MSc Thesis

(1)

MSc Thesis

Empirical analysis of resource constraints and uncertainty as antecedents for

effectuation in the field of firm creation with the moderating effect of

entrepreneurial expertise

“ How do resource constraints and uncertainty influence the use of

effectuation and causation in the creation of a new firm and does

entrepreneurial expertise moderate these relationships?”

by

LEONARD LOHMEYER

S2412985

Fongersplaats 56

9725 LC Groningen

L.C.F.Lohmeyer@student.rug.nl

University of Groningen

Faculty of Economics and Business

January 2018

Supervisor: Prof. Dr. Andreas Rauch

Co-assessor: Dr. Evelien Croonen

(2)

Abstract

This study studied the effect of uncertainty and resource constraints on the use of the

entrepreneurial logic effectuation over causation with an interaction effect of entrepreneurial expertise. Data from German entrepreneur (N=196) was gathered via an E-Mail questionnaire and analyzed with a multiple hierarchical regression analysis. The results showed that

uncertainty had a positive effect on the use of effectuation and a negative effect on causation, thereby supporting my hypotheses. No evidence for the effect of resource constraints on the logic used was found, and neither was there evidence for an interaction effect with

entrepreneurial expertise found.

(3)

Introduction

Many scholars have tried to describe the behavior of humans under conditions of uncertainty. For example, utilitarianism of Bentham and Mill (1789) describes that people when faced with a situation act to maximize their own personal pleasure and minimize their pain. But even before that, expected utility in the theory of choice under uncertainty can be traced back to Pascal (1672) who introduced the concept of maximizing expected value when faced with a choice. Daniel Kahneman (1974) recognized that human beings use heuristics, or rules of thumb, to make sense of the environment and make decisions in the absence of absolute certainty. Bentham and Mill’s (1789) theory can be described as heuristics, through which humans judge circumstances and make their decisions. Though they simplified the heuristics used to maximizing pleasure and minimizing pain.

Knight (1921) concluded that there are two types of uncertainty, one, which is now known as risk, and one that is known as uncertainty. Risk relates to a future, in which one can assign probabilities to events that will occur. A future that is characterized by uncertainty is not easily predictable by assigning probabilities to possible events. Kahneman (1974) argues that people use a set of heuristics to make sense of such a future by comparing it to past experiences. Here, theories of rational choice fall short. While creating a business, the entrepreneur faces high uncertainty as well as high resource constraints (Sarasvathy, 2001). Sarasvathy (2001) introduced the logic of effectuation and contrasts it with the logic of causation, which stresses thorough analyses and is mostly taught in business schools. Effectuation can be described as a set of heuristics precisely fitted to the context of creating new economic artifacts. The difference between effectuation and causation is that “causation processes take a particular effect as given and focus on selecting between means to create that effect. Effectuation processes take a set of means as given and focus on selecting between possible effects that can be created with that set of means“ (Sarasvathy, 2001, p: 245).

(4)

Sarasvathy (2001) makes a significant contribution by questioning the effectiveness of predominant causal logic in the creation of new economic artifacts. Moreover, she does a great job articulating the elements of effectuation as a model of how expert entrepreneurs make decisions while building economic artifacts. However, uncertainty, as well as resource constraints, has been assumed in the scenarios of current studies (e.g. Sarasvathy, 2009; Dew, Read, et al., 2008). Sarasvathy (2009) assumes in her expert study that entrepreneurs face uncertainty and resource constraints while creating economic artifacts. While she has merely shown that expert entrepreneurs use effectuation more than nascent entrepreneurs, she has not studied, but simply assumed the existence of uncertainty or resource constraints in her experiments. However, she claims, just as Kahneman and Tversky (1979) that humans use heuristics in uncertain situations by ignoring parts of the information to make effective decisions. But, thorough analyses, for example, a part of causation, might help the entrepreneur cope with uncertainty, too. Chandler, De Tienne, McKelvie and Mumford (2011) found significant evidence for the effect of uncertainty on one sub-dimension of effectuation and Harms and Schiele (2012) found no significant effect of uncertainty on effectuation as an aggregated construct.

However, Sarasvathy argues that the expert entrepreneur facing uncertainty with rather limited resources considers effectuation, however, to my knowledge, there is no study that investigated the effects of resource constraints on effectuation, neither has there been found a significant effect of uncertainty on the aggregated variable of effectuation. Moreover, is there no paper that tried to measure the moderating effect of entrepreneurial expertise on the uncertainty-effectuation and the resource constraints-effectuation relationships. This paper aims to test key antecedents of effectuation- and causation-based decision-making in the creation of new ventures. This research contributes to the effectuation literature by being one of the first papers to empirically test the concepts of effectuation and causation in a field study among entrepreneurs, thereby allowing this paper to investigate the antecedents of effectuation and causation conceptualized by Sarasvathy (2001). Moreover, this paper shows that entrepreneurs use the effectual heuristics more and causal relationships less under conditions of uncertainty; however, there has not been significant results for resource constraints being a predictor of either logic, neither was this research able to confirm the effect of entrepreneurial expertise on either way of creating new ventures.

(5)

entrepreneurs and non-entrepreneurs think aloud while solving scenarios, in which the goal is to start a new venture. However, studies that focus on the effect of uncertainty on effectuation and causation are limited to two, whereas one has found no significant results (Harms & Schiele, 2012) and the other has found results only for one sub-dimension (Chandler et. al., 2011). In addition, there is no study to my knowledge that has investigated the effect of resource constraints on effectuation and causation and neither is there a study that investigated how entrepreneurial expertise affects these relationships. Therefore, this study sheds light on the moderating effect of entrepreneurial expertise on the relationship between uncertainty and the use of effectuation as well as the relationship between resource constraints and the use of effectuation.

Research Question

“How do uncertainty and resource constraints influence the use of effectuation and causation in the situation of new firm creation and how are these relationships moderated by

(6)

Literature Review

Choice Under Uncertainty and Resource Constraints

The field of choice under uncertainty begins with a distinction between risk and

uncertainty. Knight (1921) discusses two types of uncertainty: (1) a future with a known distribution and an unknown draw, which he defined as risk and (2) a future with an unknown distribution and an unknown draw, which he defined as uncertainty. Under risky

circumstances, one can thus attach probabilities to each event occurring and figure out with simple mathematics, which choice is the best. Choice under risk with the assumptions of rationality and perfect information has been widely studied by scholars such as Pascal (1672), Bentham and Mill (1789) and MacCrimmon and Stanbury (1988). For example, the rational choice theory argues, that the rational actor, the homo oeconomicus, chooses between several distinct options by calculating the expected return and choosing for the alternative that maximizes one’s utility (MacCrimmon and Stanbury, 1988).

In 1959, Herbert Simon, a cognition scientist, hypothesized that humans are limited in their judgment due to bounded rationality, which he described as having limited information, time constraints and cognitive limitations. This was the starting point of the research on heuristics (Simon, 1959). Kahneman and Tversky (1979) criticized the return-maximization model of rational choice for over-simplifying human nature, as it assumes that the actor has complete information and can adequately assign value to each alternative. They further developed the model of rational choice by introducing the bounds of human rationality, which argues that humans have cognitive limitations and use heuristics, or rules of thumbs, and inductive logics to make decisions. In psychology, heuristics are efficient and simple rules that people use to make judgments (Lewis, 2012). Lewis (2012) describes heuristics as mental shortcuts, focusing on one part of a complex situation while ignoring others.

(7)

require less information and computation than complex and elaborate analyses (Gigerenzer et. al., 1999).

Even though heuristics are natural ways of making decisions, the accuracy varies (Hammond, Hamm, Grassia & Pearson, 1987). Hammond et. al. (1987) found out that decisions made by experts are mostly adequate, whereas novices’ decisions based on heuristics are not. Generally, experts solve problems faster and with fewer mistakes (Erricson & Smith, 1991). However, it cannot be stated that either the deep analysis with sophisticated techniques or heuristically-based decision are better or worse than the other, even if the decision-maker uses the most salient rules, substantial parts of information are ignored (Carsrud & Brännback, 2009). Heuristics are domain-specific, designed for precise tasks and can be learned (Todd & Gigerenzer, 2003).

Effectuation is a set of heuristics used by expert entrepreneurs in the uncertain situations of economic artifact creation such as firms, markets and economies (Arend, Sarooghi, & Burkemper, 2015). This framework stems from an extensive and qualitative research by Sarasvathy (2001) in the area of entrepreneurship and stresses a non-predictive logic to tackle uncertain issues and stakeholder commitment. In her expert study, she analyzed the making of expert entrepreneurs and compared it with the decision-making of novice entrepreneurs. Sarasvathy (2008) was able to identify principles from the qualitative data analysis that were present in and preferred by expert entrepreneurs. The five heuristics of effectuation, and which she found in expert entrepreneurs are (1) the

bird-in-hand principle: start with your means: who you are, what you know, and who you know, (2) the-affordable-loss principle: concentrate on the affordable loss rather than expected return,

(3) the crazy-quilt principle: form partnerships with people and organizations that are willing to commit to jointly create the future (4) the lemonade principle: remain flexible to exploit surprising opportunities, and (5) the pilot-in-the-plane principles: focus on creating rather than predicting the future.

(8)

either effectual or causal logic depending on the situation. So causation and effectuation are not substitutable types of logics in the creation of new firms, but complementary ones.

In contrast to the effectual logic described above, causal logic is characterized by the following principles: 1) gather the means to create the effects to achieve the pre-determined goal, 2) choose the alternative with the highest expected return, 3) stress competitor analysis to find your place in the market 4) exploit preexisting knowledge instead of exploiting contingencies, 5) gather data to plan long term and predict the future. Sarasvathy argues that causal reasoning falls short in the creation of new firms, because the access to resources is constrained and the effects that are created by these means as well as the final goal are uncertain. Examples for causal reasoning in practice are thorough strategic planning, market research, competitor analysis or writing a business plan (Sarasvathy, 2009). Honig and Karlsson (2004) showed that institutional coercion and mimetic forces lead to a higher probability to use causation in nascent entrepreneurs.

Sarasvathy (2008: 26) found out that uncertainty about the future is a personal belief. Downey, Hellriegel and Slocum (1977) have also argued this way. They state “uncertainty should be treated as a perceptual quality” (p. 570). They argue “uncertainty is based upon the attributes of the environment (…) and the characteristics of those individuals’ conceptual processes (p. 568). This means that each individual perceives uncertainty differently, which opens up the question whether there is a significant difference of the perception of uncertainty between expert entrepreneurs and nascent entrepreneurs (which were the two groups under study in Sarasvathy’s (2008) research) or whether the principles are learned by experience in situations of uncertainty (Todd & Gigerenzer, 2003). Uncertainty is a major problem in the building of new economic artifacts, as it can prohibit the entrepreneur to predict, whether his or her business will be successful. If the circumstances the entrepreneur faces are not predictable, it is considered as uncertain, and if the circumstances are predictable, it is considered risky (Sarasvathy, 2001). Sarasvathy argues that the common model of entrepreneurship, causation, does not work under uncertainty, as the entrepreneur cannot predict the effects of accumulated means.

(9)

relationships. It does not necessarily mean that there is uncertainty about the external environment, but rather uncertainty about the effects that are created by causes. Response uncertainty is uncertainty about what the utility of each alternative might be, by looking at the expected returns. Connecting these types of uncertainty to effectuation, it becomes clear that especially the uncertainty or certainty of cause-effect relationships are drivers of the use of either effectuation or causation, because Sarasvathy argues that expert entrepreneurs start with the means and try what effects they can create with it, implying that the effects of these means are unknown. However, she argues, nascent entrepreneurs that use more causation logic start off with the effects they want to create and gather the means to create these effects, implying that they are certain about the effects that certain means can create.

McKelvie, Haynie and Gustavssion (2011) have operationalized effect uncertainty and investigated the relationship to entrepreneurial action. They also stress the perceptual quality of effect uncertainty as opposed to the more objective measurements of uncertainty such as environmental dynamism. Further, they discovered that expertise moderates the relationship between effect uncertainty and the willingness to pursue an action, which might be because more experienced entrepreneurs downplay the importance of predicting the future, which is also what Sarasvathy (2001) argues. However, another possible explanation is that experienced entrepreneurs are overconfident and therefore overestimate their abilities, which is a widely discussed phenomenon in the entrepreneurship literature (e.g. Zhang & Cueto, 2015; Salamouris, 2013). However, as uncertainty has been measured in their article as a perceived quality, it seems that also the more experienced entrepreneurs have perceived high degrees of uncertainty, and recognized the inability to forecast in highly uncertain environments.

(10)

of the new venture and co-create this venture with the partners by leveraging own and partners’ resources without risking more than one is willing to lose (Sarasvathy, 2001).

Another theory that has acknowledged resource constraints in new and entrepreneurial companies is the behavioral theory of ”entrepreneurial bricolage”. It is a way how entrepreneurs “make do” with what they have at hand (Levi-Strauss, 1967). However, there is no empirical evidence that resource constraints cause the application of bricolage (Perry et. al., 2012). Following the logic of Downey, Hellriegel and Slocum (1977), that uncertainty should be treated as a perceptual quality only, this paper argues that resource constraints should be treated as a perceptual quality, because “it is based upon the attributes of the environment (…) and the characteristics of those individuals’ conceptual processes” (p. 568). Dolmans, van Burg, Reyman and Romme (2014) have also argued for the perceptual quality of resource constraints, because reflects the entrepreneur’s lack of available resources relative to demand. Therefore, if the entrepreneur perceives he or she is constrained in accessing the resource he or she needs to create the desired effects, he or she will use effectuation rather than causation.

Resources are important to start a business, and if the entrepreneur is constrained in his or her access to resources the effectuation processes can be a way to overcome this constraint. This paper adopts Draft’s (1983) definition of resources as “all assets, capabilities, organizational processes, firm attributes, information, knowledge etc. controlled by a firm that enable the firm to conceive and implement strategies that improve its efficiency”, but this paper will, as Sarasvathy argues include the resources that are at the immediate disposal through the entrepreneur’s and firm’s network, because it is not the control over the resources, but the access to them, which is important (e.g. Sarasvathy, 2001, Light, 1984; Zimmer and Aldrich, 1987; Bates, 1997). This is in line with the first principle of effectuation “Bird in the hands”, according to which the entrepreneur starts with the evaluation of who I am, what I

know, and whom I know, so he or she starts by assessing the resources at his or her immediate

(11)

that expert entrepreneurs start off with their means at hand, with who they are, what they know and whom they know. So these entrepreneurs start with the resources that are available to them within their immediate disposal, on which there are constraints. However, these resource do not only entail the entrepreneur’s own resources (who they are and what they know), but also his or her network’s resources (whom they know). So, the entrepreneur faces constraints on the access to resources. Sarasvathy (2001) developed the principles of effectuation from the elements in her expert-study, and argues that these principles lead to effective decisions in uncertain times while being constrained in resources by leveraging partnerships without risking more than one can lose and co-creating the future.

Therefore, I set up the following hypotheses:

H1a: The higher the entrepreneur’s perceived uncertainty the more likely he/she will use effectuation.

H1b: The higher the entrepreneur’s perceived uncertainty the less likely he/she will use causation.

H2a: The higher the entrepreneur’s perceived resource constraints the more likely he/she will use effectuation.

H2b: The higher the entrepreneur’s perceived resource constraints the less likely he/she will use causation.

Entrepreneurial Expertise

As Sarasvathy has shown expert entrepreneurs use more effectuation, because expertise in creating new economic artifacts helps to develop heuristics specifically for this domain; however, they use both forms of logic, dependent on the context. It seems that due to their expertise, expert entrepreneurs can judge their surroundings well and use heuristics they learned over the course of their career to build new artifacts. Tegtmeier and Meyer (2018); however, found no evidence for the impact of expertise on the use of effectuation among German entrepreneurs.

Sarasvathy (2008) defined her expert entrepreneur as a person who, either individually

or as part of a team, had founded one or more companies, remained a full-time founder/entrepreneur for 10 years or more, and participated in taking at least one company public (21). So, Sarasvathy has shown that expertise in entrepreneurship affected the use of

(12)

expert entrepreneurs either perceive the same situation differently than nascent entrepreneurs, or they make different decisions, even though they perceive it as nascent entrepreneurs do. The question whether or not expert entrepreneurs perceive the same environment differently is not within the scope of this study. However, the question how entrepreneurs make decisions based on their perception of the environment and how this is influenced by entrepreneurial expertise is indeed within the scope of this study.

Additionally, resource constraints as argued above should also be seen as a perceptual quality only. Therefore, the same question arises, whether expert entrepreneurs perceive the same situation differently than nascent entrepreneurs, which is again, outside the scope of this study. However, how the entrepreneur acts in face of resource constraints and how this is moderated by entrepreneurial expertise is within this study. Sarasvathy (2001) has argued that the heuristics of effectuation are best suited for the beginning situations of creating a firm, in which uncertainty and resource constraints are high. By applying these heuristics, the entrepreneur can ignore certain information, leverage partnerships and co-create the future, by not risking more than one can lose (Sarasvathy, 2009).

According to Kahneman and Tversky (1974) heuristics that help people make effective decisions can be learned. This implies that people having experienced similar situations learn the heuristics to make effective decisions. As Sarasvathy (2001) found out, causal decision-making and processes are effective in situations that are characterized by risk, not by uncertainty. For decision making under uncertainty, humans use heuristics or rules of thumb to make quick estimations by ignoring part of the information. Effectuation, which was observed by Sarasvathy, can be seen as a set of heuristics for situations where the entrepreneur creates new economic artifacts (Sarasvathy, 2001; Read et. al. 2009). Therefore, this paper argues that when entrepreneurs perceive the situation of building a firm as uncertain they will use more effectuation than causation, which is amplified by the level of expertise the entrepreneur has. Additionally, this paper argues also that when an entrepreneur perceives the situation of creating a new firm as having resource constraints they will use more effectuation than causation, which is amplified by the level of entrepreneurial expertise.

Therefore I set up the following hypotheses:

(13)

H3b: Entrepreneurial expertise moderates the relationship of perceived uncertainty and the use of causation, as such: the higher the entrepreneurial expertise, the greater the negative effect of perceived uncertainty on the use of causation

H4a: Entrepreneurial expertise moderates the relationship of perceived resource constraints and the use of effectuation, as such: the higher the entrepreneurial expertise, the greater the positive effect of perceived resource constraints on the use of effectuation

(14)

Conceptual Models

(15)

Methodology

Sample

This research is about the effects of uncertainty and resource constraints, moderated by entrepreneurial expertise, on effectuation and causation, which is about how people make decisions while they create new economic artifacts. Sarasvathy (2001) states that the creation of firms, markets and economies are examples of new economic artifacts. For the sake of practicality, I decided to investigate the creation of new firms as did other studies about effectuation (e.g. Read et. al., 2008). Based on my knowledge of the current literature, there is yet to come a study on effectuation about the creation of another form of economic artifact than firms. Therefore, the sample includes only recently started firms (incorporated on 2014 and after). The data is from the Orbis database, which is a database with globally comparable data. The sample only includes German companies. In the beginning of the survey, there were two questions, that decided whether the person answering this questionnaire was in an adequate position to do so: 1) Have you, alone or with others, started, or are you currently

starting a new business, including any self-employment or selling any goods or services to others? and 2) Have you, alone or with others, started or are your currently starting a new business or a new venture for your employer, an effort that is part of your normal work? The

questionnaire ended when neither question was answered with yes and preceded if either or both questions were answered with yes. The search criteria in the Orbis database yielded 45,935 companies, from which I contacted 24,512 random companies by e-mail and received 297 answers on my questionnaire, from which 196 were completed. This resulted in a response rate of 0,8%.

(16)

Measurements

Effectuation and Causation

(17)

Table 1 Component Matrix Component 1 2 Caus1 ,689 ,125 Caus2 ,811 ,168 Caus4 ,742 ,147 Caus5 ,745 ,033 Caus7 ,591 ,056 Caus9 ,540 ,287 Eff1 -,251 ,713 Eff2 -,377 ,602 Eff5 -,001 ,708 Eff6 -,154 ,587 Eff8 -,099 ,477 Table 3

Reliability Statistics Effectuaiton

Cronbach's Alpha N of Items

,652 5

Table 4

Reliability Statistics Causation

Cronbach's Alpha N of Items

,789 6

Extraction Method: Principal Component Analysis.a

Table 2

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. ,758

Bartlett's Test of Sphericity

Approx. Chi-Square 489,646

df 55

(18)

Independent Variables Uncertainty

McKelvie et. al. (2011) have established a measurement for effect uncertainty theorized by Miliken (1987), which entails both the predictability of demand and the predictability of technological change. The authors used a conjoint analysis to capture the behavior under uncertainty. Here, the participant has to make decisions in scenarios that differ based on uncertainty attributes. These uncertainty attributes are state uncertainty, effect uncertainty and response uncertainty all with a value for high and for low and each scenario has different configurations of each attribute. This study will make use of only effect uncertainty measures as described above in the theory section. As the questions should ask about decisions and situations from the past, I have added the historical element “At the time you were starting your company…” Also, I have converted the high and low values of McKelvie et al. (2011) to a 5-point Likert-scale with opposing statements. The opposing statements for the 5-point Likert-scale for predictability of demand change are “At the time you were starting your company, you had a strong idea of your customers’ preferences and demands with regard to your product or service, and these are predictable over time.” and “At the time you were starting your company, it was not possible to predict in advance demand changes affecting the viability of the product or service.”. For predictability of technological change I used these opposing statements: “At the time you were starting your company, your were in a strong position to predict the nature and source of innovations that affect the viability of the product or service.” and “At the time you were starting your company, it was not possible to predict with any certainty the kinds or timing of future technological innovations that will affect the viability of the product or service.”. The alterations to the questions might interfere with the scale reliability; however, precise items have been valued more in this research.

(19)

Resource Constraints

In this study, I use a subjective measure of the entrepreneur’s level of satisfaction with his or her access to resources and not an objective measure such as total funds available, or others, because as described above, resource constraints should be treated as a perceptional quality only, because it “is based upon the attributes of the environment (…) and the characteristics of those individuals’ conceptual processes” (Downey, Hellriegel and Slocum, 1977, p. 568). Dolmans et. al, (2014) have studied the effect of perceived resource positions on performance and creativity, and operationalized resource slack and resource constraints as the two extremes of attainable perceived resource position. This reflects perceived resource availability, resulting from the set of actual or potential resources at one’s disposal, relative to the perceived resource demand. Dolmans et. al. (2014) have divided resource position into three dimensions: financial resource, entailing cash or other financial means, capacity resource, entailing operational or production capacity and capability resources, entailing human resources and know-how. I developed the following dimensions from Dolmans et. al. (2014) stressing the perceptual and relative value, as well as the access beyond personally owned resources, of resource constraints: “At the time you were starting your company, how satisfied were you with your access to cash and other financial means compared to what your business needed?” I employed a five-point Likert-scale from 1 “extremely satisfied” to 5 “ not at all satisfied”. I repeated the question structure and filled in the other dimensions of resource position by Dolman et. al. (2014) creating a 3-items scale.

However, the scale reliability test Cronbach’s Alpha showed insignificant results that were below the threshold of 0,6 (Cohen et. al. 2003). It is possible that the Cronbach’s Alpha shows such low results, because this measure is dependent on the number of items in the scale (Cohen et. al., 2003). However, this led me to the inclusion of all items as independent variables in the regression analysis rather than the composite variables of those items and to the adaption of a hypothesis for each item.

Entrepreneurial Expertise

(20)

Expertise due to low frequency, as only 2 entrepreneurs of the sample (n=196) have taken a company public, and low factor loadings (<0,4) (Cohen et. al. 2003)

Control Variables

(21)

Results

The mean of causation is around 3,4 with a standard deviation of 0,84. The mean of Effectuation is 3,38 with a standard deviation of 0,83. Most of the entrepreneurs (51%) from this sample had less than 1 year of experience being the entrepreneur of another company before starting their current business, and 15,8% had more than 11 years of experience being the entrepreneur of a business before having started their current business. Only 2 entrepreneurs have taken a company to the public stock exchange, which is why this item has been dropped from the analysis (see table 5). Overall the correlations (See table 6, Appendix 1) between the independent variables are relatively modest ranging up until 0,311. Interesting to note is that there is not significant negative correlation between causation and effectuation (p>0,05). Moreover, the correlation table shows that capacity constraints and effectuation correlate positively r(194) = 0,147, p<0,05 and capacity constraints and causation correlate negatively r(194) = -0,175, p<0,05. No other independent variable concerned with resource constraints correlates with either dependent variable significantly (p>0,05). Both demand uncertainty and technological uncertainty correlate with effectuation positively r(194) = 0,412, p<0,01 and r(194) = 0,214, p<0,01, respectively. Also, both correlate negatively with causation positively r(194) = -0,246 , p<0,01 and r(194) = -0,199 , p<0,01, respectively.

(22)

Hypotheses Tests

I tested the hypotheses via a hierarchical regression analysis. I mean-centered the variables so the threat of multicollinearity in equations where we included interaction terms was minimized (Aiken & West, 2003). However, to ensure that multicollinearity is not a problem, I mean centered the variables and applied a multicollinearity diagnosis. The results of this analysis can be found in table 9 and table 12 in the appendices and it shows that the variance inflation factor are between 1,016 and 1,287, which is far below the critical value of 4,0 (Hair, Black, Babin, Anderson and Tatham, 1998). Hierarchical linear regression was used to test the models, because this type of analysis is suitable when independent variables correlate. In addition, this style of regression is used to remove the effect of potential confounding variables, included here in the control variables, as well as testing main- and interaction effects (Cohen, et. al., 2003). I started off with adding the control variables and then the independent variables, before I added the interaction effects with entrepreneurial expertise. I split the analysis into different blocks, to control for confounding effects (Cohen et. al., 2003).

Effectuation

The results show that neither of the control variables shows a significant coefficient as you can see under model 1 (see table 7 in Appendix 2). When adding the moderator variable Experience and the independent variables of Demand Uncertainty, Technological Uncertainty, Cash Constraint, Capabilities Constraints and Capacity Constraints model 2 showed significance in predicting the dependent variable Effectuation (F10, 182 = 4,529,

p<0,01). Moreover, model 2 shows a significant (p<0,01) F-Change value of 7,348 (see table

7 in Appendix 2). The addition of the interaction effects of Entrepreneurial Expertise with the independent variables in model 3 resulted in insignificant results and showed no explanatory power (p>0,05). Also the F-Change remained insignificant. Overall, model 2 fits best and was able to explain 15,2% of the observed variance in the sample (see table 7, Appendix 2).

(23)

I able to find evidence that Entrepreneurial Expertise influenced the relationships between any of the predictors with the use of Effectuation; therefore, I have to reject H3a(Demand), H3a(Technological), H4a(Cash), H4a(Capabilities) and H4a(Capacity).

Causation

The results of my second regression analysis show that neither of the control variables is able to significantly predict the outcome of the dependent variable Causation (p>0,05) as it can be seen under model 1 (see table 10). In model 2 the moderator variable Experience and independent variables Demand Uncertainty, Technological Uncertainty, Cash Constraints, Capability Constraints and Capacity Constraints were added. Model 2 was able to predict

5,9% of the variance of the dependent variable Causation (F11,181 = 2,003, p<0,05) (see table

10 and 11, Appendices 5 and 6). Moreover, the F-change value for model 2 was 3,535 and significant (p<0,05). The inclusion of the interaction effects of Entrepreneurial Expertise with the independent variables in model 3 resulted in insignificant results and showed no explanatory power (p>0,05). Overall, model 2 shows the only significant results.

(24)

Discussion

The goal of this study was to answer the question what effect resource constraints and uncertainty have on the use of effectuation over causation and how this is moderated by entrepreneurial expertise. As hypothesized, I found out that higher uncertainty leads to the use of effectuation over causation; however, I did not find significant effects of resource constraints on the use of effectuation, nor did I find significant moderating effects of entrepreneurial expertise.

Sarasvathy (2001) did a great job articulating the concepts of effectuation; however, empirical proof for the predictors of effectuation was missing in the current academic stream of literature. From the theory of Sarasvathy (2001) I conceptualized the two conditions that predict the use of effectuation and operationalized them in this study. Moreover, I tried to incorporate the level of entrepreneurial expertise in this model, which was stated explicitly as the main driver of effectuation by Sarasvathy (2001). I built new measures for resource constraints and uncertainty and was able to find significant evidence for demand- and technological uncertainty being a predictor for effectuation, which means that entrepreneurs use this framework when they face demand- and technological uncertainty. When the demand or technological change is not predictable, entrepreneurs use rules of thumb to hedge against this uncertainty. Having established prove for the predictors of effectuation brings this framework closer to the establishment as a theory (Perry et. al., 2012).

(25)

increased use of effectuation and whether it is learned or comes naturally due to the tendency to hedge against uncertainties is up to future research.

This research was unable to conclude that entrepreneurs use the principles of effectuation when they are faced with resource constraints as it was expected. The principles of effectuation, especially the “affordable loss” and the “forming of partnerships” are conceptually related to working with resource constraints; however, due to inadequate factor loadings, the “forming of partnerships” item that measured the principle accordingly was dismissed from the variable effectuation, which might have influenced the relationship between all resource constraints variables and effectuation.

Interesting was the missing correlation between effectuation and causation. I assumed that there was a negative correlation between both types of logic, as Sarasvathy has conceptualized them as opposing models. Moreover, Chandler et. al. (2009) have operationalized the measures also as opposing measures. However, as Sarasvathy (2001) argues, entrepreneurs can use both types of logics depending on the circumstances. And as the circumstances are ever changing (Milliken, 1987) cross-sectional data might be inadequate for both, measures of effectuation and causation and for measures of the circumstance such as uncertainty or resource constraints.

Moreover, it is likely to assume that entrepreneurs of rather small and less ambitious businesses answered my questionnaire. Ambitious is in this case conceptually closely related to the difference between where the company is and where the company wants to go, which can be directly seen back in the measures of constraints, which measures the current access to resources the business’ need for resources. The mean and standard deviations are leaning towards high satisfaction with the entrepreneurs’ access to resources (mean: 3,05; 2,44; 2,33 SD: 1,267; 0,944 and 1,016, for each measure of resource constraints, respectively, 1 being extremely satisfied with the access and 5 being extremely dissatisfied). Less ambitious companies are more resource constrained, which also entails constrained in the CEO’s time, which he or she would be unable to spend on my questionnaire.

Practically, my findings help to understand entrepreneurial behavior. For example, it helps to understand observable behavior of detailed planning, which likely implies that the entrepreneur is certain of demand and technology. On the other hand, only risking what one can lose, not planning and establishing partnerships to co-create the future might be a sign that an entrepreneur is uncertain about demand or technological changes in their market space.

(26)

Limitations and Future Research

This study has its major drawbacks in the operationalization of the independent variables. Due to the limited number of items it was hard to establish reliable measures for both independent variables and the moderator variable. However, the limited number is justified with the limited resources this study had at its disposal, because the dependent variables already took up 17 items in total. This implies that due to time concerns of potential respondents, the number of items had to be cut down for the dependent variables which lead to less reliable measures.

Moreover, the factor loadings for both logic models effectuation and causation were not as accurate as hoped, as I had to delete several items from both scales due to inefficient- or cross-loadings. However, especially for the effectuation measure this was foreseeable, because it is normally a multidimensional measure, but for the sake of this analysis I aggregated this measure into one, sacrificing deleting several items and more detailed results. As described above, the principle of gathering partnerships was not adequately loaded with the right factor, therefore, the effect under research were most likely affected, because forming relationships is one of the main principles that an entrepreneur uses to hedge against resource constraints or, more directly, gain access to resource.

(27)

References

Aken, J. E. van; Berends, J. J.; Bij, J. D. van der. Problem Solving in Organizations : A

Methodological Handbook for Business and Management Students, 2nd editie.; Cambridge

University Press: Cambridge, 2012.

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological bulletin, 103(3), 411.

Arend, R. J., Sarooghi, H., & Burkemper, A. (2015). Effectuation as ineffectual? Applying the 3E theory-assessment framework to a proposed new theory of entrepreneurship. Academy of Management Review, 40(4), 630-651.

Bagozzi, R. P. (1984). Expectancy-value attitude models an analysis of critical measurement issues. International Journal of Research in Marketing, 1(4), 295-310.

Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the academy of marketing science, 16(1), 74-94.

Bainbridge, L., & Sanderson, P. (1995). Verbal protocol analysis (Vol. 7). ch.

Benhabib, J., & Spiegel, M. M. (1994). The role of human capital in economic development evidence from aggregate cross-country data. Journal of Monetary economics, 34(2), 143-173. Carsrud, A. L., & Brännback, M. (Eds.). (2009). Understanding the entrepreneurial mind:

Opening the black box (Vol. 24). Springer Science & Business Media.

Chandler, G. N., DeTienne, D. R., McKelvie, A., & Mumford, T. V. (2011). Causation and effectuation processes: A validation study. Journal of business venturing, 26(3), 375-390. Chandler, G. N., DeTienne, D. R., & Mumford, T. V. (2007). Causation and effectuation: Measurement development and validation. Frontiers of entrepreneurship research, 27(13), 3. Charness, N. (1989). Expertise in chess and bridge. Complex information processing: The

impact of Herbert A. Simon, 183-208.

Cohen, J. 1978. Partial products are interactions; partial powers are curve components.

Psychological Bulletin 70: 259–267

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2013). Applied multiple regression/correlation analysis for the behavioral sciences. Routledge.

Davidsson, P., & Honig, B. (2003). The role of social and human capital among nascent entrepreneurs. Journal of business venturing, 18(3), 301-331.

Downey, H. K., Hellriegel, D., & Slocum Jr, J. W. (1977). Individual characteristics as sources of perceived uncertainty variability. Human Relations, 30(2), 161-174.

(28)

Ericsson, K. A., & Smith, J. (Eds.). (1991). Toward a general theory of expertise: Prospects

and limits. Cambridge University Press.

Fox, M. C., Ericsson, K. A., & Best, R. (2011). Do procedures for verbal reporting of thinking have to be reactive? A meta-analysis and recommendations for best reporting methods. Psychological bulletin, 137(2), 316.

Gigerenzer, G., & Gaissmaier, W. (2011). Heuristic decision making. Annual review of

psychology, 62, 451-482.

Gigrenzer, G., Hell, W. and Blank, H. (1988). Presentation and content: the use of base rates as a continuous variable. Journal of Experimental Psychology: Humam Perception and

Performance. 14: 513-525

Goldstein, D. G., & Gigerenzer, G. (2002). Models of ecological rationality: the recognition heuristic. Psychological review, 109(1), 75.

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (1998). Multivariate data analysis (Vol. 5, No. 3, pp. 207-219). Upper Saddle River, NJ: Prentice hall.

Hammond, K. R., Hamm, R. M., Grassia, J., & Pearson, T. (1987). Direct comparison of the efficacy of intuitive and analytical cognition in expert judgment. IEEE Transactions on

systems, man, and cybernetics, 17(5), 753-770.

Honig, B., & Karlsson, T. (2004). Institutional forces and the written business plan. Journal of Management, 30(1), 29-48.

Kahneman, D. and Tversky, A. 1979: Prospect theory: an analysis of decision under risk,

Econometrica, 47: 263-91.

Kaiser, H. F., & Rice, J. (1974). Little jiffy, mark IV. Educational and psychological

measurement, 34(1), 111-117.

Lawrence E. Blume and David Easley (2008). "rationality," The New Palgrave Dictionary of Economics , 2nd Edition. Abstract." by Abstract] & pre-publication copy.

Lewis, A. (Ed.). (2012). The Cambridge handbook of psychology and economic behaviour. Cambridge University Press.

MacCrimmon, K.r., Wehrung, D.A. and Stanbury, W.T. 1988. Taking Risks: The management of uncertainty. Administrative Science Quarterly. 33 (4)

McKelvie, A., Haynie, J. M., & Gustavsson, V. (2011). Unpacking the uncertainty construct: Implications for entrepreneurial action. Journal of Business Venturing, 26(3), 273-292.

Mezias, S. J. (1990). An institutional model of organizational practice: Financial reporting at the Fortune 200. Administrative science quarterly, 431-457.

(29)

Montgomery, H. E., & Svenson, O. E. (1989). Process and structure in human decision

making. John Wiley & Sons.

Neal, D. (1995). Industry-specific human capital: Evidence from displaced workers. Journal of labor Economics, 13(4), 653-677.

O’brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41(5), 673-690.

Pascal, B. ([1672] 1958):Pascal’s Pensées. E. P. Dutton and Co., New York.

Perry, J. T., Chandler, G. N., & Markova, G. (2012). Entrepreneurial effectuation: a review and suggestions for future research. Entrepreneurship Theory and Practice, 36(4), 837-861. Rangone, A. (1999). A resource-based approach to strategy analysis in small-medium sized enterprises. Small Business Economics, 12(3), 233-248.

Read, S., Song, M., & Smit, W. (2009). A meta-analytic review of effectuation and venture performance. Journal of Business Venturing, 24(6), 573-587.

S.X. Zhang; J. Cueto (2015). "The Study of Bias in Entrepreneurship". Entrepreneurship

Theory and Practice. 41 (3): 419–454

Sáenz, J. (2005). Human capital indicators, business performance and market-to-book ratio. Journal of Intellectual Capital, 6(3), 374-384.

Salamouris, I. S. (2013). How overconfidence influences entrepreneurship. Journal of

Innovation and Entrepreneurship, 2(1), 8.

Shane, S., & Venkataraman, S. (2000). The Promise of Entrepreneurship as a Field of Research. The Academy of Management Review, 25(1), 217-226. Retrieved from http://www.jstor.org/stable/259271

Tegtmeier, S., & Meyer, V. (2018). Experts of thoroughness and fanatics of planning? Daring insights into the decision-making of German entrepreneurs. International Journal of

Entrepreneurship and Small Business, 33(1), 132-157.

Von Otter, C., & Stenberg, S. Å. (2015). Social capital, human capital and parent–child relation quality: interacting for children’s educational achievement?. British Journal of

Sociology of Education, 36(7), 996-1016.

(30)

Appendix 1: Correlations

Table 6: Correlations Correlations Age Gender Le ve l o f Education Experience Parents En tre pr en eu rs Demand Uncertainty Te ch no lo gi ca l Uncertainty Cash Constr aints Capabilities Constr aints Capacity Constr aints Causation Ef fectuation Age

Pearson Correlation Sig. (2-tailed) N

Gender Pear son Cor re lation Sig. (2-tailed) N Level of Education

Experience

Parents Entrepreneurs

Demand Uncertainty

Te ch no lo gi ca l Uncertainty

Cash Constr aints Pear son Cor re lation Sig. (2-tailed) N Capabilities Constraints

Capacity Constraints

Causation

Ef

fectuation

(31)

31

Appendix 2: ANOVA; Dependent Variable: Effectuation

Appendix 3: Regression Summary; Dependent Variable: Effectuation

Table8:

Model Summary

Table 7: ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression ,687 4 ,172 ,250 ,910b Residual 129,284 188 ,688 Total 129,970 192 2 Regression 25,897 10 2,590 4,529 ,000c Residual 104,073 182 ,572 Total 129,970 192 3 Regression 26,684 15 1,779 3,048 ,000d Residual 103,286 177 ,584 Total 129,970 192

a. Dependent Variable: Effectuation

Model Summary

Model R R Square Adjusted R Square Std. Error of the Estimate

Change Statistics R Square

Change F Change df1 df2 Sig. F Change 1 2 3 ,073a ,005 -,016 ,82926 ,005 ,250 4 188 ,910 ,446b ,199 ,155 ,75619 ,194 7,348 6 182 ,000 ,453c ,205 ,138 ,76390 ,006 ,270 5 177 ,929

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education a.

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education, Cash Constraints, Technological Uncertainty, Capacity Constraints, Experience, Capabilities Constraints, Demand Uncertainty

b.

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education, Cash Constraints, Technological Uncertainty, Capacity Constraints, Experience, Capabilities Constraints, Demand Uncertainty, Demand Uncertainty x Experience, Cash Constraints x Experience, Technological Uncertainty x Experience, Capacity Constraints x

(32)

32

Appendix 3: Regression Coefficients, Dependent Variable: Effectuation

Table 9:

Coefficients Coefficientsa

Model

Unstandardized Coefficients Standardized Coefficients

t Sig.

Collinearity Statistics

B Std. Error Beta Tolerance VIF

1 (Constant) Age Gender Level of Education Parents Entrepreneurs 2 (Constant) Age Gender Level of Education Parents Entrepreneurs Experience Demand Uncertainty Technological Uncertainty Cash Constraints Capability Constraints Capacity Constraints 3 (Constant) Age Gender Level of Education Parents Entrepreneurs Experience Demand Uncertainty Technological Uncertainty Cash Constraints Capability Constraints Capacity Constraints Demand Uncertainty x Experience Technological Uncertainty x Experience Cash Constraints x Experience Capability Constraints x Experience Capacity Constraints x Experience 3,160 ,443 7,130 ,000 -,011 ,059 -,014 -,183 ,855 ,969 1,032 -,051 ,182 -,021 -,280 ,780 ,952 1,050 -,040 ,046 -,064 -,866 ,387 ,967 1,034 -,046 ,139 -,024 -,331 ,741 ,984 1,016 1,866 ,485 3,849 ,000 -,009 ,057 -,012 -,166 ,868 ,872 1,147 ,046 ,173 ,019 ,269 ,788 ,880 1,137 -,052 ,043 -,083 -1,203 ,231 ,933 1,072 -,005 ,129 -,002 -,037 ,971 ,954 1,048 ,033 ,025 ,095 1,353 ,178 ,884 1,131 ,227 ,048 ,348 4,683 ,000 ,795 1,258 ,084 ,048 ,128 1,729 ,086 ,806 1,241 ,033 ,047 ,051 ,700 ,485 ,841 1,189 ,039 ,062 ,045 ,620 ,536 ,854 1,170 ,032 ,057 ,040 ,560 ,576 ,881 1,135 1,795 ,503 3,568 ,000 -,010 ,058 -,013 -,180 ,857 ,861 1,162 ,037 ,175 ,015 ,209 ,835 ,871 1,148 -,047 ,044 -,075 -1,070 ,286 ,903 1,107 ,015 ,133 ,008 ,112 ,911 ,909 1,101 ,033 ,025 ,093 1,293 ,198 ,870 1,150 ,219 ,050 ,335 4,413 ,000 ,777 1,287 ,085 ,049 ,130 1,725 ,086 ,787 1,270 ,037 ,048 ,057 ,772 ,441 ,834 1,199 ,045 ,064 ,052 ,708 ,480 ,827 1,210 ,038 ,058 ,046 ,648 ,518 ,873 1,146 ,012 ,061 ,014 ,203 ,839 ,895 1,117 -,053 ,062 -,062 -,863 ,389 ,867 1,153 -,017 ,059 -,022 -,294 ,769 ,829 1,207 ,014 ,063 ,017 ,227 ,821 ,786 1,272 -,033 ,057 -,042 -,578 ,564 ,857 1,166

(33)

33

Appendix 5: ANOVA, Dependent Variable: Causation

Appendix 6: Regression Summary, Dependent Variable: Causation

Table 11:

Model Summary

Table 10: ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression ,553 4 ,138 ,189 ,944b Residual 137,363 188 ,731 Total 137,916 192 2 Regression 14,889 10 1,489 2,203 ,019c Residual 123,028 182 ,676 Total 137,916 192 3 Regression 17,240 15 1,149 1,686 ,057d Residual 120,676 177 ,682 Total 137,916 192 Model Summary Model R R Square Adjusted R

Square Std. Error of the Estimate

Change Statistics R Square

Change F Change df1 df2 Sig. F Change 1 2 3 ,063a ,004 -,017 ,85478 ,004 ,189 4 188 ,944 ,329b ,108 ,059 ,82218 ,104 3,535 6 182 ,002 ,354c ,125 ,051 ,82570 ,017 ,690 5 177 ,632

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education a.

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education, Cash Constraints, Technological Uncertainty, Capacity Constraints, Experience, Capabilities Constraints, Demand Uncertainty

b.

Predictors: (Constant), Parents Entrepreneurs, Gender, Age, Level of Education, Cash Constraints, Technological Uncertainty, Capacity Constraints, Experience, Capabilities Constraints, Demand Uncertainty, Demand Uncertainty x Experience, Cash Constraints x Experience, Technological Uncertainty x Experience, Capacity Constraints x

(34)

34

Appendix 7: Regression Coefficients, Dependent Variable: Causation

Table 12: Coefficients Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics

B Std. Error Beta Tolerance VIF

1 (Constant) Age Gender Level of Education Parents Entrepreneurs 2 (Constant) Age Gender Level of Education Parents Entrepreneurs Experience Demand Uncertainty Technological Uncertainty Cash Constraints Capability Constraints Capacity Constraints 3 (Constant) Age Gender Level of Education Parents Entrepreneurs Experience Demand Uncertainty Technological Uncertainty Cash Constraints Capability Constraints Capacity Constraints Demand Uncertainty x Experience Technological Uncertainty x Experience Cash Constraints x Experience Capability Constraints x Experience Capacity Constraints x Experience 3,023 ,457 6,619 ,000 ,034 ,061 ,041 ,560 ,576 ,969 1,032 ,130 ,188 ,052 ,693 ,489 ,952 1,050 ,013 ,048 ,019 ,262 ,794 ,967 1,034 ,019 ,143 ,010 ,133 ,895 ,984 1,016 3,654 ,527 6,933 ,000 ,000 ,062 ,000 ,002 ,999 ,872 1,147 ,157 ,188 ,062 ,837 ,404 ,880 1,137 ,023 ,047 ,035 ,481 ,631 ,933 1,072 ,009 ,140 ,005 ,063 ,950 ,954 1,048 ,041 ,027 ,113 1,512 ,132 ,884 1,131 -,112 ,053 -,167 -2,131 ,034 ,795 1,258 -,093 ,053 -,138 -1,775 ,078 ,806 1,241 -,008 ,051 -,013 -,166 ,869 ,841 1,189 ,044 ,068 ,049 ,648 ,518 ,854 1,170 -,098 ,062 -,117 -1,573 ,118 ,881 1,135 3,763 ,544 6,921 ,000 ,005 ,062 ,006 ,077 ,938 ,861 1,162 ,155 ,190 ,062 ,819 ,414 ,871 1,148 ,013 ,048 ,020 ,265 ,791 ,903 1,107 -,008 ,144 -,004 -,053 ,957 ,909 1,101 ,043 ,027 ,119 1,577 ,117 ,870 1,150 -,103 ,054 -,154 -1,926 ,056 ,777 1,287 -,105 ,053 -,155 -1,957 ,052 ,787 1,270 -,017 ,052 -,025 -,329 ,742 ,834 1,199 ,047 ,069 ,052 ,674 ,501 ,827 1,210 -,107 ,063 -,128 -1,702 ,091 ,873 1,146 -,039 ,066 -,044 -,590 ,556 ,895 1,117 ,024 ,067 ,027 ,363 ,717 ,867 1,153 ,048 ,064 ,058 ,745 ,457 ,829 1,207 ,026 ,068 ,030 ,384 ,702 ,786 1,272 ,073 ,062 ,089 1,175 ,242 ,857 1,166 Dependent Variable: Causation

(35)

Appendix 8: Effectuation, inverse logic of Causation

Figure 2:

Effectuation, inverse logic of Causation

(36)

Appendix 9: Figures and Tables

Figure 1: Conceptual Model

Figure 2: Effectuation, inverse logic of Causation Table 1: Factor analysis

Table 2: KMO

Table 3: Coefficient Alpha Effectuation Table 4: Coefficient Alpha Causation Table 5: Descriptives

Table 6: Correlations

Table 7: ANOVA: Effectuation

Table 8: Model Summary Effectuation

Table 9: Regression Coefficients Effectuation Table 10: ANOVA: Causation

Table 11: Model Summary Causation