less serious or not organizing at all should be added and analyzed to predict who of those perceive an opportunity and decide to enact it.
Another important factor to keep in mind when interpreting the results is that - for reasons explained before - only startups that have either emerged or quit entirely have been used in the regression. While it resulted in clearer findings it might lead to wrongful readings of the findings. Given this particular dataset the probability of successful venture creation might be best measured as portion of all startup-processes started (200/948 = 0.21), however the results above express this probability as portion of those cases that actually experienced a transition (startup or quit) - thus excluding still active startups (200/642 = 0.31).
Last, as noted above, linear (logistic) regressions are not fully capable to explain the phenomenon of interest. Indeed, increasingly scientist highlight the need for entrepreneurship researchers to equip themselves with other tools – such as Complex Adaptive Systems and Agent Based Models – to study the disorderly, non-linear, emergent dynamics of venture creation. While I strongly support these recommendations, I humbly admit the acquiring of these skills solely for the purpose of this article would be a bridge too far.
Conclusions(
Startups are intriguing entities indeed. The gradually evolving pursuit of an opportunity as a prologue to the sudden emergence of something genuinely new, is a process that surely keeps many souls awake at night. Either deeply involved in the creation of one, or rising high above studying the patterns of many, those involved in entrepreneurship will be continuously confronted by increasingly complex puzzles. By examining how entrepreneurs envision, shape and commit to their own paths of creation, this article is an
Opportunity tension – where the perceiving and enactment of an opportunity come together – is a promising concept that lies at the crossroads of entrepreneurship and complexity science. As the number of calls for further integration of these fields is growing, this particular concept is worth of further attention. Indeed, it appears to be capable of bringing together complexity science concepts such as emergence and dissipative structures with entrepreneurship concepts as new venture creation, perceived opportunity and enactment. Where process theories in entrepreneurship explain the gradual evolutions of entrepreneur and opportunity, complexity science brings in the power of explaining sudden transitions of one state to another.
Entrepreneurs, their intentions and actions are the driving forces of venture creation. It is their ability to envision a viable opportunity that holds somewhere in the future and the decision to commit to the exploitation of that opportunity that incites them to take action. With the cognitive creation of this opportunity a tension or disequilibrium arises that needs to be resolved by the entrepreneurs’ actions. Whether these actions will indeed lead to the successful creation of the venture is largely determined by the commitment of the entrepreneur. The willingness to exert effort and stick to the exploitation of the opportunity is in itself a strong predictor of reaching the state of a profitable new venture.
Moreover this opportunity tension also determines to some extent the dynamics of the organizing process and through it has its effect on the outcome of the process. It is this organizing that enables the entrepreneur to acquire the right set of resources, attract investors and reach out to consumers before a venture emerges. While for every startup the contents, the in- and outputs of every such activity are widely different, there exist identifiable patterns common to all startups powerful enough to help explain emergence.
Most notably, a higher rate, a wider spread and later timing of activity lead to higher probabilities of emergence. The count of activity alone does not seem to matter: only as part of the rate of activity is it powerful enough to make meaningful predictions.
Opportunity tension then seems to be a fruitful bridge between entrepreneurship and complexity science. Still, more research like that of Storm (2012) and Hop and Sonderegger (2014) is needed to strengthen the explanatory power of the concept. Additionally, the use of methods other then logistic regression might be able to further uncover the organizing patterns as these might be better capable of describing the dynamics that surround phase transitions, tipping points and emergence.
In the end, my hope is that in overcoming these and other shortcomings the concept of opportunity tension will eventually become the answer to the double-barreled question by Moroz and Hindle (2012, p. 781) “what is both generic and distinct about entrepreneurship (as a process)?”
References(
Alvarez,)S.)A.,)&)Barney,)J.)B.)(2007).)Discovery)and)creation:)Alternative)theories)of) entrepreneurial)action.)Strategic)entrepreneurship)journal,)1(1-2),)11C26.) Baker,)T.,)&)Nelson,)R.)E.)(2005).)Creating)something)from)nothing:)Resource) construction)through)entrepreneurial)bricolage.)Administrative)science) quarterly,)50(3),)329C366.) Bhave,)M.)P.)(1994).)A)process)model)of)entrepreneurial)venture)creation.)Journal)of) business)venturing,)9(3),)223C242.) Brush,)C.)G.,)Manolova,)T.)S.,)&)Edelman,)L.)F.)(2008).)Properties)of)emerging) organizations:)An)empirical)test.)Journal)of)business)venturing,)23(5),)547C566.) Carsrud,)A.,)Brännback,)M.,)Elfving,)J.,)&)Brandt,)K.)(2009).)Motivations:)The) entrepreneurial)mind)and)behavior.)In)Understanding)the)entrepreneurial)mind) (pp.)141C165).)Springer)New)York.) Carsrud,)A.,)&)Brännback,)M.)(2011).)Entrepreneurial)motivations:)what)do)we)still) need)to)know?.)Journal)of)Small)Business)Management,)49(1),)9C26.) Carter.)N.)M.,)Gartner.)W.)B.,)&)Reynolds.)P.)D.)(1996).)Exploring)startCup)event) sequences.)Journal)of)business)venturing.)11(3).)151C166.) Chiles,)T.)H.,)Tuggle,)C.)S.,)McMullen,)J.)S.,)Bierman,)L.,)&)Greening,)D.)W.)(2010).) Dynamic)creation:)Extending)the)radical)Austrian)approach)to) entrepreneurship.)Organization)Studies,)31(1),)7C46.) Crawford,)G.)C.,)&)McKelvey,)B.)(2012).)Strategic)implications)of)powerClaw) distributions)in)the)creation)and)emergence)of)new)ventures:)PowerClaw) analyses)in)three)panel)studies.)Frontiers)of)Entrepreneurship)Research,)32(12),) 1.) Davidsson.)P.,)&)Gordon.)S.)R.)(2012).)Panel)studies)of)new)venture)creation)C)a) methodsCfocused)review)and)suggestions)for)future)research.)Small)Business) Economics.)39(4).)853C876.) Delmar,)F.,)&)Shane,)S.)(2004).)Legitimating)first:)Organizing)activities)and)the) survival)of)new)ventures.)Journal)of)Business)Venturing,)19(3),)385C410.) Dimov,)D.)(2007).)Beyond)the)singleCperson,)singleCinsight)attribution)in) understanding)entrepreneurial)opportunities.)Entrepreneurship)Theory)and) Practice,)31(5),)713C731.) Dimov.)D.)(2010).)Nascent)entrepreneurs)and)venture)emergence)C)Opportunity) confidence,)human)capital,)and)early)planning.)Journal)of)Management)Studies.) 47(6).)1123C1153.) Edelman,)L.,)&)YliCRenko,)H.)(2010).)The)impact)of)environment)and)entrepreneurial) perceptions)on)ventureCcreation)efforts:)bridging)the)discovery)and)creation) views)of)entrepreneurship.)Entrepreneurship)Theory)and)Practice,)34(5),)833C 856.) Fletcher,)D.)E.)(2006).)Entrepreneurial)processes)and)the)social)construction)of) opportunity.)Entrepreneurship)and)Regional)Development,)18(5),)421C440.) Gartner,)W.)B.)(1985).)A)conceptual)framework)for)describing)the)phenomenon)of) new)venture)creation.)Academy)of)management)review,)10(4),)696C706.)Goldstein,)J.)(1999).)Emergence)as)a)construct:)History)and)issues.)Emergence,)1(1),) 49C72.) Goldstein,)J.,)Hazy,)J.)K.,)&)Silberstang,)J.)(2010).)A)complexity)science)model)of)social) innovation)in)social)enterprise.)Journal)of)social)entrepreneurship,)1(1),)101C125.) Honig,)B.,)&)Samuelsson,)M.)(2012).)Planning)and)the)entrepreneur:)a)longitudinal) examination)of)nascent)entrepreneurs)in)Sweden.)Journal)of)Small)Business) Management,)50(3),)365C388.) Hopp,)C.,)&)Sonderegger,)R.)(2014).)Understanding)the)Dynamics)of)Nascent) Entrepreneurship—PrestartCUp)Experience,)Intentions,)and)Entrepreneurial) Success.)Journal)of)Small)Business)Management.) Katz,)J.,)&)Gartner,)W.)B.)(1988).)Properties)of)emerging)organizations.)Academy)of) management)review,)13(3),)429C441.) Levie)and)Lichtenstein)(2010))A)Terminal)Assessment)of)Stages)Theory)C) Introducing)a)Dynamic)States)Approach)to)Entrepreneurship) Liao.)J.)J.,)&)Welsch.)H.)(2008).)Patterns)of)venture)gestation)process)C)Exploring)the) differences)between)tech)and)nonCtech)nascent)entrepreneurs.)The)Journal)of) High)Technology)Management)Research.)19(2).)103C113) Lichtenstein.)B.)B.,)Dooley.)K.)J.,)&)Lumpkin.)G.)T.)(2006).)Measuring)emergence)in) the)dynamics)of)new)venture)creation.)Journal)of)Business)Venturing.)21(2).)153C 175.) Lichtenstein.)B.)B.,)Carter.)N.)M.,)Dooley.)K.)J.,)&)Gartner.)W.)B.)(2007).)Complexity) dynamics)of)nascent)entrepreneurship.)Journal)of)Business)Venturing.)22(2).) 236C261.) Lichtenstein,)B.)B.)(2009).)Moving)Far)from)FarCfromCEquilibrium:)Opportunity) Tension)as)the)Catalyst)of)Emergence*.)Emergence:)Complexity)and)Organization,) 11(4),)15.) Lichtenstein,)B.)B.,)&)McKelvey,)B.)(2011).)Four)types)of)emergence:)a)typology)of) complexity)and)its)implications)for)a)science)of)management.)International) Journal)of)Complexity)in)Leadership)and)Management,)1(4),)339C378.) Lichtenstein,)B.B.)‘Levels)and)degrees)of)emergence:)toward)a)matrix)of)complexity) in)entrepreneurship’,)Int.)J.)Complexity)in)Leadership)and)Management,)Vol.)X,) No.)Y,)pp.000–000.) Lounsbury,)M.,)&)Glynn,)M.)A.)(2001).)Cultural)entrepreneurship:)Stories,)legitimacy,) and)the)acquisition)of)resources.)Strategic)management)journal,)22(6-7),)545C 564.) MacKinnon,)D.)P.)(2008).)Introduction)to)statistical)mediation)analysis.)Routledge.) Mathias,)B.)D.,)Williams,)D.)W.,)&)Smith,)A.)R.)(2015).)Entrepreneurial)inception:)The) role)of)imprinting)in)entrepreneurial)action.)Journal)of)Business)Venturing,)30(1),) 11C28.) McKelvey,)B.)(2004).)Toward)a)complexity)science)of)entrepreneurship.)Journal)of) Business)Venturing,)19(3),)313C341.) McMullen,)J.)S.,)&)Dimov,)D.)(2013).)Time)and)the)entrepreneurial)journey:)the) problems)and)promise)of)studying)entrepreneurship)as)a)process.)Journal)of) Management)Studies,)50(8),)1481C1512.) Menzies,)T.)V.,)Diochon,)M.,)Gasse,)Y.,)&)Elgie,)S.)(2006).)A)longitudinal)study)of)the) characteristics,)business)creation)process)and)outcome)differences)of)Canadian)
