Decision making of entrepreneurs : the effect of effectual and causal reasoning on business performance

(1)

DECISION MAKING OF ENTREPRENEURS:

THE EFFECT OF EFFECTUAL AND CAUSAL

REASONING ON BUSINESS PERFORMANCE

Author

Student number

Date of submission

Version

Course

Track

Institution

Supervisor

Erik Gyurity

11097663

24-06-2016

Final

MSc Business Administration:

Entrepreneurship & Innovation

University of Amsterdam

(2)

1

Statement of originality

This document is written by Student Erik Gyurity 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.

(3)

2

LIST OF FIGURES AND TABLES

Figure 1: Sarasvathy (2008): Dynamic process of effectual interactions ... 14

Figure 2: Conceptual Model ... 22

Table 1: Skewness and Kurtosis ... 29

Table 2: Cronbach’s alpha ... 29

Table 3: Correlation matrix ... 30

Table 4: ANOVA ... 31

Table 5: Regression Model ... 31

Table 6: ANOVA (Ad-hoc analysis) ... 33

(5)

4

Abstract

Central focus of this thesis is the relationship between the certain decision making logics (effectuation and causation) of entrepreneurs and business performance of their startups. Sarasvathys ‘effectuation and causation’ concept served as a theoretical baseline at this research, where the certain principles of decision making patterns were examined in regards of business performance. Current study is relevant and important, because it contributes to the existing effectuation literature with further empirical testing and enhances the practicability of entrepreneurial decision making principles. The research investigated 102 startups (mostly

Dutch and Hungarian) via web-questionnaire and the quantitative statistical analyses revealed

thatthere is no significant relationship between the preferred type of decision making styles and business performance of startups. Moreover, the study did not find any significant

differences in scale means for effectuation and causation, which indicates that entrepreneurs could also identify themselves with the ‘causational’ thinking. These finding imply that entrepreneurial success can be achieved with both causation (managerial) and effectuation (entrepreneurial) mindset and the low overall effect of these on performance indicates that there are other factors too that influence success.

Key words: entrepreneurship, startup, decision making, effectuation, causation, entrepreneurial success, business performance

(6)

5

1. Introduction

Being an entrepreneur and starting a business from scratch has always been an enticing career choice, but it especially became a huge trend with the emergence of the IT-technology and the internet, because it enables to create and widespread solutions simply by applying knowledge. Almost half of the companies that are ranked today as ‘most powerful brands’, (like Google or

Facebook), just appeared 10-20 years ago and were developed from brilliant ideas of college

students (Forbes, 2013). Even huge incumbents react on this trend as they recognize the disruption potential and power of startups and dedicate special attention and resources to them. A successful startup can generate tremendously high returns, no wonder that they are emerging in an extraordinary pace today, however failure rate is extremely high among them too (Neishem, 2000). So why is one startup successful opposed to other? Entrepreneurship literature has been looking for the answer first in the personality trait and skills of entrepreneurs (Gartner, 1990). Later the question was approached from a ‘process’ point of view, where the focus was on the entrepreneurs’ behavior and actions (Gartner, 1988). Some others shifted the focus from the entrepreneurs’ nature to the venture and the external environment where they operate. By studying entrepreneurial expertise, a new major stream evolved in the entrepreneurship literature, namely the effectuation theory (Sarasvathy, 1998). This theory focuses on how entrepreneurs make their decisions, how they craft their strategy and what logic do they use at their business decisions (Ucbasaran, 2008). Effectuation is associated with a flexible, ‘entrepreneurial’ mindset, where the entrepreneur experiments with alternatives, focuses on partnerships, affords potential losses and instead of predicting the future, he/she tries to control it (Chandler et al., 2011). Causation on the other hand is associated with traditional ‘managerial’ thinking, where expected returns, business planning and competitive analysis is used to predict an uncertain future (Chandler et al., 2011). Numerous studies were conducted

(7)

6

in this topic and confirmed entrepreneurial behaviors driven by effectual logic, but only few examined whether it directly leads to entrepreneurial success. Does to way entrepreneurs make their decisions effect performance? If effectuation is linked to entrepreneurial behavior than is it likely that startups using effectuation will be more successful?

Research goal and research question

The goal of this research is to contribute to the effectuation literature by examining the relationship between the certain decision making types (effectuation and causation) of entrepreneurs and business performance of their startups. Therefore, the research question is stated as following:

RQ: What is the effect of the decision making logic of entrepreneurs on business performance of their startups?

In more detail, this study examines the effect of the eight different sub constructs of effectuation and causation (will be explained later) to business performance. This research applies quantitative data collection via web-questionnaire targeting mostly Dutch and Hungarian startups with validated measures about decision making and self-reported, perceived business performance.

This thesis is structured in the following way: the next section consists of a brief overview about entrepreneurship, an extensive literature review about entrepreneurial decision making and presentation of effectuation-causation principles. Furthermore, the current state of literature is discussed along with the conceptual model of the research. The third section presents the data collection and research method applied to investigate the research question. After this, the result of the data analysis and hypothesis testing is presented. This will be followed by the discussion of the results, where the research question is answered and reasoned. Finally, conclusion will be drawn from the main findings, further the limitations of the research is presented with a proposal for future research in the topic.

(8)

7

2. Literature review

2.1 Entrepreneurship

An ‘entrepreneurial revolution’ has spread throughout the world and entrepreneurship has never received such attention of academics and scholars as today (Kuratko et al., 2015). Entrepreneurial activity is the essence of modern economies, as it enhances competition, creates jobs and value for customer, drives innovation and contributes to a societal economic well-being (Morris et al., 2015). The role of entrepreneurship in the economy was first recognized by Richard Cantillon (c. 1680-1734), who defined the entrepreneur as ‘someone who engages in exchange for profit; specifically, he exercises business judgements in the face of uncertainty’ (Hebert & Link, 1989, p42). Cantillon emphasized the function of the entrepreneur who is present in different occupations in areas such as like production, distribution and exchange. Later in the first half of the 20th century, Schumpeter focused more on the entrepreneur as a person, as he/she is the ‘persona causa of economic development’ (Hebert & Link, 1989, p43). In his view, the entrepreneur is an innovator in a dynamic economy and creates ‘new combinations’ (Bruyat & Julien, 2000). Further, Ted Schultz contributed to the entrepreneurship literature by extending the notion of entrepreneurship to non-market activities (house-hold decisions, allocation of time etc.) as well (Hebert, 1989), but later Israel Kirzner at the second half of the 20th century defines the essence of entrepreneurship as alertness to profit opportunity (Kirzner, 1997). It is clear that entrepreneurship is not a new occurrence, as the term is in use for almost three hundred years, but it is still an emerging field with various different definitions and theories. In fact, the number of studies about entrepreneurship has expanded exponentially in the past three decades (Kuratko, 2015). However, there is no single definition for entrepreneurship, there is a growing consensus among scholars, that entrepreneurship is ‘a field of business that seeks to understand how opportunities to create something new (e.g., new products or services, new markets, new production processes or raw

(9)

8

materials, new ways of organizing existing technologies) arise and are discovered or created by specific persons, who then use various means to exploit or develop them, this producing a wide range of effects’ (Shane and Venkataraman, 2000, p. 218).

2.2 Entrepreneurial decision making

The economic literature suggests three dominant streams of market process theories, which also had an influence on how the role of entrepreneurship is considered (Sarasvathy et al., 2002). The first stream, the neo-classical economics emphasizes the ‘Equilibrium theory’, which was driven by central assumptions about the market. The theory states that current prices transmit all relevant information to decide how to allocate resources (Hayek, 1945) and all information and expectations about the future can be derived from current price bids for resources. Since everyone owns the same information, decision making evolves into a ‘simple’ mathematical optimization process and all decisions are optimizing decisions (Casson, 1982).

But the second stream of economics, the ‘Austrian model’ neglects the equilibrium theory (Shane & Venkataraman, 2000) and states that prices do not convey all information about the market and cannot reflect the changes immediately. That allows a certain ‘disequilibrium’ situation, where someone has different information about the value of resources, which enables profit opportunities to be exploited by entrepreneurs (Eckhardt & Shane, 2003). Meanwhile

optimizing decisions try to maximize scarce resources across previously developed means and

ends, entrepreneurial decisions are creative and create or identify new means and ends, which were undetected or unutilized (Kirzner, 1997; Gaglio & Katz, 2001).

The third stream in economic theory describes the market as a creative process where the market is not responsible for optimizing, but the actors have the ability to create ends (Sarasvathy et. al., 2002a). This view originates from the ‘Knightian’ view of entrepreneurship. Knight introduced 3 types of uncertainties regarding the future: 1. Risk, where the distribution of future is known and can be dealt with analysis; 2. Uncertainty, where the distribution of the future is

(10)

9

unknown; and 3. ‘Knightian’ uncertainty, which says that the distribution of the future is not knowable (Sarasvathy & Kotha, 2001). The latter construct was studied by Sarasvathy (1998), where she analyzed the decision making of expert entrepreneurs in ‘Knightian uncertainty’. The results proved that expert entrepreneurs do not try to predict the future market, but they played a role to shape and control it (Sarasvathy, 1998). This was the first empirical evidence for a certain entrepreneurial decision making logic which overcomes the ‘Knightian uncertainty’ problem, and is called effectual logic which will be elaborated later in this paper (Sarasvathy, 2001).

2.3 Causation and Effectuation

Considering the development in entrepreneurship research in the past few decades till the appearance of the notion of ‘effectuation’ (Sarasvathy, 1998), it is clear that major part of literature applies the ‘neoclassical’ rational decision making models as a basis (Perry et al, 2012). This means that researchers assumed a goal-driven behavior for entrepreneurs when seeking for opportunities (Bird, 1989) and that they can be discovered through a purposeful search process (Drucker, 1998). Sarasvathy (2001) provides a well-known example for this with the ‘STP-model’ of Kotler, which is a base-model in marketing and is a standard element in management education (Sarasvathy, 2001). STP stands for segmentation, targeting and positioning, which suggests a certain pattern for bringing a product/service to market: analyzing long run opportunities, researching and selecting target markets, positioning the offering, planning and executing the product/service implementation (Kotler, 1991). This example of the STP model proves that goal-driven mentality became a central element in decision making models taught in many business schools and is referred as ‘causation model’ by Sarasvathy (2001).

(11)

10

Sarasvathy provides following definition for causation and effectuation (Sarasvathy, 2001, p245):

‘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.’

In order to better understand the underlying differences between the two processes described above, Sarasvathy introduced some contrasting metaphors (Chandler et al., 2011).

The simplest way to distinguish between causation and effectuation is to compare it to the process of cooking. A chef is assigned to cook dinner. He/she has two ways to approach the task. The causation way would be, when he/she picks out a meal in advance, checks and shops ingredients, and cooks it according to the recipe. Here it begins with a given goal (the chosen food to prepare) and selects between effective ways of preparing it (Sarasvathy, 2001). On the other hand, the effectuation way of cooking dinner would start by checking given ingredients and utensils at hand and imagining possible menus and then preparing the meal (Sarasvathy, 2001).

Further illustrative metaphor about jigsaw puzzle and a patchwork quilt was provided to contrast the differences of how the two approaches see the world (Sarasvathy, 2008). The causation approach (jigsaw puzzle) sees the world as one in which all of the pieces are there and must be assembled. The effectuation approach (patchwork quilt) develops opportunities by experimenting and changing directions and sees the world still in-the-making with an important role of human actions (Chandler et al., 2011).

Causation processes always focus on a given goal and try to exploit existing profitable opportunities, by predicting, analyzing and planning (Read et al., 2009; Alsos et al., 2014). Although the existing literature assumed a goal-driven attitude for entrepreneurs when pursuing

(12)

11

opportunities (Bird, 1989), the study of Sarasvathy (2001) showed that expert entrepreneurs do not follow the causation pattern of business schools. Experienced entrepreneurs used the practical principle of effectuation and do not spend too much time with planning and analysis, as their primal aspiration is creating a new venture and they adapt or change direction as different opportunities arise (Chandler et al., 2011).

Therefore, the appearance of the effectuation theory was a paradigmatic shift in the entrepreneurship literature and gained exceptional attention, as it questions the universal applicability of causation-processes and provides a new way to understand entrepreneurship (Perry et al., 2012; Fischer, 2012).

Causation-driven decision making always starts with a given goal to be achieved and acquires the needful set of means. Then takes the constraints on possible means in consideration and selects the alternative with the maximal return. Causation processes are best applicable when someone has the intention to choose the best, fastest, most effective or economical way to accomplish a desired effect. Therefore, ‘causal’ processes focus on the question: ‘What should we do?’. Effectuation-driven decision making on the other hand consists of the given means, set of effects and the constraints and criteria (affordable loss) for selecting them (Sarasvathy, 2001). Decision makers with effectuation in mind start with the question ‘What can we do?’ and have to imagine more possible ends and select the best by leveraging contingencies. (Ucbasaran, 2008) Causation processes always depend on the ‘effect’, the chosen goal and have great ability to exploit knowledge and deal with environment where clear laws and rules exist (exogenous market). Adversely, effectuation processes are always actor dependent and are truly effective at exploiting contingencies (Sarasvathy, 2003). Therefore, effectuation is more prevalent in an environment, which is dependent on human action (endogenous), as they are not easy to analyze or predict. Thus, the effectuation-driven process is more effective in uncertain and unpredictable situations (Sarasvathy, 2001).

(13)

12 2.3.1 Principles of Effectuation – Causation

To better understand the differences of causal and effectual decision making and to better capture the core of the theory, Sarasvathy (2001) introduced four contrasting principles of effectuation.

1. Means driven vs. goal driven

The first principle is related to the starting point of causation and effectuation processes. Causation-driven decision making always starts with a given goal or effect to be achieved and the necessary means (resources) are obtained accordingly (Sarasvathy, 2003). Conversely, ‘effectuators’ always begin with the given means, which implies that their starting point is what they have at their current disposal. First they asses three categories of ‘means’: Who I am, What I know and Whom I know (Sarasvathy, 2001). This indicates that they gauge their own traits, abilities, expertise, education, skills, knowledge and their social network (Dew et al., 2009). Every combination of given means can establish different outcomes/ends. Effectuators starting point is their own given means as they assume it is easier to control the available resources, than trying to gather and influence ones which they do not own. Sarasvathy (2008) described this phenomenon as the ‘bird in the hand principle’, which says that ‘a bird in the hand is worth two in the bush.’ This says that it is better to have ‘one bird which is in your hand’ (under your control), than two in the bush that can just fly away (not controllable). The example described earlier about cooking dinner is also very illustrative at this principle: either start with the given ingredients at hand or start with a proposed menu.

2. Affordable loss vs. expected return

The second principle relates to the manner of how different entrepreneurs craft their strategies. Causation models choose their strategies by focusing on maximizing the potential returns. Effectuators on the other hand predetermine the amount they are willing and can afford to lose and experiment with as many strategies as possible within the acceptable level of risk (Sarasvathy, 2001). This enhances creative solutions and new combinations, as they only invest

(14)

13

what they can afford to lose and want to maximize returns. Furthermore, when only investing the affordable resources, level of uncertainty and risk can be reduced and the future can be better controlled (Sarasvathy, 2008; Read et al., 2009).

3. Partnerships vs. Competitive market analysis

When it comes to the relation and attitude towards outsiders in a business, causation and effectuation mindsets represent again a different manner. Causation driven processes consider other players on the same market as competitors and rivals. Thus, they use detailed competitive market analysis and strategic planning, which can help to identify the level of risk and returns (Sarasvathy, 2001). Partnerships and collaborations are rare and allow limited influence. They are only present if they serve to achieve the predetermined goal (Dew et al., 2009). But effectuation processes approach this question differently: they focus on partnerships, collaborations and pre-commitments with stakeholders (Sarasvathy, 2001). They do not want to predict the future and calculate returns, but they make effectual partnerships with pre-commitments in order to control and shape it (Sarasvathy, 2008). In that way, effectuators do not want to compete in an existing market, but (co-)create new and controllable ones with strategic alliances and pre-commitments (Chandler et al., 2011).

4. Leveraging the unexpected vs. avoiding contingencies

The fourth principle is related to the attitude towards unexpected events. Causation processes always try to avoid contingencies and situations where unexpected appearances can happen, as they already have a given goal in mind and the detailed action plan for achieving it (Dew et al., 2009). In these unexpected situation the initial plan and target can fail and also the expected returns can alter. Conversely, effectuation processes do not try to avoid the unexpected, but they try to leverage the appeared contingencies. Sarasvathy illustrates this notion with the so called ‘lemonade principle’, which uses the famous quote: ‘when life gives you lemons, make lemonade’ (Sarasvathy, 2008). This means that effectuators do not differentiate between

(15)

14

positive or negative contingencies (Dew et al., 2009). Even if something unexpected happens that completely changes the status quo, they do not try to stick to the initial idea, but they change directions and try to take advantage of the new situation and create an opportunity out of it (Ucbasaran, 2008). As effectuators do not start with a given goal and a related plan, they are comfortable with contingencies as their initial idea is loosely defined and broad, so this helps them to imagine new possible end and solutions (Chandler et al., 2011).

To better understand the above listed principles, Sarasvathy (2008) visualized the dynamic process of effectual interactions (Figure 1). Starting point of the effectual process is the given means (Who I am, What I know, Whom I know), where the effectuator assesses his/her own identity, knowledge and social network and then imagines many possible effects (‘What can I do?’). After this, the effectuators interact with other people and stakeholders in order to achieve commitments based on affordable loss, and further co-develop the project. This involvement of stakeholders and partnerships have two effects: first it increases the set of ‘given means’, as the partners bring their own human capital (identity, knowledge, social network) together for the common project. Second, the involvement and commitment of partners and the increased set of resources now allow to better specify and constrain the possible effects (Dew et al., 2008) (Sarasvathy, 2008).

(16)

15 2.3.2 Further studies on Effectuation

Although, a couple of scholars expressed some critics about the few number of attempts to empirically test the widely known effectuation-theory (Perry et al., 2012; Arend et al., 2015), there are still many which are worth to mention. The study of Sarasvathy (1998 & 2001) is considered as a baseline of effectuation literature, which was followed by number of related studies (Dew et al., 2008). Sarasvathy (1998) examined expert entrepreneurs’ behavior and decision making and identified special patterns that inspired the effectuation model. Based on this, she elaborated the model explicitly in her well-known study of 2001 (Sarasvathy, 2001). Gustavsson (2004) and Wiltbank et al. (2006) took the initiative to further research in the topic of entrepreneurial expertise: they found that there is a significant difference in the way business problems and challenges are approached by expert entrepreneurs and control groups. Similarly, Dew, Read, Sarasvathy and Wiltbank (2009) researched the differences between expert entrepreneurs and novices (MBA students) and confirmed that while experts tend to use effectual logic, novices (inexperienced) tend to apply causation models. The theory of effectuation and the appearance of effectual logic in new venture creation was also tested with several case studies using historical data (Sarasvathy & Kotha, 2001; Sarasvathy, 2008). Based on the studies of entrepreneurial expertise, Dew et al. (2008) delineates a model of entrepreneurial firm behavior, which concentrates rather on leveraging contingencies and creating new markets than trying to compete in existing ones (Dew et al., 2008). Read & Sarasvathy (2005) emphasizes that entrepreneurship is rather a form of a skill, ability and mechanism than a personality trait or psychological characteristics. Further research states that effectuation is not an inborn competence, but a learnable and creative in nature, according to Sarasvathy (2008). Also the interrelation of effectual principles with new venture performance was analyzed by Read & Song (2007) and confirmed a significant relationship at ‘Means-driven’ attitude, ‘partnerships’ and ‘leveraging contingency’. Further research of Read et al. (2009) verified positive relationship between an effectual approach to strategy making and new

(17)

16

business performance (Read et al., 2009). Effectuation and causation processes are not only subjects of research in new ventures and entrepreneurship, but also in different areas. Brettel et al. (2012) transfers and expands the notion of effectuation to another discipline and examines effectual/entrepreneurial actions and its impacts on R&D performance. Affordable loss, partnerships, and acknowledging the unexpected were found to have a positive effect on R&D performance (Brettel et al., 2012). Chandler et al. (2011) developed validated measures and investigated if the effectuation and causation constructs are indeed definite. They define causation as unidimensional construct and effectuation as a multidimensional construct (Chandler et al., 2011). Alsos et al. (2014) employed qualitative research on Chandlers (2011) work to test the validity of the measures and found several issues with the scale. With a corrected and validated scale, quantitative research was conducted on startups in Norway. Results showed significant positive correlation between effectuation and uncertainty; and negative correlation between causation and uncertainty, which support the underlying theory of Sarasvathy (Alsos et al., 2014).

2.3.3 Relationship between causation and effectuation

There are some differences in the understanding of the relationship between causation and effectuation in the entrepreneurship literature. Some scholar like Chandler et al. (2011) believe that effectuation and causation are contrary processes and mindsets. Therefore, it assumes that they are not able to accept each other’s logic and none of the contrary principles can be present at the same time (Chandler et al., 2011). However, Sarasvathy (2008) expressed a different viewpoint regarding the relationship of causation and effectuation. She claims that entrepreneurs can easily switch between the two logics and cannot be stated obviously that they are complete opposite (Sarasvathy, 2008). Her initial study (2001) says, that ‘both causation and effectuation are integral parts of human reasoning that can occur simultaneously, overlapping and intertwining over different contexts of decisions and actions’ (Sarasvathy,

(18)

17

2001, p. 245). This idea was also supported by the study of Fischer, where he conducted research on causation and effectuation by analyzing several case studies on new ventures. He found that there were no cases were entrepreneurs associated with causation behavior only applied causation principles, as they also used effectual ones simultaneously (Fischer, 2012). Read and Sarasvathy (2005) provides further interesting implications about the occurrence of causation and effectuation, by affirming that causation logic is more probable to be used when the entrepreneur is less experienced and when the firm is bigger (Read & Sarasvathy, 2005). So assuming a process of transitioning from a small enterprise to getting a larger firm, it is not possible to fully disassociate and contrast the two different logics.

Sarasvathy (2008) examines this question thoroughly by analyzing the starting ages of abiding companies and identifies certain elements of effectual processes at these entrepreneurs. But having a look at later stages of the firm lifecycle, as it develops and gets more complex, it was clear that there happened a shift to elements of casual logics, in order to sustain competitive advantage on the market (Sarasvathy, 2008). Further she claims that there is no clear boarder line when this transition happens and can differ at each case, but generally there are certain situations where the change is inherent. The entrepreneur realizes that the company is getting large and complex and cannot be organized the same way as it was at the early stages, so he/she tries to use more elements of the casual logic, or he/she moves to start a new venture, as he/she is more comfortable with the environment that is required to the effectual logic (Sarasvathy, 2008).

2.4 Entrepreneurship and the measurement of success

After reviewing the entrepreneurship literature about casual and effectual decision making, this paper aims to analyze the connection of certain elements to venture performance. Decisions are made by the entrepreneur in a company and being successful in entrepreneurship does not inherently means success in terms of business figures and targets of the firm. Success in

(19)

18

entrepreneurship is defined as ‘generating an effective firm in the long term’ (Bouchiki, 1993, p561.) and is considered as a dynamic process. Although, Sarasvathy (2008) states that there must be a distinction between the success of the entrepreneur and the success of the firm, majority of studies link the success of the entrepreneur to the success of the venture. Simply, because entrepreneurial success is a dynamic and abstract term, which is hard to measure; and because business performance is mostly the result of the decisions/actions of the entrepreneur (Sarasvathy, 2008). Moreover, every entrepreneur has the aim to establish a sustainable business, which generates profit in order to further realize his/her visions. Even studies that are analyzing entrepreneurship with a trait approach, that means that they are more focused on the personality characteristics of the entrepreneur, use business performance as a metric to measure entrepreneurial success (Sarasvathy, 2004). This thesis also aims to analyze the connection between decision making pattern of entrepreneurs and the business performance of the underlying startup.

First it is important to provide a definition to ‘startup’, which became a ‘buzzword’ nowadays. However, the notion gained remarkable popularity recently, the literature is still lacking of consensus about a common definition. After conducting online research about the term startup, it is clear that it is a frequently used term, but there are disagreements about some certain criteria. These common elements, that define startups are age, growth, profitability and stability. Kolvereid et al. (2006) and Longenecker et al. (2016) defines startups as a new business venture created from scratch by entrepreneurs and it is called ‘de novo entry’. In this definition the

entrepreneur and new business are central elements of the notion and this allows to apply the

definitions and theories of entrepreneurship as theoretical basis of this work.

2.5 Decision making and business performance

Running a business requires making decisions all the time and the certain decisions are setting the direction of the company and that influences the overall performance of the venture. The

(20)

19

formerly discussed effectuation theory can be derived from a study, which analyses entrepreneurial expertise (Sarasvathy, 1998, 2001). According to psychology literature, experts are individuals, who have acquired unique pattern matching and pattern recognition skills and outperform the general population within their discipline (Chase & Simon, 1973; Ericsson & Lehmann, 1996). So assuming than that effectuation is linked to expertise, allows to hypothesize a (positive) connection to venture performance (Read et al., 2009). Read et al. (2009) conducted a quantitative data analysis with historical data to analyze the connection of effectual principles and new venture performance. This study serves mainly as a basis of my hypotheses regarding effectuation principles, because it applies the most extensive dataset of existing studies.

The three components of ‘means’ (What I know, Who I am, Whom I know) where measured separately, by analyzing 60 studies, which included the data of 11 180 firms (Read et al., 2009). Results showed significant positive relationship between all the three sub-items of means and venture performance, which entitles this thesis to hypothesize following:

H1: Means driven practices have positive impact on venture performance.

Brettel el al. (2012) measured the impact of effectuation and causation on R&D performance, which is not equal to the classical venture performance, but still can be relevant. They used a bipolar scale instead of a Likert-scale to contrast the certain element of effectuation and causation. It turned out that effectuation has a positive effect on R&D performance when there is a high level of innovativeness. Read et al. (2009) did not find any relationship between ‘affordable loss’ and venture performance, however Brettel et al. (2012) confirmed a positive relationship to the performance of R&D departments. Also Dew et al. (2009) agrees that ‘affordable loss’ is effective in uncertain environments, so it lets this paper to hypothesize following:

(21)

20

H2: ‘Affordable loss’ driven thinking has positive impact on venture performance.

Read et al. (2009) also analyzed the relationship of partnerships and business performance, using the data of 14 studies including the data of 3196 firms and found a positive, significant relationship. This allows to create the third hypothesis:

H3: Business-practices guided by ‘partnerships’ have positive impact on venture performance.

Further, Read et al. (2009) also found positive significant relationship between the fourth effectuation principle ‘leveraging contingencies’ and new venture performance, by analyzing 5 studies with the data of 712 firms. So the fourth hypothesis would be:

H4: Business-practices guided by ‘leveraging from the unexpected’ thinking have a positive impact on performance.

Further, this paper aims to investigate on the relationship between causation processes and venture performance.

Brinckman et al. (2010) did not take Sarasvathys theory of effectuation as a baseline, but still conducted a research which is worth to mention. They analyze the connection between business planning and business performance. It turned out that business planning had a positive effect on business performance, however different factors moderate the strength of the relationship. Certain contingencies limit the return on business planning, like uncertainty, limited prior information and absence of business planning structures and procedures (Brinckman et al. 2010). Further, Brinckmann et al. (2010) and Wheelwright & Clark (1992) advocates the notion of the ‘planning school’, which implies that having a well-defined goal leads to enhanced output and efficiency, as it guides the process, keeps the determined budget and reduces errors. This allows to hypothesize a positive relationship between planning and business performance:

(22)

21

H5: Goal driven practices have positive impact on venture performance.

Neoclassical investment theory (Campbell, 1992) suggests that maximizing expected return results in high operational performance, but Dew et al. (2009) shows that it is not effective in uncertain environments. And since startups operate mostly in highly uncertain environment, it is not probable that ‘expected return’ mindset will be beneficial. Therefore, the next hypothesis is following:

H6: ‘Expected return’ driven thinking has negative impact on venture performance.

Although ‘Competitive market analysis’ is considered to increase the chance of success on the market (Grinstein, 2008), Grewal and Tansuhaj (2001) states that in uncertain environment the knowledge about market becomes quickly outdated. So in the highly unforeseeable environment where startups operate, competitive market analysis seems to be ineffective. Therefore, next hypothesis states following:

H7: Business-practices guided by ‘competitive market analysis’ have negative impact on venture performance.

When considering the ‘planning school’, avoiding the unexpected seems to be sensible, as it helps to stick to the initial goal (Wheelwright & Clark, 1992), but in uncertain environment it is not effective (Tatikonda and Rosenthal, 2000). It does not allow adaptability and limits the capabilities of the firm to leverage contingencies. Therefore, this thesis hypothesizes following:

H8: Business-practices guided by ‘avoiding the unexpected’ thinking have negative impact on performance.

(23)

22

(24)

23

3. Methodology

The following chapter discusses the research approach and research design of this master thesis. Firstly, the research instrument of the questionnaire is discussed, then the targeted sample and the data collection procedure is presented. After this, the measurements and variables are discussed, which is followed by the statistical method part.

3.1 Research design

The philosophy of the research conducted for this thesis is positivism, as it intends to study observable and measurable variables in the principle of cause and effect. This suggests that a theory can be proposed, tested and refined until it accurately predicts reality (Saunders & Lewis, 2012). Furthermore, deductive approach is applied at this research, as its basis is a grounded theory in the business literature, namely Causation and Effectuation from Sarasvathy (1998,

2001), and the goal is to test hypotheses derived from it (Saunders et al., 2009). The research

technique used in this thesis is questionnaire and is aimed at cross-sectional analysis of the relationship of entrepreneurial decision making and venture performance. The limited timeframe available for this master thesis allows collecting/analyzing data at only one period in time (cross-sectional) (Saunders & Lewis, 2012).

The scales used in the questionnaire are all validated in past literature (Appendix 2). The original scales are in English language and was translated to Hungarian, as almost half of the population was selected from Hungary. The Hungarian scales were translated back to English by a third person in order to make sure that the content and meaning remained the same. Also, a pre-test of the survey was done with an entrepreneur to receive feed-back and fine-tune the final version.

The questionnaire was distributed online and applied multi-channel approaches such as social media (Facebook), e-mail and company websites. In order to maximize response rates and

(25)

24

increase visual appeal to potential respondents, the electronic survey of Qualtrics was used (Qualtrics, 2015).

To gather as many responses as possible, a joint survey with a fellow student was created. The unit of analysis (startups/entrepreneurs) and the dependent variable (business performance) of both constructs are the same, so a common survey enabled to unite our network and reach as many entrepreneurs as possible. The survey is divided up into four blocks: The first part is related to general information about startups, the second is about opportunity recognition (excluded from my research), the third is about decision making processes and the final one is about business performance. Obviously the survey became longer than the average and it took ~10-15 minutes to complete in average. This resulted in a higher than the average drop-out rate (~34.2%), but the large target group, which was possible because of the joint-survey compensated it.

3.2 Sample and Data Collection

As the research targeted startups, which are continuously forming and disappearing, it was impossible to achieve a complete sampling frame. In order to create a representative sample of the population, non-probable self-selective sampling (Saunders & Lewis, 2012) was used and targeted at 1560 firms. The research was conducted at the University of Amsterdam in the Netherlands by a Hungarian student, so Dutch and Hungarian startups were the most approachable and therefore they built the majority of the sample. Besides the two nations, the questionnaire also targeted German, Belgian and English entrepreneurs. After distributing the web-questionnaire to the possible respondents, they could decide whether they are willing to take part in the research (self-selective sampling). This ensured that respondents were not coerced to take part and additionally anonymity was guaranteed to all startups. The joint survey enabled collaboration with another student from different background and network, which ensured a bigger diversity of the respondents, by covering more industries and markets. The larger and more diverse the target group in terms of country,

(26)

25

industry and markets, the more likely differences will become apparent and comparable (Dew et al., 2009).

The survey was distributed online to the startups, so there was a possibility that not the right person with the relevant knowledge about the business will respond. As this paper aims to analyze the relationship of the decision making pattern of entrepreneurs and the business performance of the startup, it was indispensable to have answers from the founder/manager of the company. Several control measures were implemented to reduce possibility of bias here. The letter sent to the companies, the official cover letter and even the introduction section of the survey consisted of a statement to forward the questionnaire to the relevant person if the current recipient is not the founder/manager. Moreover, the first question in the survey was asking, whether the respondent the founder/manager of the company is. This made it possible to filter irrelevant respondents even if they were reluctant to read the instructions and started to fill it out immediately.

3.2.1 Sample Characteristics

The research targeted 1560 firms and received 102 complete responses, which resulted in a 6.54% response rate. The survey targeted startups from different countries, hence the nationality dispersion is the following: 45% of the sample were Dutch, 44% Hungarian, 7% Belgian, 2% UK and 2% German. The industrial composition of the sample is diverse: however, biggest part of the sample (~32%) is operating in the IT industry, there are responses from more than 20 different sectors. 87.3% of respondents were male, the average age was 34 years and their company was operating for 3 and a half years on average. The distribution of how the respondents considered their entrepreneurial expertise was dispersed almost equally (novice 31%- - moderately expert 34% - expert 35%)

3.3 Variables and measures

This research is investigating the relationship between certain decision making processes and venture performance. Sarasvathys (2001) study serves here as a theoretical basis, which was empirically tested and further developed by several scholars. All applied measures and scales are

(27)

26

adopted from original scales of authors from peer-reviewed journals in the field of entrepreneurship and management. The variables are discussed below and the actually used items are listed in Appendix 2.

3.3.1 Independent variables

The literature review section gives a detailed explanation about the different types of decision making procedures that this research handles. According to Sarasvathy there are two major decision making patterns: causation and effectuation, which are further divided in 4-4 sub dimensions: Means driven, affordable loss, partnership, leverage from unexpected, goals driven, expected return, competitive market analysis, avoiding the unexpected (Sarasvathy, 2001; Brettel et al. 2012; Dew et al. 2009; Chandler et al. 2011). The items were adopted from the study of Chandler et al. (2011) and measured with a 5-point Likert scale from strongly agree to strongly disagree. Chandler et al. (2011, p379) describes effectuation and causation constructs as following:

Effectuation: ‘When entrepreneurs use effectuation processes they experiment with

alternatives in which potential losses in the worst-case scenario are affordable, they use pre-commitments and strategic alliances in an attempt to control an unpredictable future, and they remain flexible so they can take advantage of changing environmental contingencies’

Causation: ‘envisioning the end from the beginning, maximizing expected returns, business

planning and competitive analyses to predict an uncertain future, and exploiting pre-existing knowledge.’

The independent variable is perceived as a categorical / nominal variable in the model.

3.3.2 Dependent variable

The dependent variable of the conceptual model is business performance of startups. The research model assumes that there is a connection between how entrepreneurs make their decisions and the success of their business. Entrepreneurship literature usually assesses

(28)

27

business performance by measuring employee and sales growth, furthermore the change in profitability of startups in the first three years of operating (Sambasivan et al., 2009; Blackburn et al., 2013; Haber & Reichel, 2007; Wiklund & Shepherd, 2005; Rauch et al., 2009). The three metrics are measured with seven possible interval-categories which were designed for each measure to offer appropriate choices that can cover the various and wide-ranging performance level of their startups. These measures are all validated and approved by peer reviewed papers: employee-growth is based on Terpstra & Olson (1993), Bruno and Tyebjee (1985) and Kazanjian (1988); the clusters of sales growth is based on Bruno and Tyebjee (1985) and Kazanjian (1988) and the categories of profitability is based on Huggins & Johnston (2009).

3.4 Statistical Procedure

First of all, the raw data of the research had to be prepared and cleaned in order to keep the level of bias as low as possible. The dataset was checked for errors and the incomplete answers were filtered out and excluded from the research list-wise. In order to test the hypotheses, new variables were created from the certain items. Reliability analysis was conducted on these measures to ensure that all of them are consistent. To be sure that all the items in one scale measure the same, or some questions should not be used for analysis, Cronbach’s Alpha test was done. Also, skewness and kurtosis of the data had to be checked in order to test normality. Correlation and Multiple Linear Regression analysis was carried out to test the underlying hypotheses. The statistical software, SPSS (v.22) was used for the analysis. The detailed analysis and presentation of results are elaborated later in the ‘Results’ section.

(29)

28

4. Results

This section is about to present the results of the data analysis, which aims to test the hypotheses of the research. First the reliability analysis will be presented, second the normality test which is followed by the correlations and linear regression. Finally, some additional analyses are presented.

4.1 Preliminary steps, scale means, normality, reliability

The data collection was carried out through web-questionnaire in two languages (English and Hungarian), so the two different datasets had to be merged first. Before starting the analysis, the dataset was cleaned list wise from the incomplete and irrelevant responses, which resulted in 102 relevant responses with a dropout rate of 34,2%.

In order to be able test the hypotheses, new variables were created from the certain items and the scale means were calculated. The first two columns of Table 3, (the correlation matrix) serves as a demonstration of these measures (M=means; SD=standard deviation). Skewness and kurtosis was calculated in order to check normality of the variables (Table 1). As all variables had a kurtosis and skewness value between -1 and 1 (except leveraging from unexpected, where

the difference is negligible, so is accepted in the model) the distribution can be considered as

normal, which enables the regression analysis. For ‘means driven’, ‘affordable loss’, and ‘competitive market analysis’ skewness is between -0,5 and -1, which indicates that the distribution is moderately negative; and kurtosis for ‘partnership’, ‘avoiding the unexpected’ is also between -0,5 and -1, which means that distribution is flatter than the normal.

(30)

29

Table 1: Skewness and Kurtosis

Variables Skewness Kurtosis

Means-driven -0.80 0.21

Affordable loss -0.61 -0.37

Partnership -0.30 -0.82

Leveraging from unexpected 0.09 -1.06

Goal-driven -0.24 -0.44

Expected returns -0.49 -0.30

Competitive market analysis -0.58 -0.32 Avoid the unexpected -0.14 -0.63

Business Performance 0.16 -0.39

To ensure internal consistency of the measures reliability analysis was conducted. Seven out of the total nine scales have a Cronbach’s Alpha value higher than (.70), which indicates that they measure consistently and can be used at the research. The variable, ‘Expected return’ had a rate of (.687), which is very close to the acceptable level (.70), so it is kept in the research. Initially business performance consisted of three items (sales growth, employee growth, profitability), which had a low level of internal consistency (.619), so one item (profitability) was deleted to keep reliability of the measure. After eliminating the risky item, the remaining two items measuring business performance are averaged and transformed into a continuous scale from 1 to 7 in order to result in sufficiently high level of measurement for regression analysis. The exact Cronbach’s Alpha values are demonstrated in Table 2below.

Table 2: Cronbach’s alpha

Variables Cronbach’s alpha

1. Business performance .752

2. Means-driven .830

3. Affordable loss .721

4. Partnership .759

5. Leveraging from unexpected .755

6. Goal-driven .706

7. Expected returns .687

8. Competitive market analysis 1.0 9. Avoid the unexpected 1.0

(31)

30

4.2 Correlations

The correlation matrix (Table 3.) shows the correlation coefficients with the combination of all

variables used in this study. Business performance (dependent variable) does not correlate with any of the independent variables (dimensions of decision making). But besides this, the model shows some relationship between the variables: ‘Avoiding the unexpected’ was the strongest predictor of the principles of causation, namely ‘Expected returns’ (r=0.392; p<0.01), ‘Competitive market analysis’ (r=0.343; p<0.01) and ‘Goal-driven’ attitude (r=0.325; p<0.05). ‘Means-driven’ attitude was the strongest predictor of two effectuation principles: ‘Affordable loss’ (r=0.292; P<0.01) and ‘Partnership’ (r=0.184; P<0.05).

Table 3: Correlation matrix

4.3 Regression analysis

Multiple Regression analysis was conducted to test the eight underlying hypotheses. Linear relationship was tested between the independent variables (means, affordable-loss, partnership,

lev. the unexpected, goal, expected returns, competitive market analysis, avoid the unexpected)

and the dependent variable (business performance). Moreover, the inbuilt analysis of variance, (ANOVA test) is presented, which tests whether the model is significantly good at predicting the outcome.

Means, Standard Deviations, Correlations

Variables M SD 1. 2. 3. 4. 5. 6. 7. 8. 9.

1. Business perform. 4.716 1.356 (.752)

2. Means-driven 3.931 0.937 0.110 (.83)

3. Affordable loss 3.792 0.783 0.126 0.292** (.721)

4. Partnership 3.118 1.111 0.026 0.184* 0.092 (.759)

5. Lev. from unexpected 3.059 1.144 -0.105 0.051 0.106 0.005 (.755)

6. Goal-driven 3.368 0.794 -0.134 0.118 0.192* -0.020 0.120 (.706)

7. Expected returns 3.441 0.926 0.136 0.157 0.189* 0.055 0.203* 0.129 (.687)

8. Comp. mark. analysis 3.500 1.079 0.041 0.126 0.159 0.103 0.209* 0.162 0.813 (1.0)

9. Avoid the unexpected 3.206 1.084 -0.007 0.124 -0.054 0.292** 0.192* 0.325** 0.392** 0.343** (1.0) *Correlation is significant at the level 0.05 (1-tailed)

(32)

31

The results of the regression analysis and the ANOVA table is presented below. (Table 4. &

Table 5.)

Table 4: ANOVA

Table 5: Regression Model

Multiple Regression Model of Business Performance

Variable R R Square B Std. Error Beta t p

.300a _.090

Means-driven .008 .111 .153 .077 .724 .471

Affordable loss .016 .216 .190 .125 1.135 .259

Partnership .000 -.008 .132 -.006 -.060 .952

Leveraging from unexpected .013 -.151 .122 -.127 -1.238 .219

Goal-driven .022 -.285 .187 -.167 -1.525 .131

Expected returns .037 .401 .261 .274 1.537 .128

Competitive market analysis -.007 -.206 .216 -.164 -.954 .343

Avoid the unexpected .000 .025 .155 .020 .161 .873

a. Predictors: Independent Variables b. Dependent Variable: BP_fin

The ANOVA’s table (Table 4.) indicates that the model is not significant with F (8, 93) = 1.154, p= .336, which means that the model is not predicting the dependent variable significantly well (because p>.05).

The result of the multiple regression model implies following findings:

All in all, the model shows a .09 R Square value, which means that it has a 9% proportion of variance in the dependent variable that can be explained by the independent variable. So the certain dimensions of effectuation and causation explain 9% variance in business performance of the startups. ANOVAa Sum of Squares df Mean Square F Sig. Regression 16.768 8 2.096 1.154 .336b Residual 168.987 93 1.817 Total 185.755 101

a. Dependent Variable: Business Performance

b. Predictors: (Constant), Means-driven,Affordable loss, Partnership, Lev. from unexpected, Goal-driven, Expected return, Competitive mark. Analysis, Avoid the unexpected

(33)

32

‘Means driven’ practices and ‘affordable loss’ driven thinking proved to have positive effect on business performance, but the level of it is low and insignificant (p>.05). Therefore:

Hypothesis 1 and 2 are not supported, because they predicted positive relationship between ‘means driven’ attitude & ‘affordable loss’ driven thinking and business performance.

’Partnership’ and ‘Leveraging from unexpected’ have an insignificant (p<.05), low and negative influence on business performance (Beta=-.006; .127), so Hypothesis 3 and 4 are not

supported, because they predicted positive relationship between ’Partnership’ & ‘Leveraging from unexpected’ driven thinking and business performance.

‘Goal driven’ attitude proved to have an insignificant (p<.05) and negative effect on business performance (Beta=-.167), so Hypothesis 5 is not supported, because it predicted a positive

relationship between ‘goal’ driven attitude and business performance.

‘Expected return’ driven practices proved to have the biggest positive effect (Beta=.274) on the outcome variable among the decision making principles, however it is still above the acceptable significance level (p<.05). Because Hypothesis 6 predicted a negative relationship between

‘expected return’ driven practices and business performance, it is not supported.

‘Competitive Market Analysis’ driven practices have negative effect on business performance (Beta=-.164), but it is still insignificant, so Hypothesis 7 is not supported, although it predicted

a negative relationship between ‘Competitive Market Analysis’ driven attitude and business performance.

‘Avoid the unexpected’ attitude has a low, insignificant and positive effect on business performance (Beta=.020), so Hypothesis 8 is not supported, because it predicted a negative

(34)

33

4.4 Further analyses and testing

Additional statistical analyses were carried out in order to identify possible relationships from the dataset. As the sample equally included respondents with three different experience levels in entrepreneurship, further relationships were tested in this regard.

Relationship between the level of experience and business performance was tested with regression analysis. As experience was measured as a categorical variable with three levels (1. experienced, 2. semi-experienced, 3. beginner), it was recoded into three dummy variables, by handling ‘beginners’ as the baseline group to which other experience levels are compared. Significant relationship (p<.05) was found between experienced (10 years+) entrepreneurs and business performance with a .03 R Square value (Beta=.678), however the whole model was found to be moderately above the .05 significance level (p=.075) with a 5.1% proportion of explanation in the level of business performance (Table 6 & Table 7).

Table 6: ANOVA (Ad-hoc analysis)

Table 7: Regression (Ad-hoc analysis)

Regression Model of Business Performance and Experience Variable R R Square B Std. Error Beta t p .226a _.051 Semi-Experienced (~5years) .021 .594 .317 .209 1.877 .064 Experienced ( >10 years) .030 .678 .327 .231 2.075 .041

a. Predictors: Level of Experience b. Dependent Variable: Business Perf.

ANOVAa Sum of Squares df Mean Square F Sig. Regression 9.474 2 4.737 2.660 .075b Residual 176.281 99 1.781 Total 185.755 101

a. Dependent Variable: Business Performance

(35)

34

Furthermore, relationship between the level of experience and the chosen type of decision making pattern was investigated with One-way ANOVA test. The results did not show any significant relationship between the certain decision making patterns (effectuation or causation) and the level of experience, however when comparing scale means there are clear differences. These will be discussed in more detail in the discussion section of this thesis.

(36)

35

5. Discussion

This section aims to provide a well-reasoned answer to the research question, namely, whether

there is an effect of decision making of entrepreneurs on the business performance of their startups. Furthermore, the most important findings of the research are discussed and reflected

to existing theories and practices. Also, the limitations of the study are presented, which are important to be able to properly assess the findings and implications.

5.1 Theory

This study aimed to contribute to the entrepreneurial decision making literature, namely to the theory of effectuation by Sarasvathy (1998, 2001). While, some scholars are very positive about the effectuation concept and consider its advent as a shift in the theory of entrepreneurship, others, such Perry et al. (2012) and Arend et al. (2015) tend to express some critics about the lack of empirical evidences. Empirical evidences can serve as justification or opposition for a theory, hence it is indispensable to conduct further research in this topic. This thesis aimed to contribute to the extension of the current examinations of the theory, moreover it tried to unveil relationship between effectuation/causation and business performance, in order to provide practical implications for entrepreneurs. The study investigates the effect of the application of the 4-4 principles of effectuation and causation to business performance of new ventures.

Effectuation theory was derived from entrepreneurial expertise and is often linked to experienced entrepreneurs (Sarasvathy, 1998, 2001), which allowed this work to assume a positive linkage to business performance and the certain effectuation principles. The first two principles of effectuation, ‘means’ driven attitude and practices guided by ‘affordable loss’ thinking proved to have a positive impact on business performance, just like the finding of Read et al. (2009) and Brettel et al. (2012). However, the identified effect is very low and insignificant, so the previous studies are not supported in this regard. Furthermore, this study found a negative effect of focusing on partnerships and leveraging from the unexpected to

(37)

36

business performance, yet the effect is very low and insignificant. Therefore, it contravenes to the findings of Read et al. (2009) and Dew et al. (2009).

Also the causation principles were analyzed in the context of new venture performance. Goal driven attitude was identified to have negative impact on business performance, although the result is not significant, which contradicts the ‘planning school’ (Brinckmann et al., 2010 and Wheelwright & Clark 1992). Brinckman et al. (2010) emphasizes the degrading effect in success when applying goal driven mentality in uncertain environments, which fits to the current sample and research, as it examines startups mostly from the fast-growing IT industry. ‘Expected returns’ mentality proved to have positive impact on business performance, which is in align with the neoclassical investment theory (Campbell, 1992), but the result is again insignificant, so the relationship is not valid. Moreover, ‘Competitive market analysis’ had a negative effect on business performance, which is supporting the notion of Grewal and Tansuhaj (2001), as it emphasizes the inefficiency of market knowledge in highly uncertain environments, such as the IT-industry. But unfortunately the relationship was found to be insignificant, therefore the connection has to be refused. Finally, ‘avoiding the unexpected’ driven mentality was found to have a very low and positive relationship to business performance, which contradicts the study of Tatikonda and Rosenthal (2000), but the relationship is again insignificant, so it is not validated.

All in all, the results of the conducted study show only a low level of impact of the certain effectuation and causation principles to the business performance of startups. Moreover, the conceptual model proved to be insignificant, hence the research question is answered as following:

There was no significant relationship found between the different decision making patterns and business performance.

(38)

37

5.2 Discussion of further analyses and tests

Additional statistical analyses were carried out on the dataset in order to unveil any relationship between the level of experience of entrepreneurs and the preferred type of decision making patterns along with achieved business performance. This analysis of relationship is relevant and sensible, because the theory of effectuation was derived from the certain behavioral patterns of experienced, ‘expert’ entrepreneurs and there are some well-known studies that handled this topic and is necessary to reflect on these. Furthermore, the sample happened to have an equal distribution of beginners, semi-experts, experts which allowed a proper comparison.

First, the effect on business performance was analyzed at the different type of experience levels. The results implied significantly better performances at higher level of experience (Appendix 3A). This means that even semi-experts and experts significantly outperformed beginners in terms of business performance of their startups. This raises the question whether which type of decision making pattern was preferred by them opposed to beginners that led to better performance?

By running a linear regression analysis on experience and on the aggregated effectuation and causation variables, there was no significant relationship found. However, the investigation of scale means implied some differences between the certain experience levels.

Beginners tend to have a higher scale mean for effectuation than for causation, which indicates that they prefer to follow the patterns of effectuation when making decisions about their business. Semi-experts also preferred effectuation against causation, but the difference in scale means were less than at beginners. The scale means at experts proved to be higher at causation than effectuation, which means that experts identified themselves better with the causation principles (Appendix 3B).

(39)

38

These statistics indicate that higher the level of experience at entrepreneurs, the less they tend to use effectuation principles. This finding contradicts the theory of Sarasvathy, that experts tend to use effectual logic and also contradicts to the study of Dew, Read, Sarasvathy and Wiltbank (2009), where they compared MBA students to experienced entrepreneurs and found causation patterns at students and effectuation patterns at entrepreneurs. The difference in the findings could be explained with the differences by the definition of ‘novice’, as in the current study, novices were entrepreneurs with less than 5-year experience, meanwhile at Dew et al. (2009) they used MBA students. Business studies are usually more associated with providing knowledge about causation patterns (business planning, expected return, market analysis), so no wonder they applied these elements. But beginners at the current study are young entrepreneurs on the market with low capital, small network and almost no experience, so they are not able to make rewarding market analysis thus, they have to rely more on their creativity, have to experiment more and have to apply new combinations.

In order to gain better insight in this topic, not just the aggregated effectuation and causation variables were analyzed, but also the scale means of the individual decision making principles. Only the mean of ‘affordable loss’ driven practices tend to be slightly higher at experts than at beginners. ‘Means’, ‘partnership’ and ‘leveraging the unexpected’ showed lower preference at higher levels of experience (Appendix 3C). Interestingly, all of the causation principles had higher values at experienced entrepreneurs than at novices (Appendix 3D). These findings imply that the entrepreneurs who have more than 10-year experience favor more secure decisions and became more goal-driven. They tend to experiment less with new combinations and mostly know from the beginning where they want to end up. Probably the experience gathered through the years enables them to have a clearer picture about the market and dispose of a better knowledge about ‘how business is done’. This apparently makes ‘competitive market

Decision making of entrepreneurs : the effect of effectual and causal reasoning on business performance