Entrepreneurial decision-making in a microcosm

Introduction

This article explores when, how and why undergraduate entrepreneurship students on a business-management degree use different opportunity management (OM) behaviours (causation, effectuation, bricolage) in a fundraising project that acts as a microcosm of the challenges faced by entrepreneurs over time (Gibb, 2002). In this article, we use the term OM rather than opportunity recognition, discovery or creation (Alvarez and Barney, 2007; Miller, 2007) as it encompasses both the identification and conversion of a potential opportunity into something valuable. We are aware that other authors have used the terms decision-making or cognitive logics (Dutta and Thornhill, 2014; Reymen et al., 2016) to describe a combination of entrepreneurial intentions and behavioural outcomes. It is beyond the scope of this article to disentangle these differences; instead, we simply use the term OM to encompass both the cognitive underpinnings and actual behaviours.

The best way of preparing university students for entrepreneurial careers has been the subject of energetic, and increasing, debate in a number of literatures (Fayolle, 2013; Gielnik et al., 2015). What they should be taught, or should learn to do, ideally ought to prepare them for becoming entrepreneurs. Historically, causation has been presumed to be the default behavioural logic that entrepreneurs use (Matlay, 2008). Causation refers to where an entrepreneur decides on a predetermined goal and then selects between means to achieve that goal, a process that involves formal planning based on competitor and market analyses (Fisher, 2012; Sarasvathy, 2001, 2003). However, it was increasingly recognised that actual entrepreneurs often do not do this, and certainly not to the extent that some had presumed. Hence, subsequent theorising on entrepreneurial decision-making identified effectuation, in which decisions are based on the identification of means which are subsequently applied to suitable entrepreneurial opportunities, and bricolage, in which neither existing means nor predetermined goals set the entrepreneur’s path, rather it is a process of ‘making do’ with whatever is at hand (Fisher, 2012).

These are therefore the behaviours and skills that students leaving higher education should have learnt. However, there is evidence that this is not what is happening. In parallel with increasing knowledge about what entrepreneurs actually do, there is increasing awareness of the limitations of entrepreneurship education which has tended to focus on the teaching of causation and not effectuation and bricolage. In this article, we reveal when and why students use effectuation, causation or bricolage when they are immersed into a microcosm of a ‘real-life’ entrepreneurial task, fundraising for a social enterprise in a resource-constrained environment. This is in order to identify the contextual imperatives that shape the adoption of the different OM behaviours and thereby provide a more nuanced understanding of the learning needs of entrepreneurship students. Our longitudinal qualitative data captured the students’ journey over time as they described and critically reflected on what they had done and why they had done it (Welter et al., 2016).

Our aim is to expose the contextual influences on students’ behaviour, and changes in their behaviour, in order to help educators facilitate students’ learning about the appropriate use of the different behaviours (Arend et al., 2015; Brettel et al., 2012; Chandler et al., 2011; Fisher, 2012; Perry et al., 2012; Senyard et al., 2015). We are able to provide insights as to why effectuation and bricolage, as well as causation, should form a stronger part of entrepreneurship education. For example, when entrepreneurs are working in a time-unconstrained and predictable situation, causation is an appropriate behavioural choice (Sarasvathy, 2001). In a time-pressured, stressed and risky context, effectuation and bricolage are useful (Fisher, 2012; Perry et al., 2012; Welter et al., 2016). Educators need to understand this in order to guide students appropriately and to provide them with the means to acquire the skills and knowledge that will allow them to make good behavioural choices.

The structure of this article is as follows. First, we review the literature on OM behaviours and the use of live projects as a microcosm in entrepreneurship education. This is followed by a discussion of the methods used to collect and analyse our data. We then unfold the students’ use of different OM approaches within the fundraising projects. The conclusion summarises the contributions of this article to entrepreneurial learning and the role of microcosms in entrepreneurship education and suggests areas for future research and practice.

Entrepreneurial decision-making

The management of opportunities is a core problem for entrepreneurs (Mair and Marti, 2006). The original construct, causation or ‘rational’ theory, is an economics-based approach in which entrepreneurs discern an opportunity and then follow a normative decision-making process to exploit it (Hindle, 2004). By collating essential information, they systematically evaluate choices (Alvarez and Barney, 2007; Miller, 2007) and assemble any necessary resources (Lanivich, 2015). A subsequent development of the theory was the defining of ‘effectuation’ (Sarasvathy, 2001: 245) as a decision-making process that does not begin with a specific goal:

… instead, it begins with a given set of means and allows goals to emerge contingently over time from the varied imagination and diverse aspirations of the founders and the people they interact with, bringing in his or her skills, resources, people, and networks.

Effectuation is means-driven, as opposed to the goal-driven causation. Effectuation’s original five principal components: ‘bird-in-hand’, ‘affordable loss’, ‘lemonade’, ‘crazy-quilt’ and ‘piloting the plane’ (Sarasvathy, 2003) were subsequently operationalized by Chandler et al. (2011) and Fisher (2012) into seven categories that we used as the basis for our analysis.

The most recent aspect of OM to be conceptualised is ‘bricolage’ (Lévi-Strauss, 1967), the application of new combinations of whatever resources are at hand to address an opportunity that has been identified (Di Domenico et al., 2010). Baker and Nelson (2005) characterise three aspects of bricolage: ‘resources at hand’, ‘recombination of resources for new purposes’ and ‘making do’. Bricolage has the additional connotations of refusing to recognise, or be constrained by, existing definitions of a problem (Mair and Marti, 2006).

Causation tends to be used in a flourishing or stable and predictable environment. Here, opportunity creation is incremental, such as the differentiation of a product where its market and competitive environment are already known (Sarasvathy, 2001). Causation involves identifying a gap and developing a plan to address it (Alvarez and Barney, 2007; Sarasvathy, 2003). Effectuation is more common in situations where the future is unknowable and unmeasurable. Effectuation’s principles (Sarasvathy, 2001) are to find the means to adapt resources and create new opportunities (Welter et al., 2016) as well as reducing complexity in an environment that is lonely, uncertain and full of risk (Gibb et al., 2013). Bricolage, which has been less researched than the other two categories, tends to be found in conditions where resources are especially scarce and penurious, and uncertainty-based risk is higher (Senyard et al., 2015; Welter et al., 2016). These conditions are particularly common in the case of business start-ups in a new field (Beckett, 2016; Fisher, 2012). There is a growing literature on how entrepreneurs respond to resource adversity through the use of bricolage, but little work to date has been done on the competences or skills necessary to use bricolage effectively (Baker et al., 2003).

As theoretical constructs, causation, effectuation and bricolage have some critics (Arend et al., 2015) who suggest that each element remains to be fully developed conceptually. A number of scholars have criticised the OM categories as being rather repetitive and overlapping (Welter et al., 2016) and recommend further refinement of the categories. Bricolage appears especially problematic (Baker and Nelson, 2005; Welter et al., 2016). Given this lack of clarity, it can be difficult to understand what aspects of the different concepts are used, and whether they can be used sequentially, or even simultaneously, to address specific aspects of the entrepreneurial decision-making process (Laine and Galkina, 2017).

However, increasing numbers of scholars are examining their use (Fisher, 2012; Mäkimurto-Koivumaa and Puhakka, 2013; Welter et al., 2016) and finding that entrepreneurs use a range of different decision-making logics to deal with the creation of a new venture under conditions of uncertainty (Baker and Nelson, 2005; Dew et al., 2015; Fisher, 2012; Read et al., 2009; Sarasvathy, 2001). One of the important inputs into decision-making generally, and this holds true for entrepreneurs as well, is the role of experience (Dew et al., 2009; Sarasvathy, 2001; Wiltbank et al., 2006). Experience leads to the development of decision-making heuristics, which shape behaviour (Fodor et al., 2016; Fodor and Pintea, 2017; Foo et al., 2015; Delgado-Garcia et al., 2015). Bad experiences can leave psychological ‘scars’ which steer the individual away from using the same decision-making logics, even if it would be an appropriate strategy in different circumstances; in contrast, positive experiences can lead to the same behaviour being repeated, even if it would be ineffective in the new circumstances. However, few studies have examined how entrepreneurs decide which logic to use (Middleton and Donnellon, 2014; Salusse and Andreassi, 2016) or how entrepreneurship students are taught which are the most appropriate. An awareness of these gaps underpinned our study.

Methodology and data analysis

Our intention was to understand the use of OM behaviours by entrepreneurship students in a live project that mimicked, as far as was possible within the higher education context, the entrepreneur’s world (Dew et al., 2015; Lehner and Kansikas, 2012). Understanding the contextual influences on these processes, and how they change over time, requires longitudinal qualitative data and an interpretive epistemology (Gray, 2007; Jones et al., 2011).

Students were tasked with raising money for a social enterprise without any prior budget. This took place within an elective entrepreneurship module on an undergraduate business degree in a UK business school during the second year of a 3-year degree programme. It lasted 10 weeks. Students were not required to have any prior knowledge of entrepreneurship concepts or theory in order to join the module, although all had previously taken core business and management modules (e.g. marketing, accounting, economics). Our objectives were to enable students to be immersed in a real project, a form of ‘learning by doing’ (Pittaway and Cope, 2007). The microcosm, as described above, would allow students to learn the skills that entrepreneurs need and also develop some empathy for the entrepreneur’s world. Mentoring was provided by academic staff who met weekly with the students in order to provide feedback and discuss options. The amount of money raised was not to form part of the assessment: instead, students were graded according to the quality of evidence of their activities and their reflection on their learning.

A total of 124 students participated in the module, divided into 25 teams. Each team was tasked with raising funds for one of six social enterprises that staff had pre-selected and that had agreed to make themselves available to the students (Table 1).

OPEN IN VIEWER

Table 1.

Demographics of the student groups and commentary on group functioning.

Group no.	Social enterprise	Male	Female	International diversity (%)	Experienced member (%)	Performance against target (%)	Comments on group processes
1	International aid charity	2	3	40	0	45	Weeks 1–3, 100% attendance, with all tasks completed. Weeks 4–6, attendance at meetings went down to 60%. Conflict led to reduction in ambitions for the project. By weeks 7–9, attendance returned to 100%, aided by tutor’s intervention. Tended to improvise during the event in week 8 as they reflected on their performance.
2	Hospice	2	3	20	0	62	High attendance during initial weeks, individualistic, collective, low motivation-poor allocation of task but full attendance in weeks 9 and 10.
3	Hospice	4	0	0	0	87	Highly motivated group with maximum weekly attendance. Worked collaboratively with clear allocation of tasks. Sought to resolve conflict. Self-appointed task leaders, based on perceived expertise.
4	Hospice	5	0	0	0	25	Lacked motivation and poor attendance at the start; 60% until week 6 when members became conscious of assessment requirements. Lack of communication within the group. Sought to resolve conflict at later stages. Event cancelled (funds came from sponsor).
5	Hospice	1	4	20	40	134	Highly motivated group with consistent attendance and engagement. Full engagement from members. Clear allocation of tasks, but tended to be highly detailed in weekly plan with too many action points, and repeated reminders of the task from one experienced member caused conflicts between weeks 5 and 7. Poor donation levels in week 7 led to conflict resolution in order to achieve the target.
6	Teenagers’ charity	5	1	0	20	117	Highly motivated and engaged group with clear allocation of tasks and roles between weeks 1 and 6. Conflict developed as the experienced member accused the inexperienced members of being less engaged. Stung by the accusation, one of the inexperienced members initiated a collection in weeks 8–10.
7	South African women’s charity	1	4	40	0	123	Highly engaged and motivated group throughout the 10 weeks. Good level of discussion led to clear and detailed allocation of tasks and roles.
8	Head injuries charity	4	0	50	0	42	Attendance was consistent throughout. Enthusiastic, but poor in their execution of tasks. No conflict or argument over proposed ideas, but actually little discussion too. Most of the tasks and roles were appointed by one self-appointed ‘ringleader’.
9	Head Injuries charity	3	2	60	20	107	Highly engaged and motivated group throughout the 10 weeks. High-quality discussion of ideas led to clear and detailed allocation of tasks and roles.
10	Children’s charity	3	3	0	0	8	Group conflict was at the centre of this group over the whole 10 weeks. Members were not united and had different goals for the event which were not resolved. Allocated tasks were not executed. There were a lot of accusations flying about, with members not trusting each other. Despite support from tutors, group members were still at loggerheads until week 10. The event did not happen; the only income came from donations.
11	Children’s charity	1	4	20	20	160	Highly engaged and motivated group throughout the 10 weeks, high level of discussion of ideas led to clear and detailed allocation of tasks and roles.
12	Children’s charity	2	2	0	0	93	Highly engaged and motivated group throughout the 10 weeks, High-quality discussion of ideas led to clear and detailed allocation of tasks and roles.
13	Children’s charity	3	2	20	0	33	Group attendance hovered around 60% between weeks 3 and 6. By week 7, attendance increased to 100% as members became aware of the need to develop ideas for the project. Shortage of time was the prime reason for the poor performance.
14	Children’s charity	2	3	20	0	117	Highly engaged and motivated group throughout the 10 weeks. High-quality discussion of ideas led to clear and detailed allocation of tasks and roles.
15	Children’s charity	5	0	20	0	22	Characterised by group conflict throughout the 10 weeks as one dominant international student thought he worked harder than others and members were avoiding him. He was complaining that others did not complete the work that he had allocated. Tasks achieved were never to his satisfaction. Other group members eventually contributed to the event, but this was despite his wishes.
16	Hospice	4	2	0	0	83	A motivated group where any conflict over ideas was resolved. Roles and tasks allocated well and were always achieved. Attendance averaged more than 80% over the 10 weeks. Did not achieve their intended income target through no fault of their own.
17	Children’s charity	5	0	40	0	67	Enthusiastic group. Full attendance from weeks 1 to 6, then attendance at 80% in weeks 7–9 and full attendance at week 10. Conflict arose when roles and tasks were not clearly defined in the action plan, were not in actionable form and therefore weekly targets were missed.
18	Hospice	3	3	17	0	105	Highly engaged and motivated group throughout the 10 weeks, with the exception of two international students who were not engaged for the first 4 weeks. However, the group worked hard in order to get the two members involved. High-quality discussions of ideas led to clear and detailed allocation of tasks and roles.
19	Children’s charity	3	2	80	0	85	Highly ambitious, internationally diverse, group. Extremely motivated towards putting ideas and plans into action. However, by weeks 7–8, motivation of members had dropped due to absenteeism and conflict over the allocation of tasks. Targets were over-ambitious and the group was not able to achieve the targeted amount. One member remained in disagreement till the very end.
20	Head injuries charity	5	0	20	0	54	Members clearly enjoyed working with each other. There was no sign of conflict. However, the weekly action plans were rather descriptive and roles, tasks and targets to be achieved were not clear and rather vague.
21	Head injuries charity	2	3	0	0	83	A very enterprising and motivated group with clear allocation of task and roles. No indication of conflict throughout the 10 weeks.
22	Head injuries charity	1	4	20	0	68	A highly motivated group with full weekly attendance. However, were rather inward looking, preferring to use their own resources instead of working with sponsors or suppliers
23	Hospice	2	2	25	0	52	A group of very reserved and quiet individuals. Initially developed a lot of ideas but because of procrastination little action was taken to develop the ideas until week 6 where time became pressing. Tutors needed to put more effort into guiding this group as members were rather timid and risk averse.
24	Teenagers’ charity	3	2	0	0	53	A group that experienced conflict from week 1 due to the religious beliefs of a member, who did not want to visit the SE because of its perceived lack of cleanliness. Nevertheless, all members worked collaboratively with high levels of attendance each week. Members were courteous to one another, with clear task allocation and feedback. Getting sponsorship was only successful from week 7 onwards, and on the day of the sale, members could not turn out which resulted in poor performance.
25	Teenagers’ charity	3	2	20	0	42	The one international member in the group attempted to set himself up as the role model to the rest of the group. This became the source of conflict and frustration. Eventually, the members came together as a team in order to work on the fundraising event. By weeks 7–9, they were working closely with sponsors, but not potential consumers.

Fundraising activities included social events, food fairs, salsa evenings, charity auctions, sales of T-shirts, computer games competitions, radio advertisements and a photo gallery competition. Each team was required to report on their progress weekly, reflecting on what they had done, and why, and what had worked and what hadn’t (Lindh and Thorgren, 2016). These reflective logs comprise our primary data, onto which we subsequently mapped the CEB constructs and the contextual influences on them (Pittaway and Cope, 2007; Pittaway et al., 2011). The logs averaged between 9000 and 10,000 words per group, about 250,000 words in total. All groups gave permission for their data to be used for research purposes. The data from the student accounts were supplemented by information from tutorials, observation of student interactions, discussions with students, social enterprises and tutors, all of which were recorded and discussed between the teaching and research team. These helped to inform the analysis and judgement of why and how students behaved as they did.

We used Fisher’s (2012) scale to identify inductively indications of the use of the different OM behaviours within the logs. These emerged through a process of theme identification and interpretative synthesis (Howard-Grenville et al., 2013; MacKay and Chia, 2013) rather than the positivistic application of a predetermined framework (Åsvoll, 2014; Gray, 2007). Students were not asked to categorise their behaviours; instead the authors went through the transcripts and categorised any behaviours that could be identified. They agreed on the themes and the OM behaviours identified. Where there were differences of interpretation, these were discussed and a common conclusion reached. There was no attempt to measure the degree of disagreement or inter-coder reliability ratings; all classifications were discussed and an agreed decision reached.

Each behaviour was allocated exclusively to one of the CEB categories as described in the ‘Findings’ section. Almost all of our data could be fitted into one of the CEB categories. Behaviours which appeared to fit into two or more categories were also noted and allocated to what appeared to be the most appropriate. This process allowed us to believe (as we discuss later) that the cognitive logics’ categories need some refinement in order to remove duplication and overlap.

For each instance of a particular behaviour, we looked for explanations for why they had been chosen, for example, the characteristics of the project environment (e.g. stable, pressured, complex), team characteristics (e.g. diverse, experienced) and group dynamics (e.g. conflicted, demotivated). In doing this, we were looking for causal and moderating factors that underpinned the choice of behaviours as the fundraising project developed over time.

Findings

In this section, we describe the ways in which students used the different categories of OM behaviours, when they used them and what stimulated their use. We especially identify changes in behaviours as the fundraising project progressed and differences in the behaviours in different groups. In total, nearly half of the behaviours that we discerned could be categorised as causation, a quarter could be defined as effectuation, and another quarter as bricolage, but the preference for one over the others changed over the lifespan of the project (Figure 1).

OPEN IN VIEWER

Figure 1.

Frequencies of Opportunity Management.

We discuss in depth the reasons for these choices and the influences on any change below. For now, we simply note that causation was prevalent between weeks 2 and 5 when students used their learning from previous courses such as accounting, finance and marketing. Week 2 was when most groups visited their social enterprise and became aware of its ethos and need for regulatory compliance. Bricolage became particularly apparent at key dates later in the project, as the project path turned out to be not quite as the students had predicted. Both effectuation and bricolage increased over time as the students learned what worked rather than what they had learned in the classroom.

In the following sections, we discuss in more detail the OM behaviours used and provide explanations for these choices.

Causation

In looking for indicators of causation, we based our analysis on Fisher’s (2012) definitions. Figure 2 shows the pattern of causation behaviours over the 10-week period. Examples of students’ causation behaviours are shown in Table 2.

OPEN IN VIEWER

Figure 2.

The pattern of causation behaviours over the 10-week period.

OPEN IN VIEWER

Table 2.

Examples of students’ causation behaviours.

Causation category	Example	Week	Stimulus	Consequences (intended and unintended)
Develops a project plan to develop the product and/or services	We need to arrange the next group meeting. Next Wednesday before classes – NB: **cannot join, however he will contribute on Facebook or in mail. Learn by reading how this idea could be marketed. (Gp 17)	1	Awareness of the need to conform to academic requirements and achieve the targeted fundraising objectives	Helped to identify options and allocate tasks and roles within the group. Identified expertise and inexperience. In some groups led to individual attempting to dominate in order to try to achieve the intended outcomes.
Organises and implements control processes	This week we aim to catch up on the tasks we were not able to achieve last week … our biggest objective is to secure and guarantee use of our chosen venue. Once this has been achieved we can be a lot more specific with our tasks and team goals. (Gp 1)	3	Awareness of failure to achieve targets and pressing time frame for decisions to be taken; desire to succeed; application of tools and techniques that they had learned previously.	Made the internal environment more predictable. Gave structure to decision-making and confidence that the process was on course.
	This week we focused on the Risk Assessment, pinpointing the types of uncertainties that could affect the progress of the event. The risk could arise from two main variables: the venue and stakeholders. (Gp 12)	5	Increasing awareness of the potential of university regulatory non-compliance.	For dysfunctional groups provided a source of stress and increased dysfunctionality, leading to demotivation and absenteeism, and eventually to greater use of bricolage later in the Project.
Calculates the returns of various opportunities	84% of the students interviewed showed interest in a theme party, and told they would come. This confirms our belief that we will fulfill unmet needs within our target. Then, the 4 options submitted got close results, so we need to choose a theme which satisfies everybody. (Gp 12)	4	Initial research brings about an awareness of the need to identify potential customers and the most profitable returns.	Applied learning from previous classes. In some cases, spurious confidence in the planning and financial calculation process led to errors that might have been prevented if the group had taken a more experimental route.
Gathers information about competitors and compared their offerings	Our main competitors would be the other students who are doing this same module and have the same assignment. All the information about the ideas for fundraising have been kept confidential. However, we were able to gather some information about the events through word-of-mouth … Since this event is completely different from the one we are hosting, we don’t think it would be a potential competitor or a threat. Another competition might be the Messy Mondays event which takes place every Monday. However, we plan to collaborate our event with Messy Mondays’ in order to attract more people. (Gp 19)	5	Increasing awareness of the performance (and potential threat from) other students within the cohort, desire to do well, recognition of the synergistic possibilities from collaboration.	Increasing awareness of competitors and potential for other options. Increasing awareness of the need to exploit additional resources and break out from conventional ‘zero-sum game’ thinking.

Effectuation

In identifying effectuation, we again looked to Fisher (2012: 1030) and Chandler et al.’s (2011) seven categories of effectuation behaviours (Figure 3). Examples of students’ effectuation behaviours are shown in Table 3.

OPEN IN VIEWER

Figure 3.

Effectuation behaviours over 10 weeks.

OPEN IN VIEWER

Table 3.

Examples of students’ effectuation behaviours.

Effectuation category	Example	Week	Stimulus	Consequences (intended and unintended)
Experimented with different ways to sell and/or deliver the product or service in arriving at a commercial offering.	Create Just Giving Sample page and write down IN IT a step-by-step on how to do it in order to help contestants create their own. Use Freemium social media for research and getting sponsors: Use Time-Out Magazine to source contact numbers for various art galleries. Use Google to research sources we can get prizes from (e.g. PC World, Argos, Amazon etc) (Gp 5)	6	Experienced group was aware that different offerings would produce different results. Also experienced with digital media and research.	Ability to compare possible outcomes. Excitement as the opportunities became apparent.
Changed the product or service substantially as the venture developed.	(Following an attempt to broadcast on radio which did not raise sufficient funds) In week 6, we decided to execute another event to raise money – using the [donation] bucket available from SE along the street. (Gp 6)	6	Commitment to achieve the goal; motivated to achieve the target, and awareness of the lack of success of the original idea.	Stress of failure stimulated awareness of the need to change course. Even experience (Group 6) did not prevent poor performance and the need to rethink what needed to be done. Stimulation of creative problem solving and effectual experimentation.
	There is no interest in participating in the talent show. Found a Zumba trainer [group decided to hold a Zumba dance], need to find another trainer as a backup in any case our primary can’t attend. (Gp 4)	8		Stimulation of seeking out of external resources.
Committed only limited amounts of resources to the venture at a time	Also, we thought about extra marketing actions that would generate more benefits for SE …. For that we need to contact:Suppliers to provide the communication media (tee shirts, bracelets and gadgets)Suppliers to do the printingEventually, sponsors to cover the extra costs. (Gp 12)	3	Awareness of internal resource limitations.Desire to not lose their own money.	Inexperienced group took longer to realise that they have limited resources of their own and needed to seek external resources.Some groups decided to use their own money rather than seek resources from elsewhere.
Adapted what they were doing to the resources at hand	After talking with the course instructor about our project and the viability of it, it was decided that the project can be done but would be more interesting for our clients to be able to buy a piece of cake and a hot beverage [they decided to hold a British Tea Party to capitalise on the London Olympics ‘vibe’ that was happening at the time]. We will be selling cakes and tea and coffee. (Gp 11)	3	Recognition of the need to be realistic and creative in terms of maximising resource utility.	Inexperienced group thought they should use their own money in order to make the event happen, even though this was not an officially sanctioned option.Group with experienced members knew earlier, and better, what they could do.
Entered into agreements with customers, suppliers and other organisations	We will also physically meet or contact the student union at their campus offices on Monday inquiring on how and where to get permission to use the Uni yard for our cake stall.	2	Recognition of need to obtain the external resources provided by customers, suppliers and so on in order to ensure success of the Project.	Learning about the potential expansion of resource base from ecosystem members.
	Meeting with coffee machine sponsors on Monday to see what their decision is in securing us a machine for the event(s). (Gp 11)	3

Bricolage

In looking for bricolage, we looked to Fisher (2012), Chandler et al. (2011) and Senyard et al. (2009) to identify eight categories of bricolage (Figure 4). Table 4 gives some examples of students’ bricolage behaviours.

OPEN IN VIEWER

Figure 4.

Bricolage behaviours over 10 weeks.

OPEN IN VIEWER

Table 4.

Examples of students’ bricolage behaviours.

Bricolage category	Example	Week	Stimulus	Consequences (intended and unintended)
Combination of resources for new purposes. Combined existing resources in creating solutions	I can … promote my shoe business and at the same time tell the world about how my business is helping the SE. (Gp 11)	3	Increased understanding of the potential for synergies.	Improved resource allocation.Increased awareness on the part of inexperienced groups especially that they could achieve more by being creative in the way that they brought together resources.
Reused resources for purposes other than those for which they were originally designed	Fortunately, the Student Union has a licence to screen a movie for student entertainment and they allowed us to use this licence to generate income for the SE. (Gp 17)	8	Awareness of the limitations of the group’s own resources, therefore had to cast around for others’.	Inexperienced groups lacked ideas and at the last minute needed to depend on staff suggestions, for example, Gp 17.Experienced groups had the ability to assess the potential of what was available, leading to earlier and better use of resources.
Physical inputs. Used forgotten, discarded, worn or presumed ‘single-application’ materials to create new solutions	[The sunflower was an iconic symbol used by the SE in fundraising campaigns]Decided to come out with the Sunflower Event as SE had a lot of Sunflower badges left from a previous event. The sunflower was to make awareness of the SE. It can be bundled as a package [with other donated goods] for sale. (Gp 18)	7	Opportunistic recognition of spare resources.	Learning of the potential synergies between SE’s ethos and underexploited resources.Awareness of the potential to increase revenue without spending anything on inputs.
Skills inputs. Encouraged the use of amateur and self-taught skills that would otherwise go unapplied	This week, as we have changed idea drastically we need to get a big move on and the new idea is based on selling products … Therefore this week each member must contact at least one company and ask them, in detail, what can be done to help us. In short, it is a mini phone pitch to them for sponsorship. (Gp 1)	5	Awareness of the pressing need to avoid the event failing.	Fear of failure, stimulating the last-minute use of amateur skills.

Discussion

Our study investigated students’ use of causation, effectuation and bricolage within a fundraising project that provided a microcosm of the entrepreneur’s world. We were interested in when, and why, students used the various OM behaviours over time (Figure 5). Drawing on previous process-based research, we found that the three OM behaviours adopted by students during the 10 weeks were triggered by different kinds of stimulus, teleological, dialectical and evolutionary (Rasmussen, 2011; Reymen et al., 2016; Van de Ven and Poole, 1995).

OPEN IN VIEWER

Figure 5.

The use of opportunity management behaviours over time.

Each of these stimuli focuses on different aspects of the new venture process. Similar to Rasmussen (2011), the contours of a stage or life-cycle process (causative goal formulation, implementation, evaluation and modification of goals), which assumes that change processes proceed through defined steps or stages of development, were visible in all of the teams, but progression through the life-cycle stages were by no means linear or uniform across the groups. Progression was affected by dialectics (academic and SE input; conflict) leading to changes in the balance of power between opposing entities. Teleological processes (all groups had the final event in mind) guided the groups’ path towards the goal, although this path was shaped by experience, emotions (Foo et al., 2009) group diversity and stress, and evolutionary theory, which assumes that change processes move through a continuous cycle of variation, selection and retention was influenced by unpredictable events, environmental changes and history.

Our results also concur with Dew et al. (2009) who found that MBA students made decisions regarding possible opportunities using a ‘predictive frame’. Established entrepreneurs use causation sparely (Dew et al., 2015), yet causation tends to dominate business school teaching, including our students (Kickul et al., 2010; Matlay, 2008). We suggest that this conditioned them to behave in ways (Calvard, 2015; Minniti and Lévesque, 2008) that may have been ineffective given the nature of the task, its time frame and the resources that they had available (Duxbury, 2014). We also speculate that causation, being less risky and psychologically expensive (Grupe and Nitschke, 2013; Pittaway and Cope, 2007), prevailed over experimentation because of the students’ need to address two goals simultaneously – to engage in entrepreneurial value creation and to pass the module.

Groups with an experienced member, who acted as coach and conflict-mediator, used more effectuation and also exhibited more of all three types of OM behaviours than the inexperienced groups. Although competitive performance was not part of the students’ brief or assessment, five of the groups that surpassed their fundraising targets used more OM behaviours, specifically more effectuation, than the less successful groups. Those that encountered conflict, typically because of the international diversity of the group, tended to perform worse. Intriguingly diverse groups which also had a member with entrepreneurial experience achieved more than their intended goals more than low diversity groups with an experienced member. We suppose that experienced students could exercise their knowledge of the range of OM behaviours, capitalising on the resources within the group and choosing which behaviours to use to their greatest effect. This suggests that groups with entrepreneurial experience are able to capitalise on diversity, whereas those without cannot.

Tight deadlines and perceived task-difficulty act as stressors (De Clerccq et al., 2014), putting pressure on team functioning (Costa et al., 2015; Lepine et al., 2005). Students became increasingly aware of the multiple expectations that they were faced with, having to pass the module at the same time as raise money for a deserving cause. These may reduce the propensity to take brave decisions or risks (Bradley et al., 2012; Dutta and Thornhill, 2014), something which experience appeared able to counter. Inexperienced students tended to rely on what they had been taught, whereas experienced students had a greater repertoire of knowledge to draw upon. The groups only needed one person of this kind to influence choices of OM behaviours (Brettel et al., 2012); for some reason, more than one experienced member made little difference. Nevertheless, it suggests that educators may need to intervene in the membership of groups so that entrepreneurial experience is part of the mix. Inexperienced students were dependent on rationalisation and analysis rather than intuition or other heuristic-based decision-making processes that more experienced students could draw upon (Kickul and Gundry, 2011). Experience may give confidence in decision-making processes and is also likely to increase awareness of the types of resources needed (Berends et al., 2014), the means of obtaining them and the consequences of failure (Brinckmann et al., 2010; Gielnik et al., 2015; Rasmussen and Sørheim, 2006).

Conclusion and recommendations for further research and practice

Our findings contribute to theoretical debates in a number of areas of entrepreneurial learning. First, there is little empirical research on the use of OM behaviours, less on the underlying reasons for the choice of these behaviours and even less on the use and causes of these behaviours by students (Dew et al., 2009; Perry et al., 2012; Welter et al., 2016). Our research also introduces the microcosm as a learning environment that allows students to enact OM behaviours, thereby mixing practice-oriented and theoretical knowledge in OM education (Fayolle, 2013). Currently, its teaching is largely disconnected from the exigencies of entrepreneurial practice (Edelman et al., 2008; Vanevenhoven, 2013). The approach that we describe in this article helps to bridge this gap in showing how students develop their ideas into outcomes in a specialised-task setting which in our case focused on fundraising for a social enterprise.

The use of a fundraising project as a substitute for learning in an actual entrepreneurship context focuses attention on the role of microcosms as pedagogical devices (Kapranos, 2016; Welter et al., 2016). Kyrö’s (2015) view is that microcosms offer a radically new vision of learning because of their creative, responsibility-inducing and risk-exposed dimensions. Although they have some differences, they have some similarities to the entrepreneur’s world and mimic to some extent the pressures that entrepreneurs actually face. They encourage, and even force, students to engage in a range of resource-creation activities. However, what a microcosm should look like is more uncertain. We believe that the fundraising project that we used in this study provided a useful pedagogical device for engaging students in the entrepreneurial process and creating empathy among novices about the entrepreneurial life (Chang and Rieple, 2013; Gibb, 2008; Kapranos, 2016). There are few empirical studies on what makes for effective learning environments of this nature (Fayolle et al., 2016). For example, how difficult and time-pressured should the task be? Complexity, risk, uncertainty and working with different stakeholders are all contextual factors known to shape behavioural choices in the ‘real’ world. Should a microcosm attempt to imitate this? If so, how? Students have different levels of skill and different experiential backgrounds to entrepreneurs, and the nature of the task needs to reflect this if students are not to be frightened off from entrepreneurial careers.

Responding to a resource-constrained task forced our students to use a range of skills (e.g. creativity, negotiation, working with stakeholders). And from their logs we could see a developing awareness of their need to change behaviours, and arguably become more resourceful, over time. This was despite not having been formally taught these skills. However, we caution that coaching was used to support the students, which may have shaped the students’ behaviours as some of the academic coaching team were experienced entrepreneurs. As Fayolle (2013) suggests, coaching is inherently effectual as it provides a feedback loop that encourages students to reflect on what they have done, what has worked and what now needs to be done differently. Arguably, this is why we saw the use of effectuation only at a later stage of the project (Sarasvathy, 2001). We would recommend that the principles of effectual or bricolage logic should be taught, if only to make students aware of their possibilities and perils. Further research could attempt to understand the differences between students who have been taught about the OM behaviours in a conventional classroom and those who learn about them through discussions with experienced entrepreneurs or through learning-by-doing. We also did not examine the learning process and further research could usefully identify which platform provides for the most insightful personal development.

We did not specifically explore the issues around diversity and numbers of relevant groups were small. This is something that we believe would warrant further investigation. Our data revealed that the national background of group members was an important influence on behaviours. Diversity is a known contributor to group conflict, as its opposite, homogeneity, is a known factor in group agreement or groupthink (Apfelbaum et al., 2014; Moreland et al., 2013). The cultural background of students and the socio-psychologically derived attitudes that are the result of innate factors in combination with socialisation processes (Autio et al., 2013) are likely to have influenced decision choices through enhancing the potential for conflict within groups, given that they were under considerable pressure (De Wit et al., 2012; Nouri et al., 2013). How, specifically, diversity influenced behaviours remains to be answered. However, we hypothesise that a student’s previous entrepreneurial experience likely to make a diverse group more able to resolve any conflict that does develop, as skills in the resolution of disagreements and handling of stressors are brought to the discussions on task behaviours (Yeung et al., 2015).

Similarly, we did not evaluate the effect of the students’ intended future careers on behaviour, for example if, whether they had already decided to create their own ventures when they left university they were more motivated to experiment with different behaviours.

A question that remains to be answered is whether the fund-rising performance would have improved if students had been more familiar with the different types of OM behaviours through being introduced to them in the classroom. Although this was not part of our study, it appears that there were links between the use of a wider repertoire of OM behaviours and financial performance. We speculate that learning about OM concepts would help students to use them more effectively, improving their performance accordingly. This is something that would benefit from further research. Whether students would use the same types of behaviours given a different microcosm has also not been studied. A better understanding of the links between context and behaviours would help both educators and practitioners alike to understand when certain categories of behaviours are more useful. Our study did not attempt to measure the effectiveness of the different behaviours or link them with the types of fundraising activities that the students chose. Future studies could also include the backgrounds of students, for example, whether those from specific academic disciplines or different cultures, influences which OM behaviours they prefer, and why.

Other types of entrepreneurial behaviours were not considered in our study. An example of this is the role of ‘improvisation’ by entrepreneurs as a possible additional approach to the categorisation of entrepreneurial activity (Duxbury, 2014). Baker and Nelson (2005) suggest that organisational improvisation can be an important precursor to bricolage, yet this is not considered in Fisher’s model and we did not investigate it. Students working under time and physical resource constraints faced unpredictable and unanticipated consequences (Duxbury, 2014; Fayolle, 2013), making improvisation appropriate. As a side issue, we would concur with those that have criticised the OM model as being rather repetitive and overlapping (Welter et al., 2016) and would recommend further refinement of the categories. Bricolage is where our data encountered unclear boundaries between the different categories (Baker and Nelson, 2005, 2012); we sometimes found it challenging to decide in which classification data should be placed. For example, ‘using resources at hand’ (effectuation) and ‘making-do with what we have’ (bricolage) seemed to overlap (Welter et al., 2016).

Finally, although not discussed in detail in this article, we saw evidence of students identifying themselves as ‘entrepreneurs-in-the-making’. We argue that this is the effect of the microcosm, which imitated as far as possible the real world and the real pressures, experienced by entrepreneurs. We suggest that the students’ obvious engagement and immersion with the entrepreneurial decision-making process reflects the powerful effect of ‘real-life’ action learning (Gielnik et al., 2015). Thus, another avenue for further research is to understand how the construction of an entrepreneurial identity can be encouraged through engagement with actual entrepreneurs engaged on genuine entrepreneurial tasks.