Thesis
What Defines a Viable Concept?
Case Study Boerenfluitjes
What Defines a Viable Concept?
Case Study Boerenfluitjes
University of Groningen
Faculty of Economics and Business
Msc Business Administration
Small Business & Entrepreneurship
August 2013
Supervisor: dr. ir. H. Zhou
Second supervisor: dr. M.J. Brand
Principal: Countus accountants en adviseurs
Supervisors: Geert Kuipers en Diane Versteeg
ABSTRACT
This research examines the viability of Boerenfluitjes. Boerenfluitjes is a social start up which focuses on closing the gap between customers and farmers/growers. A framework is developed to determine this viability. This framework exists of four elements 1) the entrepreneurial team; 2) the process; 3) the organizational strategy; and 4) the environment Furthermore, willingness to pay is identified as an indicator of viability. The contingent valuation method is chosen as method that best fits this research. This framework is applied to Boerenfluitjes and the outcome of this analysis is that it seems that Boerenfluitjes is not a viable start up in its present form. It should pay more attention to the communication between the team members and to the documentation of agreements. Furthermore, it should pay attention to its environment. And when developing a business model, it should not focus on daytrips, since potential customers are not willing to pay enough to cover the cost price of the daytrips.
Key words: Social entrepreneurship; start up; viability; Boerenfluitjes; willingness to pay
Acknowledgements: Writing my master thesis was not an easy task, and I encountered some
TABLE OF CONTENT
1. Introduction 6
2. Theoretical framework 8
A. Description of a viable social enterprise 8
Social enterprise 8
The creation of a new enterprise 8
Entrepreneurial team 9 Process 10 Organizational strategy 10 Environment 11 Framework 12 Viability 13 B. Willingness to pay 13
Methods to measure willingness to pay 14
Open-ended questions format 14
Conjoint analysis 14
Contingent valuation 15
Comparison of the different methods 16
Conclusion 17
Contingent valuation 17
Adaptation of the contingent valuation 18
3. Methodology 20
Introduction of Boerenfluitjes 20
A. Measurement of social entrepreneurship and of the viability of an enterprise 21
Content analysis 21
Observations 22
B. Measurement of willingness to pay 22
Questionnaire 22
Biases 23
Dependency upon the scenario 23
Non-response error 24
Development of the questionnaire 24
Target population 24
Distribution 25
Reliability 26
Validity 27
Willingness to pay 28
4. Findings 31
A. Social entrepreneurship 31
B. The viability of Boerenfluitjes 31
The entrepreneurial team 31
The process 33
The organizational strategy 34
The environment 34
C. Willingness to pay 35
Influence of need for the service 35
Influence of the income level 36
Influence of the amount of information 36
Influence of the image of the Dutch country side 37
Influence of the interest in the origin of food 37
Influence of age 38
5. Conclusion and discussion 39
A. Viable concept 39
B. Willingness to pay 39
C. Advice to Boerenfluitjes 39
D. Limitations and further research 40
6. Epilogue 42
References 43
Appendices
A. Summary of the content analysis 46
B. Questionnaire 50
C. Results of the questionnaire – descriptives 57
D. Description of the target population 65
E. Reliability 68
F. Factor analysis of willingness to pay 69
G. Analysis of open question 70
H. One-way ANOVA 82
INTRODUCTION
Almost twenty per cent of all start ups will not survive their first year. Roughly fifty per cent of all start ups will not survive the first five years (Wijnen and Pouwels-Urlings, 2012). This number indicates that only the strongest start ups will survive. Therefore it is important to enter the market well prepared, since this means the start up has analyzed the market carefully and knows what might happen and how to react to certain events.
Boerenfluitjes is such a start up. It is an organization that was introduced on Twitter three years ago by Ester van Aalst. Ester van Aalst lives in a city in the west of the Netherlands. When she was in the supermarket one day, she realized that she did not know where here food comes from. She likes to know the story behind her food and she found that many more people who live in the big cities in the Netherlands want to know this to. Therefore, she started Boerenfluitjes. It’s goal is to make customers more aware of the food they buy in the supermarket. It tries to accomplish this goal by bringing customers and farmers/growers in contact with each other. A farmer is someone who keeps livestock and a grower is someone who cultivates fruits and vegetables. At the moment, the activities of Boerenfluitjes are mainly on the social media platform Twitter. Boerenfluitjes has two main activities on Twitter. First, it wants to simulate customers to ask questions about the food they buy in the supermarket. And second, it wants to stimulate farmers and growers to explain the story behind their products in an easy, understandable way.
Organizations that exploit opportunities which are aimed at the enhancement of social wealth are called social enterprises. These organizations create social value by addressing social needs and the whole society benefits from this process (Hoogendoorn, 2011: 12; Zahra, Gedajlovic, Neubaum and Shulman, 2009). The social need Boerenfluitjes tries to address is to raise the awareness of customers about their food. When this need is fulfilled, customers will make more conscious choices when they buy or consume food. The whole society will benefit from this, because less food is wasted, which is good for the environment. For these reasons, Boerenfluitjes can be seen as a social enterprise.
One can say that Boerenfluitjes already survived the first two difficult years. But at the moment, the organization does not have a business model, since it is not possible to generate a turnover with the activities on Twitter alone. Therefore, Boerenfluitjes is looking for other activities, for example daytrips to farms. A question rises, is it possible to attach a business model to this organization? Are potential customers interested in the activities Boerenfluitjes wants to offer? And are potential customers willing to pay for them?
To date, a lot of research has been done to the viability of start ups. This research will be combined and applied to Boerenfluitjes in order to determine if it is a viable concept.
The main question of this research will be:
Is Boerenfluitjes a viable social concept?
The sub questions that will be used to answer this main question are: 1. What defines a viable social concept?
To answer the main question, a theoretical framework will be developed. In this framework the first two sub questions will be answered. Therefore, the framework will consist of two parts. In the first part the focus will lie on the definition of a viable social concept. At the end of this part, a framework to determine the viability of a social concept will be presented. In the second part of the theoretical framework, the focus lies on the assessment of willingness to pay. Different methods that exist to measure willingness to pay are described and compared to each other. After this description, the method that best fits the purpose of this research, the viability of a new social concept, will be chosen and adapted to correct the possible disadvantages.
After the development of the theoretical framework, an empirical study will be conducted to answer the last sub question. The framework to determine the viability of a social concept and the method to measure willingness to pay will be applied to Boerenfluitjes. This will be done by means of a questionnaire, by observations of the researcher and by an analysis of the available documents about Boerenfluitjes.
At the end of this report, a conclusion about the viability of the concept Boerenfluitjes will be drawn and an advice will be given to the board of Boerenfluitjes.
THEORETICAL FRAMEWORK
Boerenfluitjes is defined as a social concept in the introduction. Therefore, in this theoretical framework, the focus will lie on the definition of a viable social concept and on the assessment of willingness to pay for such a concept.
A. Description of a viable social enterprise
A social enterprise is an organization that tries to enhance social wealth for the society as a whole (Hoogendoorn, 2011; Zahra, et al, 2009). In this section, the term social enterprise will be described in more depth and a framework that describes the creation of a new enterprise will be developed.
Social enterprise
“Social entrepreneurs play the role of change agents in the social sector, by:
Adopting a mission to create and sustain social value (not just private value);
Recognizing and relentlessly pursuing new opportunities to serve that mission;
Engaging in a process of continuous innovation, adaptation, and learning;
Acting boldly without being limited by resources currently in hand; and
Exhibiting a heightened sense of accountability to the constituencies served and for the
outcomes created” (Dees, 1998: 4).
A social enterprise is not necessarily a non-profit organization, often it has both social and economic goals. However, profit maximization is not the most important goal of a social enterprise, it rather pursues social change and improvement of social wealth (Zahra et al, 2009). What distinguishes a social enterprise from commercial enterprises is its social mission (Austin, Stevenson & Wei-Skillern, 2006; Hoogendoorn, 2011). “A mission is a statement of
the organization’s purpose, range of activities, character, and uniqueness” (Cummings and
Worley, 2008: 751). A mission statement is stable over time.
The creation of a new enterprise
Several frameworks to assess the creation of a new enterprise. For example the PCDO framework (Sahlman, 1996) and the framework of Gartner (1985). According to these two authors, creating new enterprises is a multidimensional phenomenon and enterprises differ between industries. Gartner’s framework exists of four interrelated dimensions that, together, describe the creation of a new venture. These four dimensions are 1) the entrepreneur; 2) the organization; 3) the environment; and 4) the process. The four elements of the framework of Sahlman (1996) are similar to those of Gartner (1985). These are 1) people (is similar to the entrepreneur); 2) context (comparable to the environment); 3) deal (comparable to the organization); and 4) opportunity (is similar to the process).
Entrepreneurial team
According to Gartner (1985), the entrepreneur is the one founding the enterprise. Sahlman (1996) broadens this definition by including everybody who participates in the new enterprise. These people can be inside and outside the enterprise. In many cases, the entrepreneur does not start the enterprise on its own, but he/she gets help from others. Therefore, in this research the focus will lie on the entrepreneurial team instead of on only the entrepreneur.
According to Hoogendoorn, Pennings and Thurik (2010) there is not much evidence that social entrepreneurs and commercial entrepreneurs differ a lot from each other. The most important difference between social entrepreneurs and commercial entrepreneurs is their mission. Social entrepreneurs are more need-driven than driven by an opportunity. Also their focus does not lie on the outcomes or the revenues, but on the process of doing something (Gartner, 1985). Since it might be difficult to measure the outcomes of a social enterprise, it should adopt effective mechanisms to monitor the enterprise.
To be able to value (social) entrepreneurs, Driessen and Zwart (2006) developed the e-scan. This scan identifies the entrepreneurial strengths and weaknesses of persons who wish to start an enterprise. The e-scan creates self-awareness. This helps individuals to better understand their capabilities and to be able to develop them to become a better entrepreneur. This will increase the chance of survival of the newly founded enterprise. The e-scan is based on the competence theory. A competence can be described as “something a person is very good at” (Driessen & Zwart, 2006: 2), and consists of four components: motivation, characteristics, capabilities and knowledge.
1) Motivation – this is an internally driven component and has to do with wanting something. It depends on ambition, motives and values of the individual. Three internally driven motives for starting a business can be found. These are: 1) need for autonomy; 2) need for achievement; and 3) need for power (Brockhaus, 1982).
2) Characteristics – traits of the entrepreneur. Characteristics which are common for an entrepreneur are 1) a need for achievement; 2) an internal locus of control; 3) a risk taking propensity; 4) a need for autonomy; 5) a need for power; 6) tolerance of ambiguity; 7) need for affiliation (the need to feel part of a certain group); 8) endurance (Brockhaus, 1982).
3) Capabilities – this is about being able to do something. Capabilities are more easy to learn and change than characteristics. Capabilities are more or less related to the company life cycle. In the early phase of the enterprise, market orientation, creativity and flexibility are important. And in the mature phase of the enterprise, leadership, organizing and planning, motivating and financial controlling are important.
4) Knowledge – refers to the more rational part of entrepreneurship. It is about knowing certain rules and principles. Also experience belongs to this component.
Process
The process describes the actions the entrepreneur undertakes to start an enterprise. The process exists of the following six behaviors: 1) location of the business opportunity; 2) accumulation of resources; 3) marketing of products or services; 4) production of the product; 5) building of the enterprise; 6) responding to government and society (Gartner, 1985).
The social mission of the enterprise, the organizational sustainability and the environmental dynamics all influence the process of opportunity identification simultaneously. Innovation is an important driver, together with pro-activeness and risk management. But they are bound by the social mission and sustainability of the organization and the environmental dynamics. Before the actual organization is created, resources should be acquired and a network should be created. Flexibility and learning as you go are key elements of success (Hoogendoorn, Pennings and Thurik, 2010).
Before a new product/service can be marketed, the enterprise should perform a market research, including a customer analysis. An important part of the customer analysis is the determination of the willingness to pay of the (potential) customers. The willingness to pay indicates the maximum amount of money customers are willing to sacrifice in order to fulfill their need for the product or service (Wertenbroch and Skiera, 2002). This points out that the willingness to pay also reveals if (potential) customers have a need for the product/service. When the willingness to pay of customers is high, it can be assumed that customers have a high need for the product or service, which potentially can be fulfilled by the new enterprise. Therefore, willingness to pay will be explored in more detail in the next section of this literature framework.
Organizational strategy
Gartner (1985: 698) defines organization as “the kind of firm that is developed”. According to
him this encompasses the generic strategy of the enterprise, the development of a new product or service and the acquirement of resources. Strategy is defined as “the actions an
organization takes to pursue its business objectives” (Wickham, 2006: 349). Structure is
defined as: “the predefined way in which individuals will relate to each other” (Wickham, 2006: 349). Since the activities Gartner (1985) describes are more about the strategy of a new enterprise than about its structure, this element is called organizational strategy.
The kind of enterprise that is developed is determined by the strategy of that enterprise. Porter (1980a) developed three generic competitive strategies an organization can follow:
1) Cost leadership – produce the product or service more cost-effective than its competitors. 2) Differentiation – serve individual customer needs by adding value to the product. In this
way a price premium can be charged.
3) Focus – specialize in a special market niche to gain a monopoly.
Figure 1: Generic competitive strategies Source of competitive advantage
Cost Differentiation
Scope Broad Cost leadership Differentiation
Narrow Focus
When an organization has chosen its strategy, it will try to grow. However, without resources, it will not be possible to grow. During start up, social enterprises are hindered by the lack of access to capital. Since the results of the social enterprise are hard to measure, it is difficult to get resources from investors. Often the social entrepreneur will creatively combine the available resources. Social entrepreneurs tend to mobilize local, already existing assets of their clients rather than using outside resources. These local assets are often not available for commercial entrepreneurs. (Hoogendoorn, Pennings and Thurik, 2010).
Environment
The environment is defined as “the situation surrounding and influencing the new enterprise” Gartner, 1985: 698). The entrepreneurial team has no influence on the environment. According to Hoogendoorn, Pennings and Thurik (2010) social entrepreneurship is highly responsive to and constrained by the environment surrounding the new enterprise. Therefore, it is important to analyze the environment carefully. Porter (1980a, 1980b) developed a model that can be used for this analysis, the five forces model. Porter states that organizations should not only focus on their competitors in the industry, but also on the consumers, the suppliers, the potential entrants and the possible substitutes. The strength of each of these forces is determined by the structure of the industry. The collective strength of these five forces should be considered when developing a strategy.
1) Threat of entry – new entrants bring capacity to the market and try to gain market share and resources from the existing enterprises. Existing enterprises are trying to prevent this by creating barriers to entry. The major sources of entry barriers are: economies of scale (on the supply-side as well as on the demand-side, network effects); product differentiation and associated switching costs; capital requirements; cost advantages independent of size (learning curve, technology, access to resource, location, reputation); access to distribution channels; government policy (can enhance or decrease the existing entry barriers); and the expected reaction of the existing competitors.
2) Bargaining power of suppliers – suppliers can use their power by raising prices or by reducing the quantity of sales. This results in a lower profitability of the industry. Supplier power is high when suppliers are more concentrated than the industry it sells to; when they deliver an unique or differentiated product; when there are high switching costs; when there are no substitutes for the products/services they deliver; when the pose a credible threat of integrating forward; and when they are not heavily dependent on the industry for its revenues.
4) Threat of substitutes – substitutes place a maximum on the price organizations can charge for their products/services, because when the selling price exceeds the price of substitutes, customers will switch to the substitute. Substitutes are threatening when their relative value for the customer is higher; when switching costs are low; and when there are changes in other industries that make them attractive substitutes. Ways to overcome these restrictions are by differentiating the product and by increasing the quality of the product/service.
5) Rivalry among competitors – when there are many competitors in an industry, intense rivalry is often high. Rivalry gets more intense when competitors have the same size and power; when the growth of the industry is slow; when there are no differentiation or switching costs; when the fixed costs are high; when the exit barriers are high; when the capacity is increased with large amounts; when the product is perishable; and when the competitors have very different strategies and cannot read each other’s signals well. Evenly important as the dimensions of competition is whether the competitors compete on the same dimension.
(Source: Porter, 1980a; 1980b)
Framework
From the discussion of the different elements above, the following framework can be deducted. This framework can be used to determine to viability of a new enterprise.
(Adapted from: Gartner, 1985; Sahlman, 1996) Entrepreneurial
team
Environment Organizational
strategy Process
Figure 3: Framework to describe a new social enterprise Figure 2: Five forces model
Threat of new entrants
Viability
These four concepts together (the entrepreneurial team, the process, the organizational strategy and the environment) can be used to determine the viability of an enterprise. An enterprise is considered viable when it is able to achieve its mission, to create value for the society as a whole. As what outlined in the part about the organizational strategy, viability is, in part, determined by the willingness to pay of potential customers. After all, when customers are not willing to pay for the products or services a new organization wants to offer, it will be difficult to survive. Therefore, the remainder of this theoretical framework will focus on the assessment of willingness to pay.
B. Assessment of willingness to pay for a social concept
Willingness to pay (WTP) is defined as the maximum amount a customer is willing to sacrifice to fulfill a particular need. It is a subjective measurement that enables a customer to express the perceived value of a good or service in monetary terms (Wertenbroch and Skiera, 2002; Le Gall-Ely, 2009). When an enterprise has developed a new service, it is important to determine the WTP of (potential) customers before it is brought to the market. The WTP of (potential) customers can be used as an indication of the maximum price that can be charged and of the need that exists for the service. The main factors that influence the WTP of customers for a new service are:
The need for the service – “A need is a physiological or psychological deficiency that
arouses behavior. A need can be strong or weak and is influenced by environmental factors. Needs can vary over time and place” (Buelens, van den Broek and
Vanderhuyden, 2006: 177). A need is not only related to the characteristics of the service, but also the service before and after the sale and the selling points have to be taken into account. Potential customers have to have a need for the service. When this is not the case, they are not willing to pay anything for it (Abell and Hammond, 1979).
The perceived importance of the service – a distinction can be made between the needs of
respondents. There are needs that need to be fulfilled immediately, and there are needs that do not have to be fulfilled. Customers are generally willing to pay more for a need that has to be fulfilled immediately than for a service that is perceived as less important (Aaker and McLoughlin, 2007; Abell and Hammond, 1979).
The amount of information that is available – the scenario is important for the quality of
the results. This implies that the more information about the service the respondent receives, the better they can determine their WTP (Cummings and Taylor, 1999; Mitchell and Carson, 1989)
The income level of the respondents – customers with a high income have more money
they can spend freely after all fixed costs are paid than customers with a lower income. Therefore, customers with a higher income are often willing to pay more than respondents with a lower income
The method that is used to measure willingness to pay – many methods to measure
Methods to measure willingness to pay
The WTP can be measured based on two types of data, actual market data and survey data. Actual market data are for example transactions data. This type of data is relatively easy and inexpensive to obtain. But it does not reveal the true WTP of the customer, since it only show what the customer buys. Actual market data are not available for a new concept, since it has not sold any products or services yet. Survey data, also called real data or hypothetical data, is suitable for existing as well as for new products and services. Since this research is aimed at the measurement of the WTP for a new service, the focus of this research will lie on the methods that utilize survey data.
The most frequently used methods that utilize survey data and that are appropriate to measure the willingness to pay for a new concept are the open ended-questions format; the conjoint analysis; the contingent valuation (Wertenbroch and Skiera, 2002; Miller, Hofstetter, Krohmer and Zhang, 2011).
These methods will be described shortly below. After this description a comparison of the different methods will be made. And, based on this comparison, one method will be chosen as input for the development of a questionnaire.
Open-ended questions format
The open-ended questions format is a direct approach to determine the WTP of customers for all types of services. Several service options are stated and customers are asked to indicate how much they are willing to pay for each of these options. It is a quick and straightforward approach. A disadvantage of this approach is that it is hypothetical, it does not oblige the respondent to buy anything. This may cause customers to not take the relation between price and quality or the constraints that would affect his/her choice in the real world, for example their budget, into consideration. A hypothetical approach can lead to overstatements of the value the service has to the respondent and to overstatements of the WTP. Another disadvantage is that people have the tendency to think concretely, and therefore, will often indicate tangible characteristics as more important than intangible characteristics (Alsem, 2007).
Conjoint analysis
“A conjoint analysis (CA) is a technique that is used to determine the relative importance customers attach to salient attributes and the utilities they attach to the level of attributes”
(Malhotra, 2008: 674). According to Green and Rao (1971) CA determines the ordering of a dependent variable indirectly, according to the joint effect of two or more independent variables. It deducts the willingness to pay from the choices customers make. This makes it a realistic approach.
Advantages of this technique are that it is simple to answer and that it is possible to include interactions between the characteristics of the service. It is also possible to simulate different market scenarios and to make predictions about the market share that may be gained (Alsem, 2007; Backhaus, Wilken, Voeth and Sichtmann, 2005; Cattin and Wittink, 1982; Wittink and Cattin, 1989).
Assuming the most important attributes of a service can be identified is a limitation of the CA. Also, it is not possible to include all attributes in the study and it assumes that the attributes chosen are the most important ones. Another limitation is that CA cannot be used when attributes are soft or highly correlated. There is the possibility of hypothetical, informational (when the questionnaire points the respondent in a particular direction) and representativeness (when only a few characteristics of a subject are revealed and the respondent links this to a stereotype) biases (Malhotra, 2008; Cattin and Wittink, 1982; Le Gall-Ely, 2009; Wittink and Cattin, 1989). According to Ding (2007) a disadvantage of the CA is that respondents are not incentive aligned to report their true WTP. He showed that data collected in a CA has a weaker external validity than when respondents were incentive aligned. This may be due to an unrealistic task or to the fact that respondents may be unable or unwilling to answer the questions correctly.
Contingent valuation
The contingent valuation (CV) is a method that tries to gather the WTP of respondents by means of a questionnaire. The values that are revealed are dependent on the simulated market that is presented in the survey. An advantage of the use of questionnaires is that it is possible to include all kinds of questions, for example questions about socio demographic variables. Next to this, it is time and cost efficient method, since all respondents are asked the same questions and it is possible to distribute the questionnaire via the Internet. To date, this method is mainly used to determine the WTP for public goods. However, it is applicable to all types of services, public as well as private; existing as well as new (Green and Wind, 1975; Hoevenagel, 1994; Mitchell and Carson, 1989).
Comparison of the different methods
As the previous part explained, there are many different ways to measure the WTP of (potential) customers. The most important advantages and disadvantages of each method are summarized in table 1.
Table 1: The advantages and disadvantages of the different methods to measure willingness to pay Method to measure WTP Advantages Disadvantages Open-ended questions format Direct measurement of WTP
Quick and straightforward
approach
For all types of products and
services
Strategic, hypothetical,
informational and representativeness bias
Respondents do not think about
the relation between price and quality
Respondents have the tendency
to think concretely
Conjoint analysis
For all types of products and
services
Indirect measurement of WTP
Uses combinations of attribute
levels
Can be used in new service
development
Rank-ordered input yields
interval-scaled output
Interactions between
characteristics are included
Possible to simulate different
market scenarios
Hypothetical, informational,
representativeness biases
Subjective evaluations
Used attributes are determined
by the researcher
Assumes that all important
attributes can be identified and included
Possibility of information
overload
Data collection may be difficult
Unusable for soft attributes
No incentive-alignment
Contingent valuation
Time and cost efficient
Indirect measurement of WTP
Appropriate for all types of
products and services
Possible to include socio
demographic variables
Strategic, hypothetical,
informational and representativeness bias
Dependent upon the scenario
Mainly used with public goods
Conclusion
Looking at the different methods that exist to measure WTP, each method has its advantages and disadvantages. So, when choosing a method, it is important to keep the purpose of this research in mind. In this research, the objective is to determine the viability of a new social enterprise.
The most important difference between the open-ended questions format, the conjoint analysis and the contingent valuation is that the open-ended question format measures the WTP directly while the CA and CV use an indirect measurement. Since it is hard to make a representation of something that does not yet exists, it is also difficult to determine the WTP for this. Therefore, it is better to use a method that measures WTP indirectly. An indirect method deducts the WTP from the choices customers make. This makes it more realistic, because in a real purchase situation, the respondent also has to evaluate a bundle of attributes instead of only the price of the service (Green and Rao, 1971). The open-ended questions format is therefore not the appropriate method for this research.
Backhaus, et al., (2005) compared the CA method with the CV method. They found a significant difference in results, the CA yields a higher WTP than the CV. A possible explanation for this difference is that it is easier to answer strategically in the CV than in the CA method. Therefore, it seems likely that the WTP in the CA is overestimated (Backhaus, et al., 2005; Wertenbroch and Skiera, 2002). Another disadvantage of the CA is that it is not suitable when the attributes of the study are soft. And attributes of a (new) service are often soft and intangible.
Advantages of the CV method over the CA method are that the researcher does not have to identify all attributes of the service, but the service is described in a scenario. And there is room to ask for social-demographic data in the CV method, which is useful to describe the sample. Therefore, the CV is preferred over the CA to measure the WTP for a new service.
In the next section, the contingent valuation will be described in more detail and it will be adapted to correct for the disadvantages that are identified in the previous section.
Contingent valuation
The contingent valuation makes use of a questionnaire to determine the WTP of respondents. Because different studies all have different purposes and ask for different kinds of questions and questionnaire designs, there is not a pre-specified design for the questionnaire. However, there are three elements that should be present in the questionnaire:
1) A scenario or description of the hypothetical service the respondent is asked to value. This scenario must be understandable and meaningful to the respondent and it has to be free of incentives and assumptions in order to prevent biases. The goal of the scenario is to develop a clear picture of the /service.
2) A mechanism for deducting the respondent’s willingness to pay for the hypothetical service that is being valued.
The sample that is used in the CV has to be selected randomly. To be able to generalize the results for the population as a whole, adjustments for non-responses and for respondents who deliver an uncompleted questionnaire or a questionnaire of poor quality have to be made (Mitchell and Carson, 1989).
Adaptation of the contingent valuation
In the description of the contingent valuation, several disadvantages of the method are described. To make sure this research does not suffer from those disadvantages, the method is adapted to correct for them.
The disadvantages that were mentioned in the description are:
Strategic, hypothetical, informational and representativeness bias;
Dependent upon the scenario
Non-response error
Cummings and Taylor (1999) found that a good way to remove the hypothetical bias is the use of cheap talk in the scenario. With the use of cheap talk, the respondent is made aware of the bias and the respondent is asked to answer the questions if it were a real situation. Their research shows no significant difference between the outcomes of hypothetical situations with cheap talk and real-life situations with actual payments.
Use of cheap talk can also reduce the other three types of biases. Because respondents are provided with all the information they need to complete the questionnaire, informational biases will be reduced. Furthermore, the possibility of strategic-overestimation is reduced because respondents answer if they were in a real purchase-situation. And finally, when the scenario describes the subject well, also the representativeness bias will be reduced.
This conclusion is also drawn by Mitchell and Carson (1989). According to them, the wording of the scenario is very important. When there is not paid enough attention to this, customers can be directed into a certain direction. If the scenario and the rest of the questionnaire are carefully constructed, this will motivate respondents to answer truthfully. Strategic behavior requests a lot of information, and acquiring all these information can be very expensive. Therefore, it is often easier and less expensive to answer truthfully. Next to this, the structure of the CV questionnaire gives the impression that many respondents are asked to complete the questionnaire. This implies that over- or understating the WTP will not significantly influence the outcomes of the study.
The wording of the scenario is very important in the CV study. It influences the outcomes of the questionnaire, because it determines how much and which information respondents receive and it can influence the chance on biases.
The disadvantage of non-response can be reduced by using one of the following methods to adjust for non-response:
Trend analysis – the researcher tries to distinguish a trend between early and late
respondents. This trend is then projected on non-respondents to estimate their response to the questionnaire.
Weighting – adjust for non-response by assigning weights to the data depending on the
response rates. Weighting can correct for the differential effects on non-response, but it destroys the self-weighing nature of the sample. Information from the Central Bureau of Statistics in the Netherlands can provide information regarding the parameters for weighting.
Imputation – adjust for non-response by assigning non-respondents the characteristics of
respondents based on similarity to variables that are available for both respondents and non-respondents.
(Malhotra, 2008; Mitchell and Carson, 1989)
METHODOLOGY
The methodology consists of three parts, in the first part, Boerenfluitjes is introduced. Second, social entrepreneurship and the measurement of the viability of a social entreprise is outlined, and finally, the measurement of willingness to pay is described.
Introduction of Boerenfluitjes
Boerenfluitjes started on Twitter three years ago as a communication network that connects customers, producers and partners in the food chain with each other. The goal of Boerenfluitjes is to bridge the gap between farmer and customer. It tries to achieve this by asking farmers questions about their products, and passing the answers through to its followers, without a judgment. There are a lot of farmers that tell about their daily live and work, the production of our food on social media.
Boerenfluitjes believes that transparency in the (local) food chain may ensure that customers make more informed choices in the supermarket. It may also ensure that farmers can develop a business in which they can market the added value of their product. In this way Boerenfluitjes wants to contribute to making the food chain more sustainable.
After one year, it was time to expand their activities, and to add money-generating activities to their activities on Twitter. Boerenfluitjes decided a good way to close the gap between customer and farmer is to bring the customer to the countryside. In this way, customers get the opportunity to meet the farmer in person, to ask him/her questions and to see the production process with their own eyes. Boerenfluitjes started to develop daytrips to the countryside.
For this purpose, Boerenfluitjes B.V. was founded. Three parties are involved in this limited company. The first party is Ester van Aalst, who is the founder of Boerenfluitjes on Twitter. Ester van Aalst realized that she likes the countryside, but she has no clue what happens over there. She wants to know the story behind her food, and therefore she started Boerenfluitjes. Since Ester van Aalst realized she does not have all capabilities necessary to start and manage an enterprise on her own, she approached to other parties, Praedium and Countus, to participate in the development of Boerenfluitjes. Both organizations are active in the agricultural sector and have large networks in this sector.
Praedium is an organization that wants to contribute to strengthening the regional (rural) economies and making them more sustainable. One way of doing this is by exploiting natural cemetery Heidepol at Arnhem. This cemetery is located on an estate near national park the Hoge Veluwe. At this cemetery, you can choose where you want to be buried, under a tree, or in the open field. Another example of an enterprise that Praedium is working on is a cooperative farm. On this farm, several citizens buy a farm together and hire a farmer who exploits the farm. The produced products are distributed among the owners of the farm.
Countus is an organization specialized in consultancy and accountancy. They focus on the agricultural market as well as on small and medium sized firms (MKB). For these clients, Countus produces financial statements, does tax declarations and business consultants can give them advice about financial, fiscal and juridical issues.
There are a lot of plans to generate money for Boerenfluitjes. Among others, a partnership with Gijs Streekproducten; a collaboration with Vrouwen van Nu to develop mutual activities and to publish them in their magazine; the organization of an event together with restaurant Fifteen from Jamie Oliver; and The Greenery wanted to introduce vegetables with stickers of Boerenfluitjes.
Boerenfluitjes also generates a lot of attention in the media (Telegraaf; agricultural magazines; TV West; Radio 1; and a blog on the Milk Story). Actions that Boerenfluitjes organized are ‘koop waar het klopt’ where customers were encouraged to look for the story behind their food and to share their experiences on Twitter; and ‘dag van de burger’, this day is devoted to the interests of citizens, but Boerenfluitjes encouraged customers to look for the origin of their hamburger.
Boerenfluitjes started with the development of the website and by attracting farmers to become members of Boerenfluitjes. Members of Boerenfluitjes get space on the website, so they can show their organization and promote activities they organize, get access to the communication platform and network of Boerenfluitjes and receive merchandise from Boerenfluitjes. To attract farmers, a Boerenfluitjesboerendag was organized in February 2012. This day was a success and about 15 farmers signed up as member of Boerenfluitjes (see appendix A for a summary of the documents used for the introduction of Boerenfluitjes).
A. Measurement of social entrepreneurship and the viability of an enterprise
In the theoretical framework, a framework to describe the viability of a new social enterprise was developed. This framework will be used to determine if Boerenfluitjes is a viable social enterprise. The four components of the framework (the entrepreneur, the process, the organization and the environment) will be described based on observations of the researcher and a content analysis of the available documents about Boerenfluitjes will be done.
1) The entrepreneurial team – the entrepreneurial team of Boerenfluitjes consists of three parties, Ester van Aalst, Countus and Praedium. The e-scan of Driessen and Zwart (2006) will be applied on these three parties.
2) The process – the process of founding Boerenfluitjes will be described based on the six behaviors described by Gartner (1985).
3) The organizational strategy – the building of the organization will be analyzed based on the generic strategies developed by Porter (1980a).
4) The environment – the five forces model developed by Porter (1980a; 1980b) will be used as framework to describe the environment surrounding Boerenfluitjes.
Content analysis
A content analysis is “the objective, systematic, and quantitative description of the manifest
content of communication” (Malhotra, 2008: 205). This kind of analysis includes both
Observations
Observation is “the recording of behavioral patterns of people, objects and events in a
systematic manner to obtain information about the phenomenon of interest” (Malhotra, 2008:
202). Observations are done by a researcher. The researcher has no contact with the subject under study, he/she merely records what happens. In this research, the observation will be unstructured, which means that the observer does not specify what has to be observed and how these observations should be recorded. But the researcher observes everything that seems relevant. This is an appropriate method when flexibility is needed. The observations will be done during meetings of the Boerenfluitjesteam (natural observation). Advantages of observation are that actual behavior of the parties involved can be analysed. Because the observer is a neutral person, he/she is not involved in the day-to-day business of the enterprise, this means the results can be regarded as objective. Since there is no contact between the observer and the subjects under study, there is no reporting or strategic bias. Disadvantages of observation are that it may be time-consuming and expensive and that it is not possible to determine the underlying reasons for certain behavior. Because observations can provide valuable information, the best way to use it is as a complement to other forms of data gathering (Malhotra, 2008).
B. Measurement of willingness to pay
To determine the willingness to pay of potential customers for the services Boerenfluitjes wants to develop, a questionnaire will be developed based on the contingent valuation (see appendix B). It is not possible to determine the willingness to pay of potential customers by observing them or by doing a content analysis, since the service does not exist yet.
Questionnaire
Using a questionnaire has several advantages and disadvantages. The most important advantages are: 1) low cost; 2) the person administering the questionnaire needs minimal training; and 3) it is easy to reach many people. The most important disadvantages of using a questionnaire are: 1) chance of a low response rate; 2) respondents have little or no contact with the researcher; 3) the researcher has no control; 4) questionnaires may return incomplete; and 5) data has to be re-keyed and entered into analysis software.
Most of these disadvantages can be overcome by distributing the questionnaire through the Internet. With a web-based questionnaire, 1) the researcher is able to control the order of the questions; 2) only completed questionnaires are shown in the data; 3) it is easier to reach people who are interested in a specific issue; 4) it is easier to contact people who are otherwise difficult to reach; and finally 5) it is possible to transfer the data directly into analysis software, so rekeying is unnecessary. Disadvantages of a web-based questionnaire are that 1) the response rate is usually lower than for postal questionnaires; and that 2) it is not possible to control the sample and to track the non-response. The researcher cannot control how many people are reached, and those who do not have access to a computer are automatically excluded (Wright, 2005; Jones, Murpy, Edwards & James, 2008).
data entry); and the good quality of the data, since data collection is automated and the control is larger.
Mitchell and Carson (1989) found that a questionnaire that is developed according to the contingent valuation method should consist of the following elements:
1) A scenario or description of the hypothetical product or service the respondent is asked to value. This scenario must be understandable and meaningful to the respondents and it must be free of incentives and assumptions to prevent biases.
2) A mechanism for deducting the respondent’s willingness to pay for the hypothetical product/service that is being valued.
3) Questions about the socio demographic variables of the respondents, like age, gender, education, income, etc. This information is used to explain the estimated WTP.
In the literature review several disadvantages of the contingent valuation method and the ways to overcome these disadvantages were described. These will also be taken into account in the development of the questionnaire. The most important disadvantages and their solutions were:
Disadvantages Solutions
Biases Use of cheap talk in the scenario
Dependent upon the scenario Pilot testing; use of open-ended questions to
provide room for remarks
Non-response error Imputation
Biases
As was described in the theoretical framework, the disadvantage of bias can be solved by the use of cheap talk in the scenario. In the scenario of the questionnaire about Boerenfluitjes, respondents are informed that anonymity is secured and that their own opinions and attitudes are important. There are no right or wrong answers. In this way, respondents are stimulated to answer the questions with regards to their own situation.
Furthermore, in the introduction of the daytrips, respondents are given all information necessary to be able to answer the question. This will reduce the chance that respondents are going to fill in details of the daytrips themselves, and will reduce informational bias.
Dependency upon the scenario
The wording of the scenario is important for the outcomes of the research, and to prevent biases from occurring. Therefore it is important to pay a lot of attention to writing the scenario.
After the questionnaire is developed, a pilot test will be performed among experts from the field and among members of the Boerenfluitjes-dreamteam. These experts will complete the questionnaire and will give their opinions. Based on this, the questionnaire will be adapted to make sure that the questionnaire will lead to the desired results.
the end of the questionnaire the respondent will be given the opportunity to give some final remarks and suggestions.
Non-response error
Because of the distribution of the questionnaire, through Internet, it is difficult to define the sample, since it cannot be known who gets the message and who not. However, it is possible to make use of imputation. This means that the non-respondents are assigned characteristics of respondents based on variables that are available for both groups. When a respondent fails to fill in the questionnaire completely, the answers will be complemented with the answers of other respondents with the same characteristics who did complete the questionnaire correctly. In this way, an adequate size of the sample can be assured.
Development of the questionnaire
Input of the questionnaire are the main factors that are influence the willingness to pay. These factors are 1) the need for the service; 2) the perceived importance of the service; 3) the amount of information that is available; 4) the income level of the respondents; and 5) the method that is used to measure the willingness to pay.
In the questionnaire first some introductory questions will be asked about daytrips and the Dutch country side, this will give the researcher an idea of how much value respondents attach to having the opportunity to go on a daytrip (measurement of the perceived importance of the service). After these introductory questions, three possible daytrips will be described (the scenarios) and the respondent is asked if he/she would be interested in this daytrip and what the maximum price is that he/she is willing to pay for it. These questions measure the need for the service and the willingness to pay. Finally, some general questions about the respondent, like gender, age, income, education will be asked. This measures the level of income of the respondent and this information will be used to describe the population. The influence of the amount of available information will be measured by developing two different versions of the questionnaire. The first version will have long descriptions of the daytrips and version two will have shorter ones. The method that is used to measure willingness to pay is already chosen in the literature framework, the contingent valuation method. Therefore, this factor will not be measured in the questionnaire.
Most questions in the questionnaire will be multiple choice. For respondents, it is sometimes difficult to state the maximum price they are willing to pay, especially when they are unfamiliar with the service. They normally decide whether they want to buy something based on a known price. Therefore, it is decided to use multiple choice questions to measure the willingness to pay of the respondents (Kaas and Ruprecht, 2006). However, because it is believed that it is important for respondents to explain their answer, they will be asked for an explanation after some multiple choice questions to give respondents the opportunity to do so. This method is chosen, because it is straightforward and efficient and it has the least biases.
Target population
The target population of this research exists of people who are interested in the true story behind their food and who are already familiar with Boerenfluitjes.
them. When the people in the target group are not interested in the daytrips, it is not likely that the rest of the Netherlands will be. Furthermore, the people who are in this target group are easy to reach, because they already follow Boerenfluitjes on Twitter and/or Facebook.
The minimum sample size is 100 respondents. When the sample has this size, it is possible to attach significant conclusions to the conclusions (Malhotra, 2008).
In total 115 respondents filled in the questionnaire and 80 of those respondents filled in the questionnaire completely. Therefore only the answers of those 80 respondents will be used when analysing the results. The average age of the respondents is 31,51 years old and ranges between 15 and 58 years. 31 per cent of the respondents are male and 69 per cent of the respondents are female. The main part of the respondents already knew Boerenfluitjes before they filled in the questionnaire (65 per cent). 58 per cent of the respondents had finished an HBO or WO education. The other 42 per cent finished an HAVO/VWO or MBO education. The largest part lives in the north of the Netherlands (Friesland, Groningen and Drenthe, 44 per cent). But the other part of the respondents is dispersed over the other nine provinces. Also the income level of the respondents is almost equally divided between the different answer options. This gives a varied sample (for a more detailed presentation of the descriptives and the target population see appendix C and D).
Distribution
The questionnaire will be distributed through the Internet by Boerenfluitjes and members of the Boerenfluitjesteam and the message will be repeated several times, until enough respondents completed the questionnaire. In this way, the respondent will be reminded about the questionnaire each time, and this will lead to a higher response rate (Cooper and Schindler, 2006).
The main distribution channel will be Twitter. Twitter is a social network site on which members can send messages of up to 140 characters. Members can follow each other, read each other’s messages and react on one another on their timeline (public) or in a direct message (private).
According to Boyd and Ellison (2008: 211) a social network site is “a web-based service
that allows individuals to 1) construct a public or semi-public profile within a bounded system; 2) articulate a list of other uses with whom they share a connection; and 3) view and traverse their list of connections and those made by others in the system.” Mostly, members
of social media sites connect to people who are already part of their offline social network, or to people who have the same interests.
With an Internet survey it is possible to reach samples that are identified by their interests on Internet. This is appropriate for Boerenfluitjes, since the target group exists of people who are already familiar with Boerenfluitjes, there is a high chance they will already follow Boerenfluitjes on Twitter. Therefore, distribution through Twitter is the only way to get access to this sample.
An important characteristic of distribution through the Internet is self-administration. The downside of this is that it does not represent the whole population, but that it only attracts people who are already interested in the arrangements Boerenfluitjes wants to offer. Because of this self-selection, it is expected that the estimates of the WTP may be biased and higher than the WTP of the population in general. But since this is known beforehand, it is possible to anticipate on it. For example by approaching certain people and by adjusting the weights that are given to the different respondents to make sure the distribution corresponds to the population as a whole.
The lack of personal contact between the interviewer and the respondent can also be a disadvantage. The interviewer cannot read the body language of the respondent. Therefore it is important to also have open questions in the questionnaire, so the respondent has the opportunity to explain his/her answers. Next to this, the interviewer is unable to give extra explanations or to motivate the respondent with their physical presence. When this is not possible, there is the tendency to use long descriptions and try to explain everything in the questionnaire, but this will scare respondents off. It is better to keep the scenarios short and simple. When the respondent is familiar with the subject under study, the lack of personal contact is often not a problem. Therefore, an Internet survey is preferred over in-person and telephone questionnaires (Cooper and Schindler, 2006; Mitchell and Carson, 1989; Nocella, Hubbard and Scarpa, 2009; Patino, Pitta and Quinones, 2012).
Reliability
A measure is reliable when it produces consistent results on repeated measurements, even when time and conditions change. This means the measure is free of random error. The reliability of a measure can be assessed by determining the proportion of systematic variation in a scale. Methods to determine this are 1) test-retest; 2) the use of alternative forms of the measure; and 3) testing the internal consistency (Cooper & Schindler, 2006: 321-323; Malhotra, 2008: 284-286).
The method that best fits this study is the internal-consistency test. When one of the other two methods is used, the same group of respondents has to be reached twice. Since the questionnaire in this study is distributed through the Internet and makes use of self-administration, it is difficult to reach the same group of respondents twice. Therefore it is not possible to use test-retest or alternative forms of the measure to test the reliability.
The internal consistency test is used when several items are added together to measure a scale. It determines the degree to which these items all measure the same underlying attribute. There are three ways to measure the internal consistency of a scale:
2) Cronbach’s alpha – the average of all possible split-half coefficients resulting from different ways of splitting the items. The internal consistency is sufficient when the alpha is above 0,7. The coefficient tends to increase when the number of items on the scale increases.
3) Inter-item reliability – an alternative measure for the Cronbach’s alpha. In inter-item reliability, the average of inter-item correlations is calculated. This method gives a better understanding of the reliability when short scales (shorter than 10 items) are used, since in these cases the Cronbach’s alpha is usually quite low. The optimal range for the inter-item correlation is between 0,2 and 0,4.
(Briggs & Cheek, 1986; Cooper & Schindler, 2006; Malhotra, 2008)
In the questionnaire (see appendix B) there is made use of two scales, one scale has seven items and the other one has four. This means that the scales are quite short and this makes the inter-item reliability test the best method to measure the reliability of the questionnaire. Cronbach’s alpha is not the right method since it tends to work best with large scales (more than ten items) and the split-half reliability test is not the right method because it is too much dependent on the way the scale is split in halves.
The inter-item reliability test is used to test the reliability of two scales that are composed of more than one item.
Booking a daytrip – for this scale the inter-item reliability coefficient is 0,202. This means
that the sale is assume to be reliable.
Going on a daytrip – for this scale the inter-item reliability coefficient is 0,249. This
means that this scale is also assumed to be reliable. See appendix E for a detailed description of the results.
This means that it can be concluded that these two scales are reliable and produce consistent results. This is important, because reliability is a necessary condition for validity (Cooper and Schindler, 2006). Since the items of the two scale measure the same underlying attributes, the mean of these items is taken en composed into two new variables 1) Booking a daytrip and 2) Going on a daytrip.
Validity
Validity determines if a construct measures what is should measure. In this paper the questionnaire is valid when it measures the true willingness to pay for the arrangements of Boerenfluitjes. Validity can be divided into internal and external validity.
Internal validity – measures if the observed effects on the dependent variables are caused
by the changes in the independent variables and not by other variables that are not included in the test. Without internal validity, it is not possible to attach conclusions to the results of the study. Internal validity can be divided into three, interrelated, subcategories: 1) Content validity – or face validity. It is a subjective, but systematic evaluation of how
well the attribute is covered in the content of the scale. To be able to measure this, it is important that there is agreement about the elements that constitute adequate coverage. Content validity can be judged by the researcher or by a panel of experts, who can pilot test the questionnaire before it is distributed.
3) Construct validity – determines which construct or characteristic the scale measures. Both the theory and the measuring instrument are considered.
a) Convergent validity – the degree to which the scale correlates positively with other measures of the same construct.
b) Discriminant validity – the degree to which the scale does not correlate with other measures of different constructs, to which it ought to differ.
External validity – the possibility to generalize across persons, settings and time. When
interactions with other relevant variables are not taken into account, the external validity is endangered. External validity can be guaranteed by using questions that have already proven their validity in other academic articles.
(Cooper & Schindler, 2006: 318-321; Malhotra, 2008: 225-226)
The content validity of the questionnaire can relatively easy be tested by pilot testing. A panel of experts, composed of an assistant professor; friends; members of the Boerenfluitjes-team and employees of Countus are asked to complete the questionnaire and to judge it. After this pilot test, the questionnaire was slightly adjusted and hereafter approved to be content valid.
Mitchell and Carson (1989) researched the criterion and construct validity of the contingent valuation method. They concluded that it is difficult to measure both the criterion and the construct validity, since this method is based on testing the willingness to pay in a hypothetical market. This means that it concerns a new market, for which no market measure is available yet. However, the authors concluded that the method is criterion and construct valid, because for services which are familiar to respondents, the hypothetical and simulated markets show similar results. In this particular paper the main topic of the questionnaire are the daytrips Boerenfluitjes wants to offer. Since respondents are familiar with daytrips, it is believed that the questionnaire is criterion valid.
To guarantee external validity, the questionnaire that is developed for this study will, for a large part, be composed of questions that are already used by other academics and that have already proven their external validity. Next to this, the questionnaire will exist mainly of multiple choice questions, because this will point respondents into a certain direction and leaves not much room for own interpretation of the questions. But there will be asked for explanations of the answers on the multiple choice questions.
These results show that the contingent valuation method and the questionnaire that is developed for this particular paper will give results that are internally as well as externally valid.
Willingness to pay
The willingness to pay is asked three times, one time for every daytrip described in the questionnaire. To be able to test if the independent variables have an influence on the willingness to pay, it would be convenient to combine these three variables together into one (new) variable.
Here, Cronbach’s alpha will be used. Cronbach’s alpha is the average of all possible split-half coefficients that result from different ways of splitting the items. The internal consistency is sufficient when the alpha is above 0,7.
As can be seen in appendix F the Cronbach’s alpha of the three variables that measure willingness to pay is 0,797. This means that they are internally consistent and that it is appropriate to combine them into one variable.
The second step is to combine the three variables into one new variable. This will be done by means of a factor analysis. This is a technique that is used to discover patterns among the variables to determine if there exists an underlying combination of the original variables (the factor) that can summarize the original set (Cooper and Schindler, 2006: 590). The aim of a factor analysis is to reduce the data. The method that will be used to perform the factor analysis is the Principal Component Analysis. The outcome of this method is a set of factors that are linear and are not correlated with each other. The linear combinations of the factors account for the variance in the data as a whole. The number of factors (new variables) will be determined based on:
Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy – an index that is used to
check if a factor analysis is suitable. Values > 0,6 indicate that factor analysis is suitable.
Bartlett’s test of sphericity – tests the hypothesis that the variables in the population are
uncorrelated. Values should be significant (e.g. < 0,05) for factor analysis is considered suitable.
Eigenvalue – the total variance explained by each factor. This value should be > 1,0.
(Cooper and Schindler, 2006; Malhotra, 2008).
Inspection of the correlation matrix (see appendix F) revealed that the Kaiser-Meyer-Oklin value is 0,697, which exceed the recommended value of 0,6 and that the Barlett’s Thest of Sphericity reached statistical significance (< 0,001). Therefore the factorability of the correlation matrix is supported.
Other interesting results
Boerenfluitjes is interested in some other results that follow from the questionnaire. The relations that will receive some more attention are:
If the need for the service of the respondents influences their willingness to pay.
If the income level of the respondents influences their willingness to pay.
If the amount of information of the respondents influences their willingness to pay.
If the extent to which respondents believe they have a good image of the Dutch country
side influences their willingness to pay.
If the extent to which respondents are interested in the origin of their food influences their
willingness to pay.
If age has an influence on the interest in the origin of food.
These relationships will be researched into some more depth using one-way analysis of variance (one-way ANOVA). In one-way ANOVA the mean scores of two or more groups are compared. The null hypothesis that is used is that the means of the populations are equal. The significance level that is used is 0,05.
