Business possibilities due to the creation of ecological value through the use of electric cars

(1)

Bachelor Thesis Supervisor: Willem Dorresteijn

Business possibilities due to the creation of

ecological value through the use of electric cars

A quantitative research about the influence of ecological value on loyalty intentions and willingness to pay in businesses offering transportation and delivery services.

Boris Kreecke 10784039

Word count: 10.691

27 June

17

(2)

Statement of Originality

This document is written by, student, Boris Kreecke who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract ... 4 1.Introduction ... 5 2. Literature review ... 6 2.1 Green consumption and ecological value ... 6 2.2 Consumer loyalty and willingness to pay ... 7 2.3 Transportation and delivery services ... 8 2.4 Conceptual framework ... 10 3. Methods ... 11 3.1 Data gathering ... 11 3.1.1 Questionnaire ... 11 3.1.2 Sample ... 12 3.2 Data analysis ... 13 3.2.1 Quantification ... 13 3.2.2 Testing on significance of variables ... 13 3.2.3 Testing relations between independent variables ... 13 4. Results ... 14 4.1 One sample t-test on variable statements ... 14 4.1.1 Green consumption ... 14 4.1.2 Emotional value ... 15 4.1.3 Social value ... 16 4.1.2 Ecological value ... 18 4.1.5 Loyalty intentions ... 19 4.2 One sample t-test on willingness to pay questions ... 20 4.2.1 Public transport services ... 20 4.2.2 Food delivery services ... 21 4.2.3 Taxi services ... 22 4.2.4 Moving car rental services ... 24 4.3 Paired sample t-tests on difference between willingness to pay questions ... 25 4.4 Test on relations between variables ... 27 4.3.1 Relation between green consumption and emotional value ... 27 4.3.2 Relation between green consumption and social value ... 28 4.3.3 Relation between emotional value and ecological value ... 30 4.3.4 Relation between social value and ecological value ... 31 4.3.5 Relation between ecological value and loyalty intentions ... 32 4.3.6 Relation between loyalty intentions and willingness to pay ... 33 5. Discussion ... 35 5.1 implications ... 35 5.2 Limitations ... 36 5. Conclusion ... 37 References ... 38 Appendix 1 ... 41

(4)

Abstract

This research examines to what extent businesses that offer transportation and delivery services can benefit from using electric cars. Green products create ecological value for customers which results in higher willingness to pay. However, this does not hold for the automotive industry since there are too many technical and social barriers to switch from fossil fuel driven to electric cars. Nonetheless, there are many technical options for electric driving. Businesses can potentially be better able at coping with the disadvantages of electric driving through economies of scale. A transportation or delivery service performed with electric cars is a form of green consumption. Therefore, this research examines if people are willing to pay more for these services. This is examined with an online questionnaire among young adults living in Amsterdam and surrounding areas. The collected data through the questionnaire is statistically analysed with SPSS. The results indicate that people are willing to pay more than five per cent more for these types of services when performed with electric cars. This information could be used to create new business possibilities. This is socially relevant, because electric driving can contribute to limiting the consequences of climate change.

(5)

1.Introduction

In the era we live in it is widely scientific acknowledged that global warming is a treat for humanity caused by anthropogenic activity, mostly from energy production (Aghion, Dechezleprêtre, Hémous, Martin, Van Reenen, 2016). Climate change is caused by the emission of carbon dioxide and other greenhouse gasses (Aghion et. al., 2016). The emission of greenhouse gasses is mainly an occurrence of energy production (Environmental protection agency, 2015). A transition in the way humanity produces its energy is therefore necessary to limit the consequences of climate change (Chu & Majumdar, 2012). A transition in one of the most fundamental industries in the economy comes with many challenges. However, it also creates a lot of opportunities especially for new business ideas.

According to the European Commission cars contribute for 12% to the total CO2 emissions in the EU (2017). Therefore, a reduction in emissions from cars seems necessary to contribute as a partial solution to the problem. There are already many technical possibilities for creating cleaner cars. For example, plug in hybrid electric vehicles and vehicle to grid car concepts (Sovacool & Hirsh, 2009). However, these types of electric cars come with multiple limitations. Even though, there has been a surprising increase of electric cars in the automobile market (Dijk & Yarime, 2010), there are still many barriers for a total transition from fossil fuel to electric driven cars (Sovacool & Hirsh, 2009).

Green consumption can add ecological value to products (Koller, Floh, Zauner, 2011). Since electric cars are a form of green consumption, it would be logical to assume that these cars have ecological value. However, next to technical and infrastructural barriers, there are also social and cultural barriers for customers to switch to electric driving (Sovacool & Hirsh, 2009). Nonetheless, electric cars can be used for more options than the direct personal transport of customers. For example, electric cars can be used for transportation services, such as taxis, delivery services, and rental services. Therefore, this research aims to find out if increased ecological value has a positive effect on loyalty intentions for these types of businesses and if customers have increased willingness to pay for their services. This resulted in the following research question:

- To what extent can businesses that offer transportation and delivery services benefit from increased willingness to pay that is created by the use of electric cars?

In order to get a better understanding of the concepts related to this question the following sub questions will be answered in this paper as well:

- What is the relation between green consumption and ecological value?

- What is the relation between ecological value, consumer loyalty and willingness to pay?

- Why are business offering transportation and delivery services more likely to profit from ecological value than businesses in the automotive industry?

In this research firstly a literature review will be provided. This literature review answers the three sub questions stated above. Based on the concepts and variables described in the literature review, a visualized conceptual framework is drawn. After that the research strategy is described in methods. Following, the results of the research will be described and statistically analysed. The results will afterwards be discussed and finally a conclusion will be drawn to answer the main question.

(6)

2. Literature review

This part of this research answers the sub questions. The three questions are each answered in a separate paragraph. Before these questions are answered it needs to be stated that there are multiple technical options to switch from fossil fuel to electric driven cars. To simplify, this research does not make distinctions and defines them all as electric cars. 2.1 Green consumption and ecological value The first paragraph of this literature review answers the first sub question: What is the relation between green consumption and ecological value? Next to that, the definitions of these concepts are provided in order to get a better understanding of the subject. It also explains how these concepts are related to the transportation and delivery industry. Green consumption is a term that is widely used and can be interpreted in multiple ways. In todays society there is a trend towards green consumption, which is consuming with the general consensus that something good for the environment needs to be done (Meijers & Stapel, 2011). In this research green consumption is used in its environmental dimension, meaning a sustainable form of consumption (Gilg, Barn, Ford, 2005). More specifically, this means consumption of products or services that are not or less polluting than similar traditional variants of these products. In this research this concept is applied to the car industry. Therefore, the term green consumption is here specifically used as the use of electric cars instead of the use of fossil fuel driven cars by businesses. In this research green consumption is used as the starting variable, because this research focuses on businesses using electric cars and not customers. Therefore, there is no direct consumption of electric cars by the customer. Thus, green consumption is in this research more the idea of consuming services that are better for the environment.

Ecological value is the increased customer value of products based on the idea that these products are generally better for the environment, especially compared to similar traditional variants of these products (Koller et. al., 2011). Marketing activity is fundamentally based on creating value for customers (Holbrook, 1994). Therefore, creating ecological value could be an interesting focus for businesses. Both cognitive and effective elements influence the perceived value of customers. Cognitive elements are functional and economic value. Functional value is related to functionality and quality of products. Functional value can positively influence ecological value, when sustainable variants of products also have better quality or functionality (Koller et. al., 2011). Economic value is related to the costs of products. Koller et al. state that it can have a positive effect on ecological value when sustainable variants of products are relatively cheaper than traditional ones (2011). Affective elements are emotional and social value. Emotional value is about feelings and attachment for products or brands. It can have a positive effect on ecological value when consumers experience positive feelings when consuming sustainable products (Koller et. al., 2011). Social value is about the status and signals related to products. It has a positive effect on ecological value when consumers use sustainable products as a signal or statement to their social environment (Koller et. al., 2011). More recent research has shown that these different types of value can be used to predict green consumption behaviour (Goncalves, Lourenco and Silva, 2016). In this research these variables will be used to predict willingness to pay for services, because there will be no direct use of electric cars by customers.

The specific relation between green consumption and ecological value is the following. In general green products can have a positive effect on al four types of

(7)

customer value. Therefore, green consumption can have an indirectly influence on ecological value (Koller et. al., 2011). 2.2 Consumer loyalty and willingness to pay The second paragraph of this literature review answers the second sub question: What is the relation between ecological value, consumer loyalty and willingness to pay? It is important to get a better understanding of the meaning of these concepts and how they relate to the subject of this research.

Consumer loyalty is about the post consumption behaviour of customers regarding products and brands. It is shown that consumers who feel connected to a brand have loyalty intentions to this brand (Loureiro, Ruediger, Demetris, 2012). For example, this means that loyal customers are more likely to repurchase. Consumer loyalty can be used as an indicator for a company’s success in a competitive market (Park, Joon Kim, Jib Kwon, 2017). Companies with higher ethical standards participate better in social corporate responsibility (CSR). CSR is a concept that describes that companies have to do what is right, not only from the firms perspective but also for all other stakeholders involved (Muthuri, 2013). This implies that companies involved with the use of electric cars are somehow participating to CSR, because they are doing something good for their environment. Research has shown that companies involved with CSR have more satisfied customers and that these customers are more likely to remain loyal to the company (Park, Joon Kim, Jib Kwon, 2017). Related to this subject it is therefore assumed that businesses that create ecological value by using electric cars for their services have higher loyalty intentions among their customers.

Willingness to pay is a concept that speaks for itself. It represents the amount that customers are willing to pay for a certain product or service. There is scientific uncertainty about the influence of customer loyalty on willingness to pay. However, some researches have shown that loyal customers seem to have less price sensitivity then regular customers (Umashankar, Bhagwat, Kumar, 2016). This generally means that loyal customers are willing to pay more for products. Furthermore, it is also shown that satisfied customers have a higher willingness to pay for products (Homburg, Koschate, Hoyer, 2005). Advise company Mckinsey claims that 80% of the people are willing to pay 5% more for green products or services (Miremadi, Musso, Weihe, 2012). Therefore, it is assumed in this research that people are willing to pay 5% or more for transportation and delivery services performed with electric cars. Moreover, advise company Mckinsey claims that willingness to pay decreases when the amount increases, which is visualized in figure1 (Miremadi, Musso, Weihe, 2012). This indicates that price has a moderating effect on the relation between loyalty intentions and willingness to pay.

Figure 1: Share of consumer picking green (Miremadi, Musso, Weihe, 2012)

(8)

The relation between ecological value, customer loyalty and willingness to pay is the following. Ecological value has a positive impact on the loyalty intentions of customers (Koller et. al. 2011). Based on the above it can be assumed loyalty intentions has a positive effect on willingness to pay, which is moderated by the variable price.

2.3 Transportation and delivery services

This paragraph answers the third sub question: why are businesses offering transportation and delivery services more likely to profit from ecological value than businesses in the automotive industry? This paragraph explains how the concepts in the first two paragraphs are related to the automotive and the transportation and delivery industries. The automotive industry is all businesses involved with design, development, manufacturing and selling of automotive vehicles. Firstly the relation of all four types of customer value to the automotive industry is explained. Cognitive elements regarding value creation can be applied to the electric car industry in the following way. Functional value regarding cars is based on quality, mobility, technology and speed of cars (Tertoolen, Kreveld, Verstraten, 1998). Functional value regarding electric cars is questionable, because electric cars have some functional disadvantages compared to fossil fuel driven cars. For example, fully loaded electric cars can only drive up until 30 to 100 km a day (Denholm, Short, 2006). More recent research also confirms that batteries of electric cars still have limited capacity (Alam, Muttaqi, Sutanto, 2016). Therefore, they are not yet applicable for long distance drives. On average the quantification of this disutility for customers is around 10.000 US dollar (Sovacool & Hirsh, 2009). Therefore, customers would feel like this amount has to be compensated for in the price of the cars. Which therefore negatively affects economic value. Economic value is the perceived value from financial savings compared to substitute products (Koller et. al., 2011). Related to this subject economic value has a negative effect on ecological value, because electric cars are more expensive than comparable conventional cars (Griskevicius, Tybur, Van den Bergh, 2010). This is mostly because, the purchase costs of electric cars are often higher then those of conventional ones (Romm, 2006). However, electric cars can also have economic benefits for customers. For example, consumers can save money they would otherwise spend on fuel. Furthermore, the batteries of their cars could be used for electricity storage, which could later on be sold back to the grind (Sovacool & Hirsh, 2009). So, there are multiple reasons driving electric could be profitable for consumers. Nonetheless, it seems that most consumers do not make a proper financial analysis when buying a car (Turrentine & Kurani, 2006). More specifically, consumers do not consider fuel savings at the moment of their purchase. Research has shown that customers expect a payback period of less than five months when investing in energy efficient products, which is not reasonable for fuel savings (Sovacool & Hirsh, 2009). From the above it can be concluded that functional and economic values do not contribute to ecological value in the electric car industry.

Effective elements regarding value creation can be applied to the electric car industry in the following way. Emotional value can be linked to electric cars because consumers can experience positive feelings, because consumers believe they are doing something good (Corral-Verdugo, et al., 2009). The main benefit of electric cars is that they do not use fossil fuels (Kinter-Meyer, Schneider, Pratt, 2007). Therefore, they have no direct emissions of CO2 and other greenhouse gasses. The use of electric cars could therefore result in less air pollution and limit the effects of climate change (Sovacool & Hirsh, 2009). The knowledge of these benefits makes people feel like they are

(9)

participating to limiting climate change, which they experience as something good. Therefore, emotional value has a positive effect on ecological value of electric cars. Social value is linked to electric cars when customers use electric cars to make a social statement. Research shows that some consumers drive electric cars to signal that they are pro-social instead of pro-individual (Griskevicius et. al., 2010). Therefore, social value also has a positive effect on ecological value in the electric car industry.

Besides the drawbacks for customers there are also greater barriers to a transition to electric cars. Established companies in the car and fuel industry have a reserved position regarding electric cars, because they do not want to risk their market share (Sovacool & Hirsh, 2009). These companies as well as established infrastructure are related to the conservative character of the car industry, due to path dependency (Aghion et. al., 2016). Furthermore, a negative feedback loop arises when electric cars will be used on a larger scale. If this happens, it is expected that the price of fossil fuels will drop due to lower demands (Hargreaves, 2007). Therefore, the financial benefits of electric driving can decrease.

The positive relation between green consumption, ecological value, customer loyalty and willingness to pay seems not to hold for businesses selling electric cars for direct personal use. However, businesses can also use electric cars for multiple types of services. Think about transportation services such as public transport and taxis. Other options are delivery or rental service, such as food delivery services or rental of moving vans. The drawbacks of electric cars for customers do not hold when making use of these services. When transportation time is assumed to be equal, these services would be valued higher when provided with electric cars. This is because these services would be considered as green consumption when offered with electric cars. The functional and economic drawbacks of electric cars would not be applicable to the customers, because they only buy the service and not the car. Therefore, it is likely to assume that they would be willing to pay more for their services.

Of course it must be noted that the functional and economic drawbacks also apply to businesses. However, they can be more easily bearable for companies due to economies of scales (Kaldellis, Spyropoulus, Liaros, 2016). This is because companies can make use of their own electric vehicle charging stations (EVCS). For example, solar electric vehicle charging stations are considered to be the most environmentally friendly solutions for the decarbonisation of the transport sector. These stations require a high investment, which can be paid of by complete fuel saving (Kaldellis, Spyropoulus, Liaros, 2016). Furthermore, research has shown that electric cars are suitable for short taxi trips (Smith, 2017). This is confirmed by the practical example of TaxiElectric in Amsterdam. This is the first electric taxi service in Europe that offers rides between the city and the airport only (TaxiElectric, 2015). It is also shown that in smaller communities electric cars can be used for food delivery (Cheng, Jiang, Meysamifard, Petersen, 2014).

Concluding, businesses could potentially be better able to handle the functional and economic drawbacks of electric cars. Electric cars are therefore useable for businesses offering short distance services. These businesses can earn their return on investment from fuel savings. Next to that, this research is set up to find out if these businesses can also charge higher prices due to increased consumer loyalty.

(10)

2.4 Conceptual framework In this paragraph a visualized conceptual framework is provided, see figure 2 below. It shows the relation between the main concepts that are applied in this research. So far it is shown that green consumption has a positive effect on both emotional and social value, which influence ecological value. The effect of functional and economic value on ecological value is more uncertain. Therefore these cognitive aspects will not be taken into consideration in this research. Furthermore, increased ecological value has a positive effect on loyalty intentions. Previous research has shown that this generally results in a higher willingness to pay, however this relation is assumed to be moderated by price. This results in the following conceptual framework. Figure 2: Visualized conceptual framework

(11)

3. Methods

This part describes how the research is set up and what methods are applied. Firstly, the setup of the questionnaire is explained and the distribution sample is described. Each of the variables in the conceptual framework are linked to the corresponding questions in the questionnaire. After that, it is explained how the data is analysed. The different statistic test that are used, are mentioned and linked to the corresponding questions in the questionnaire. 3.1 Data gathering 3.1.1 Questionnaire This research is a quantitative research. The data is gathered via a survey in the form of an online questionnaire. It is a cross-sectional research, which means that it is taken at a particular point in time and does not take changes over time into consideration (Brotherton, 2015). Furthermore, it is a descriptive research, which means that it is aimed to link characteristics to certain preferences, attitudes, without necessarily explaining these relations (Brotherton, 2015). It is an indirect survey, meaning that there will be no direct contact with the respondents.

The aim of this questionnaire is to check the relations between the variables shown in the conceptual framework and link them to the transportation and delivery industry. In the questionnaire these variables will be linked to questions related to businesses offering transportation and delivery services. The questionnaire is attached in appendix 1. The set up of the questionnaire with the subject of each question is shown below in table 1. Table 1: Set up Questionnaire General information / personal

characteristics Q1: Gender Q2: Age

Q3: Education level Q4: Student Q5: Employment Q6: Income Green consumption Q7: Green consumption Q8: Feeling good Q9: Signalling Q10: Price sensitivity Q11: Brand connection Willingness to pay Q12: Public transport Q13: Food delivery Q14: Taxi Q15: Moving car rental In question one to six respondents are asked about their personal characteristics, such as gender, age, education level, employment and income. This information is not used in this research. However, it could be used in further to check if respondents with different characteristics have different preferences or attitudes. This information improves the scientific and social relevance of the data. Questions seven to eleven are statements about green consumption. Respondents have to indicate how much they agree to a certain statement on a five point likert scale with answer possibilities; strongly disagree,

(12)

disagree, neutral, agree, and strongly agree. The choice for a five point likert scale is made, because it is hard for people to sensibly distinguish between more than five options (van Casteren, 2017).

Last, there are four questions on the respondents spending behaviour. There are four hypothetical situations regarding services where vehicles are used, for each situation they have to indicate how much they would be willing to pay more if the service was offered with an electric vehicle. These four questions are multiple-choice questions with the options 0, 5, 10 or 20 per cent. These questions quantify the willingness to pay of the respondents. Each situation is in a different price category with different estimated costs. Respectively, they are public transport with costs of €3,-, food delivery with costs of €10,-, taxi ride with costs of 30,- and moving car rental with costs of €100,-. The survey is set up with this different price categories, to check whether there is a difference between respondents respective willingness to pay as the prices increase. Table 2 below shows the variables of the conceptual framework and the questions to which they correspond. Table 2: Variables and corresponding questions Green Consumption Q7 Emotional value Q8 Social value Q9 Ecological value Q10 Loyalty intentions Q11 Willingness to pay Q12 Q13 Q14 Q15 3.1.2 Sample It would be most preferable to gather a sample from a broad population. However, Given the scale of this research it will be difficult to obtain a sample that is a good representative of the whole population. To prevent a non-random sample, certain biases are designed into the sample, which is necessary following Bortherton (2015). The research survey will be distributed via social media and university platforms. Therefore, the reach will most likely be students and young adults. To conform to this expectation the aimed research sample are young-adults in the self-defined age category 16-30 year old living in Amsterdam and surrounding areas. This geographical area is chosen, because it is highly likely that most respondent live in this area and the aim is to prevent geographical biases.

This sample is interesting for this research, since it aligns with the theory that young-adults in general have a higher level of concern about climate change and lower levels of scepticism (Cornet et. al. 2015). Even though young people find governments mostly responsible for responding to climate change, they are less fatalistic about combating climate change than older age groups (Cornet et. al. 2015). Linking back to the conceptual framework, this would include that young-adults would be more sensitive for emotional and social value for green consumption. Therefore it could be expected that this age group would be willing to spend more on sustainable car alternatives for delivery or transport services.

(13)

3.2 Data analysis 3.2.1 Quantification Firstly, the collected data will be statistically analysed by using SPSS. The data collected through the questionnaire is mostly qualitative data. This is because the questions were set up as likert scales. Most statistical tests require quantitative data for their analysis. Therefore, the data from the survey is adjusted to quantitative data if necessary. This is possible, since the likert scale questions in this survey can be treated as quantitative data. This is because the distance between the answers is presumably equal. For example, the distance between the following likert scale answer-possibilities is presumed to be equal; strongly disagree, disagree, neutral, agree and strongly agree. Therefore, these answers could also be treated as the following answer possibilities; 1, 2, 3, 4, 5. If necessary to check for significance, relations, dependence, variance, correlation and moderation, this is the way the data is quantified.

3.2.2 Testing on significance of variables

Secondly, the data is tested on significance. Significance is not necessarily important for testing the relations in the conceptual framework. However, it is interesting to know if the independent variables of the theory turn out to be existent in the population. For example, based on the theoretical framework it is assumed that people have emotional value for green products. However, if the data shows that people do not feel good when consuming green products, then this variable would become irrelevant to test the relation in the framework. Therefore, questions seven to eleven are all tested on significance. To test for significance the data is quantified and tested via a one-sample t-test.

The questions on willingness to pay are also tested on significance. This is to check whether or not the data indicates if people are truly willing to pay more. Again, this is done with a one-sample t-test for each question. These questions ask the same question for different price categories. Multiple paired-sample t-tests are used to test whether or not the willingness to pay decreases when the price increases. If this is the case, than this is an indication that price is a moderating variable.

3.2.3 Testing relations between independent variables

Thirdly, the relation between these variables is tested. Individual relations are tested with a chi-square test. This is a test that is used for relations between two variables (van Casteren, 2017). Therefore, the data does not have to be quantified to use this test. A chi-squared test tests for dependence between two variables. The variable willingness to pay is checked for in four questions. To check if this variable is reliable, which means that all four questions are somehow answered in the same manner, a reliability test is performed. This is done with a Cronbach’s Alpha test in SPSS. If the variable is reliable, then all four questions are summated to one scale. This scale in then used to check for relations with the chi-square test. All chi-squared tests are tested on significance using the p-values and the degree of freedom. If the expected counts of row*column in one of the chi-squared test > 5, then cells are merged for this specific test. The chi-squared tests only indicate whether or not there is a relation between the variables. Therefore all these test are also be tested on the direction and the degree of this relation. This is done using Kendall’s tau-b or Kendall’s tau-c tests in SPSS.

(14)

4. Results

This part shows the results of this research. The collected data is presented visualized. The data will be statistically tested on significance and on relations between variables via SPSS as described in methods. All SPSS output is shown an explained. The questionnaire had a total of 69 respondents. 4.1 One sample t-test on variable statements In this part the results of questions seven to twelve are presented in a bar chart. It must be noted that each bar chart has a different count on the y-axis. Please keep this in mind while studying the charts. Each question is tested on significance with a one-sample t-test. The formula of a t-test is shown in formula 1. Therefore the data of these question is quantified to strongly disagree=1, disagree=2, neutral=3, agree=4 and strongly agree=5. The hypotheses are set up before looking at the results. All variables in the conceptual framework are assumed to be existent. Therefore all test are performed right-sided. This is because it is assumed that people agree to the statements in the questions. The test value in each test is 3, because this value corresponds with being neutral. These test are all valid because the sample is random and N = 69 > 30. All test are performed with a significance level of 5%. Formula 1: t-test df = N-1 4.1.1 Green consumption Green consumption is tested in questions seven. Q7 is the following statement ‘I believe that the consumption of green products contributes to limiting the consequences of climate change’. In bar chart 1 below the results of Q7 are shown. It shows that most people agree to the statement in the question. There are no people who fully disagree. The following hypothesis is tested: 1. H0 (Q7): Respondents are neutral regarding the contribution of green products as a solution to climate change (U=3) H1 (Q7): Respondents agree that the consumption of green products contributes to limiting the consequences of climate change (U>3) The statistic results of Q7 are shown in table 3 and 4. They show a mean of 3,83 with a t-value of 8,539. The corresponding P-value of a right-sided test is 0,000/2=0,000. This means that the change of finding this t-value is 0,0% when the population mean would be 3. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general agree that the consumption of green products contributes to limiting the consequences of climate change.

(15)

Bar Chart 1: Q7: I believe that the consumption of green products contributes to limiting the consequences of climate change. Table 3: Statistics Q7 Table 4: Test results Q7 4.1.2 Emotional value

Emotional value is tested in question eight (Q8). Q8 is the following statement ‘I feel better when consuming green products instead of a similar traditional product’. Because the statement is about feelings it is corresponding to the variable emotional value. The data of Q8 is shown in bar chart 3 below. The bar chart shows that most respondents agree to the statement. The following hypothesis are used to test for significance of these results:

1. H0 (Q8): Respondents are neutral about feeling better when consuming green products instead of a similar traditional product (U=3)

H1 (Q8): Respondents agree to feel better when consuming green products instead of a similar traditional product (U>3).

(16)

Bar chart 2: Q8: I feel better when consuming green products instead of a similar traditional product The results of Q8 are shown in table 7 and 8 below. They show a mean of 3,43 with a t-value of 3,440. The corresponding P-value of a right-sided test is 0,001/2=0,0005. This means that the change of finding this t-value is 0,1% when the population mean would be 3. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general feel better when consuming green products instead of a similar traditional product. Table 5: Statistics Q8 Table 6: Test results Q8 4.1.3 Social value

Social value is tested in question nine (Q9). Q9 is the following statement ‘I would consume green products to signal to my social environment’. Because the statement is about signalling it is corresponding to the variable social value. The data of Q9 is shown in bar chart 4 below. The bar chart shows that an equal amount of respondents agrees to or disagrees to the statement. However it is assumed that social value is existent. Therefore, the following hypothesis are used to test for significance of these results:

(17)

2. H0 (Q9): Respondents are neutral about consuming green products to signal to their social environment (U=3)

H1 (Q9): Respondents agree to consume green products to signal to their social environment (U>3).

The results of Q9 are shown in table 9 and 10 below. They show a mean of 3,17 with a t-value of 1,271. The corresponding P-The results of Q9 are shown in table 9 and 10 below. They show a mean of 3,17 with a t-value of a right-sided test is 0,208/2=0,104. This means that the change of finding this t-value is 10,4% when the population mean would be 3. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that it is not significantly tested that people in general would consume green products to signal to their social environment. Bar chart 3: Q9: I would consume green products to signal to my social environment Table 7: Statistics Q9 Table 8: Test results Q9

(18)

4.1.4 Ecological value

Question ten (Q10) asks for willingness to buy from a green or a regular brand with a similar product. If respondents indicate to prefer to buy form a green imaginary brand, than this means they have ecological value for this imaginary product. The results to Q10 are shown in bar chart 2 below. It shows that most respondents agree to the statement. The following hypothesis are used to test for significance: 3. H0 (Q10): Respondents are neutral about willing to buy from a green brand than from a regular brand with a similar product (U=3) H1 (Q10): Respondents agree to be more willing to buy from a green brand than from a regular brand with a similar product (U>3). Bar chart 4: Q10: I am more willing to buy from a green brand than from a regular brand with a similar product The results of Q10 are shown in table 9 and 10 below. They show a mean of 3,35 with a t-value of 2,574. The corresponding P-value of a right-sided test is 0,012/2=0,006. This means that the change of finding this t-value is 0,6% when the population mean would be 3. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general are more willing to buy from a green brand than from a regular brand with a similar product. Table 9: Statistics Q10

(19)

Table 10: Test results Q10

4.1.5 Loyalty intentions

The variable loyalty intention is tested in question eleven (Q11). Q11 is the following statement ‘If a brand has a green image I feel more connected to this brand’. Because the statement is about brand connection it is corresponding to the variable loyalty intentions. The data of Q11 is shown in bar chart 5 below. The bar chart shows that most respondents disagree to the statement. However, based on the theory it is assumed that respondent would agree to the statement. Therefore the following hypothesis are tested: 4. H0 (Q11): Respondents are neutral about feeling connected to a brand with a

green image (U=3)

H1 (Q11): Respondents agree to feeling connected to a brand with a green image (U>3). Bar chart 5: Q11: I would consume green products to signal to my social environment The results of Q11 are shown in table 11 and 12 below. They show a mean of 2,93 with a t-value of -0,560. The corresponding P-value of a right-sided test is 1-(0,577/2)=0,7115. This means that the change of finding this t-value is 71,2% when the population mean would be 3. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that it is not significantly tested that people in general agree to feeling connected to a brand with a green image.

(20)

Table 11: Statistics Q11 Table 12: Test results Q11 4.2 One sample t-test on willingness to pay questions In this part the results of questions twelve to fifteen are presented in a bar chart. It must be noted that each bar chart has a different count on the y-axis. Please keep this in mind while studying the charts. Each question is tested on significance with a one-sample t-test. The hypotheses are set up before looking at the results. Based on the theory it is assumed that people are willing to pay 5% more for the transportation and delivery services. Which results in all right-sided tests with a test value of 5. These test are all valid because the sample is random and N = 69 > 30. All test are performed with a significance level of 5%.

4.2.1 Public transport services

Question twelve (Q12) is about public transport services. Respondents are asked how much they would be willing to pay more for a ride with public transportation operated with electric vehicles. The price of this ride is €3,- .They can choose between 0%, 5%, 10% or 20%. In bar chart 6 below the results of Q12 are shown. It shows that most people would be willing to spend 10% more for this type of service. The following hypothesis is tested:

5. H0 (Q12): Respondents are on average willing to spend 5% more for public transport services (U=5) H1 (Q12): Respondents are on average willing to spend more than 5% more for public transport services (U>5). The statistic results of Q12 are shown in table 13 and 14. They show a mean of 9,64 with a t-value of 7,075. The corresponding P-value of a right-sided test is 0,000/2=0,000. This means that the change of finding this t-value is 0,0% when the population mean would be 5%. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general are willing to spend more than 5% more for public transport services operated with electric vehicles.

(21)

Bar chart 6: Q12: Public transport services Table 13: Statistics Q12 Table 14: Test results Q12 4.2.2 Food delivery services

Question thirteen (Q13) is about food delivery services. Respondents are asked how much they would be willing to pay more for food delivery services operated with electric cars. The price of this service is €10,-. They can choose between 0%, 5%, 10% or 20%. In bar chart 7 below the results of Q13 are shown. It shows that most people would be willing to spend 5% more for this type of service. The following hypothesis is tested:

6. H0 (Q13): Respondents are on average willing to spend 5% more for food delivery services (U=5)

H1 (Q13): Respondents are on average willing to spend more than 5% more for food delivery services (U>5).

(22)

Bar chart 7: Q13: Food delivery services

The statistic results of Q13 are shown in table 15 and 16. They show a mean of 6,59 with a t-value of 2,893. The corresponding P-value of a right-sided test is 0,005/2=0,0025. This means that the change of finding this t-value is 0,25% when the population mean would be 5%. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general are willing to spend more than 5% more for food delivery services operated with electric vehicles. Table 15: Statistics Q13 Table 16: Test results Q13 4.2.3 Taxi services Question fourteen (Q14) is about taxi services. Respondents are asked how much they would be willing to pay more for taxi services operated with electric cars. The price of this service is €30,-. They can choose between 0%, 5%, 10% or 20%. In bar chart 8 below the results of Q14 are shown. It shows that most people would be willing to spend 5% more for this type of service.

(23)

Bar chart 8: Q14: Taxi services The following hypothesis is tested: 7. H0 (Q14): Respondents are on average willing to spend 5% more for taxi services (U=5) H1 (Q14): Respondents are on average willing to spend more than 5% more for taxi services (U>5). The statistic results of Q14 are shown in table 17 and 18. They show a mean of 6,01 with a t-value of 1,906. The corresponding P-value of a right-sided test is 0,061/2=0,0305. This means that the change of finding this t-value is 3,05% when the population mean would be 5%. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that it is significantly tested that people in general are willing to spend more than 5% more for taxi services operated with electric vehicles. Table 17: Statistics Q14 Table 18: Test results Q14

(24)

4.2.4 Moving car rental services Question fifteen (Q15) is about moving car rental services. Respondents are asked how much they would be willing to pay more for moving car rental services operated with electric cars. The price of this service is €100,-. They can choose between 0%, 5%, 10% or 20%. In bar chart 9 below the results of Q15 are shown. It shows that most people would be willing to spend 5% more for this type of service. The following hypothesis is tested: 8. H0 (Q15): Respondents are on average willing to spend 5% more for moving car rental services (U=5) H1 (Q15): Respondents are on average willing to spend more than 5% more for moving car rental services (U>5). The statistic results of Q15 are shown in table 19 and 20. They show a mean of 5,36 with a t-value of 0,727. The corresponding P-value of a right-sided test is 0,470/2=0,235. This means that the change of finding this t-value is 23,5% when the population mean would be 5%. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that it is not significantly tested that people in general are willing to spend more than 5% more for food delivery services operated with electric vehicles. Bar chart 9: Q15: Moving car rental services Table 19: Statistics Q15

(25)

Table 20: Test results Q15

4.3 Paired sample t-tests on difference between willingness to pay questions

In this part paired sample t-tests are performed to check if willingness to pay decreases when the price increases. If this is the case, than this is an indication that price is a moderating variable. The hypothesis is set up before looking at the results. Based on the theory it is therefore assumed that willingness to pay decreases when the price increases. The tests are all valid because the sample is random and N = 69 > 30. All test are performed with a significance level of 5%. First Q12 and Q13 are compared. This results in the following hypothesis: 9. H0: U(Q12) – U(Q13) = 0 H1: U(Q12) – U(Q13) > 0 SPSS gives the output shown in table 21. Table 21: U(Q12)-U(Q13) This output shows a mean difference of 3,043, while a mean difference of 0 is assumed. The t-value is 4,337. The corresponding p-value for a right-sided test is 0,000/2=0,000. This means there is a chance of 0,0% of finding this t-value when the population mean would be 0. This is less than 5% and therefore H0 is rejected and H1 is supported. This means that the relative average of the willingness to pay decreases, when the price increases from €3,- to €10,-. This indicates that price is a moderating variable on willingness to pay. Secondly, Q13 and Q14 are compared. This results in the following hypothesis: 10. H0: U(Q13) – U(Q14) = 0 H1: U(Q13) – U(Q14) > 0 SPSS gives the output shown in table 22. This output shows a mean difference of 0,580, while a mean difference of 0 is assumed. The t-value is 1,183. The corresponding p-value for a right-sided test is 0,241/2=0,1205. This means there is a chance of 12,1% of finding this t-value when the population mean would be 0. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that there is no

(26)

significant evidence that the relative average of the willingness to pay decreases, when the price increases from €10,- to €30,-. Table 22: U(Q13)-U(Q14) This output shows a mean difference of 0,580, while a mean difference of 0 is assumed. The t-value is 1,183. The corresponding p-value for a right-sided test is 0,241/2=0,1205. This means there is a chance of 12,1% of finding this t-value when the population mean would be 0. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that there is no significant evidence that the relative average of the willingness to pay decreases, when the price increases from €10,- to €30,-. Thirdly, Q14 and Q15 are compared. This results in the following hypothesis: 11. H0: U(Q14) – U(Q15) = 0 H1: U(Q14) – U(Q15) > 0 SPSS gives the output shown in table 23. Table 23: U(Q14)-U(Q15) This output shows a mean difference of 0,652, while a mean difference of 0 is assumed. The t-value is 1,241. The corresponding p-value for a right-sided test is 0,219/2=0,1095. This means there is a chance of 11,0% of finding this t-value when the population mean would be 0. This is more than 5% and therefore H0 is not rejected and H1 is not supported. This means that there is no significant evidence that the relative average of the willingness to pay decreases, when the price increases from €30,- to €100,-.

Finally, Q12 and Q15 are compared to get an overall indication. This results in the following hypothesis: 12. H0: U(Q12) – U(Q15) = 0 H1: U(Q12) – U(Q15) > 0 SPSS gives the output shown in table 24. This output shows a mean difference of 4,275, while a mean difference of 0 is assumed. The t-value is 6,527. The corresponding p-value for a right-sided test is 0,000/2=0,000. This means there is a chance of 0,0% of finding this t-value when the population mean would be 0. This is more than 5% and therefore H0 is rejected and H1 is supported. This means that there is significant evidence that the relative average of the willingness to pay decreases, when the price increases from €3,- to €100,-. This indicates that price is a moderating variable on willingness to pay.

(27)

Table 24: U(Q12)-U(Q15)

4.4 Test on relations between variables

In this part chi-square (X^2) tests are performed to check relations between the individual variables of the conceptual framework. The formula of the chi-square test is shown in formula 2. The chi-square table is shown in table 25. These tests are used for analysis of the relation between two categorical variables. All variables in the conceptual framework are categorical. Therefore the data does not have to be quantified for these tests. The variables are visualized in cross-tabs. Each cell shows the amount of observed (Oj) and expected (Ej) respondents per cell. A chi-square test is only valid under the condition Ej>5 for each cell of the cross-tab. When this is not the case, rows or/and columns are merged. A chi-square test only tests for association between the variables. Therefore, a Kendall’s tau-b or Kendall’s tau-c test is performed to check the degree and the direction of the association. The hypothesis is set up before looking at the results. The tests are all valid because the sample is random and N = 69 > 30. Formula 2: chi-square DF = (r-1)(c-1), r=rows, c=columns. Table 25: chi-square 4.4.1 Relation between green consumption and emotional value

Firstly, the relation between green consumption (Q7) and emotional value (Q8) is tested. To do this, the following hypothesis is used:

(28)

H1: Green consumption and emotional value are dependent.

The cross-tab results of these questions are shown in the table 26 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 27 below shows the same data with merged rows and columns. These numbers in table 27 are used to calculate X^2. Table 28 shows (Oj-Ej)^2/Ej per cell.

Table 26: Association Oj/Ej per cell

Q8\Q7 Strongly

disagree Disagree Neutral Agree Strongly agree TOTAL Strongly disagree 0/0,00 0/0,7 1/0,20 0/0,55 0/0,17 2 Disagree 0/0,00 2/1,16 7/3,25 7/8,81 0/2,78 16 Neutral 0/0,00 3/1,01 3/2,84 7/7,71 1/2,43 14 Agree 0/0,00 0/2,03 3/5,68 20/15,42 5/4,87 28 Strongly agree 0/0,00 0/0,72 0/2,03 4/5,51 6/1,74 10 TOTAL 0 5 14 38 12 69 Table 27: Association Oj/Ej per cell with merged rows and columns

Q8\Q7 Disagree*neutralNEW AgreeNEW TOTAL Disagree*NeutralNEW 16/8,53 15/22,45 31 AgreeNEW 3/10,56 35/27,54 38 TOTAL 19 50 69 Table 28: (Oj-Ej)^2/Ej per cell Q8\Q7 Disagree*neutralNEW AgreeNEW Disagree*NeutralNEW 6,54 2,47 AgreeNEW 5,41 2,02

This results in X^2 = 16,44 with DF = 1. X^2 > 6,63 and therefore, the probability of finding this X^2 < 0,01, which can be derived from table 25. This means there is a chance of less than 1% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that green consumption and emotional value are dependent variables. To check the degree and direction of the association a Kendall’s tau-b test is performed. This test is chosen because r = c in table 26. SPSS gives the output shown in table 29. Table 29: Kendall’s tau-b This output shows a Kendall’s tau-b value of 0,517. This indicates a moderate positive association. 4.4.2 Relation between green consumption and social value Secondly, the relation between green consumption (Q7) and social value (Q9) is tested. To do this, the following hypothesis is used:

(29)

14. H0: Green consumption and social value are independent H1: Green consumption and social value are dependent.

The cross-tab results of these questions are shown in the table 30 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 31 below shows the same data with merged rows and columns. These numbers in table 31 are used to calculate X^2. Table 32 shows (Oj-Ej)^2/Ej per cell. Table 30: Association Oj/Ej per cell Q9\Q7 Strongly disagree

Disagree Neutral Agree Strongly agree TOTAL Strongly disagree 0/0,00 0/0,22 3/0,61 0/1,65 0/0,52 3 Disagree 0/0,00 3/1,52 5/4,26 13/11,57 0/3,65 21 Neutral 0/0,00 2/1,09 2/3,04 9/8,26 2/2,61 15 Agree 0/0,00 0/1,52 3/4,26 14/11,57 4/3,65 21 Strongly agree 0/0,00 0/0,65 1/1,83 2/4,96 6/1,57 9 TOTAL 0 5 14 38 12 69 Table 31: Association Oj/Ej per cell with merged rows and columns

Q9\Q7 Disagree*neutralNEW AgreeNEW TOTAL Disagree*NeutralNEW 15/10,74 24/28,26 39 AgreeNEW 4/8,26 26/27,54 30 TOTAL 19 50 69 Table 32: (Oj-Ej)^2/Ej per cell Q9\Q7 Disagree*neutralNEW AgreeNEW Disagree*NeutralNEW 1,69 0,64 AgreeNEW 2,20 0,09 This results in X^2 = 4,62 with DF = 1. X^2 < 6,63, while X^2 > 3,84 and therefore, the probability of finding this X^2 is in between 0,01 and 0,05, which can be derived from table 25. This means there is a chance of more than 1%, but less than 5% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that green consumption and social value are dependent variables. To check the degree and direction of the association a Kendall’s tau-b test is performed. This test is chosen because r = c in table 30. SPSS gives the output shown in table 33. Table 33: Kendall’s tau-b This output shows a Kendall’s tau-b value of 0,4,31. This indicates a moderate positive association.

(30)

4.4.3 Relation between emotional value and ecological value Thirdly, the relation between emotional value (Q8) and ecological value (Q10) is tested. To do this, the following hypothesis is used: 15. H0: Emotional value and ecological value are independent H1: Emotional value and ecological value are dependent. The cross-tab results of these questions are shown in the table 34 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 35 shows the same data with merged rows and columns. These numbers in table 35 are used to calculate X^2. Table 36 shows (Oj-Ej)^2/Ej per cell.

Q8\Q10 Strongly

This results in X^2 = 31,77 with DF = 1. X^2 > 6,63 and therefore, the probability of finding this X^2 < 0,01, which can be derived from table 25. This means there is a chance of less than 1% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that emotional value and ecological value are dependent variables. To check the degree and direction of the association a Kendall’s tau-b test is performed. This test is chosen because r = c in table 34. SPSS gives the output shown in table 37.

(31)

Table 37: Kendall’s tau-b This output shows a Kendall’s tau-b value of 0,688. This indicates a moderate positive association. 4.4.4 Relation between social value and ecological value Thirdly, the relation between social value (Q9) and ecological value (Q10) is tested. To do this, the following hypothesis is used: 16. H0: Social value and ecological value are independent H1: Social value and ecological value are dependent. The cross-tab results of these questions are shown in the table 38 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 39 shows the same data with merged rows and columns. These numbers in table 39 are used to calculate X^2. Table 40 shows (Oj-Ej)^2/Ej per cell.

Q9\Q10 Strongly

This results in X^2 = 18,84 with DF = 1. X^2 > 6,63 and therefore, the probability of finding this X^2 < 0,01, which can be derived from table 25. This means there is a chance of less than 1% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that social value and ecological value are dependent variables. To check the degree and direction of the association a Kendall’s tau-b test is performed. This test is chosen because r = c in table 38. SPSS gives the output shown in table 41.

(32)

Table 41: Kendall’s tau-b This output shows a Kendall’s tau-b value of 0,604. This indicates a moderate positive association. 4.4.5 Relation between ecological value and loyalty intentions Now, the relation between ecological value (Q10) and loyalty intentions (Q11) is tested. To do this, the following hypothesis is used: 17. H0: Ecological value and loyalty intentions are independent H1: Ecological value and loyalty intentions are dependent. The cross-tab results of these questions are shown in the table 42 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 43 shows the same data with merged rows and columns. These numbers in table 43 are used to calculate X^2. Table 44 shows (Oj-Ej)^2/Ej per cell.

Q11\Q10 Strongly

Q11\Q10 Disagree*neutralNEW AgreeNEW TOTAL DisagreeNEW 25/14,38 6/16,62 31 Neutral 6/7,88 11/9,11 17 AgreeNEW 1/9,75 20/11,26 21 TOTAL 32 37 39 Table 44: (Oj-Ej)^2/Ej per cell Q11\Q10 Disagree*neutralNEW AgreeNEW Disagree 7,84 6,79 Neutral 0,45 0,39 AgreeNEW 7,85 6,78

This results in X^2 = 30,1 with DF = 2. X^2 > 9,21 and therefore, the probability of finding this X^2 < 0,01, which can be derived from table 25. This means there is a chance of less than 1% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that emotional value and loyalty intentions are dependent variables. To check the degree and direction

(33)

of the association a Kendall’s tau-b test is performed. This test is chosen because r = c in table 38. SPSS gives the output shown in table 45. Table 45: Kendall’s tau-b This output shows a Kendall’s tau-b value of 0,644. This indicates a moderate positive association. 4.4.6 Relation between loyalty intentions and willingness to pay

Finally, the relation between loyalty intentions (Q11) and loyalty willingness to pay (Q12, Q13, Q14 and Q15) is tested. To do this, the following hypothesis is used: 17. H0: Loyalty intentions and willingness to pay are independent H1: Loyalty intentions and willingness to pay are dependent. The variable willingness to pay is corresponding to four questions. In order to check if these questions are consistent with each other a reliability check is performed by using a Cronbach’ Alpha test. The results of this test are shown in table 46. Table 46: Cronbach’s Alpha The results show Cronbach’s Alpha = 0,810. This is an excellent indication for reliability between the variables. Therefore these variables are merged to one summated scale (SUM). The summated scale shows the average answer per respondent. This scale is used to test for the variable willingness to pay.

The cross-tab results of these questions are shown in the table 47 below. It shows Oj/Ej per cell. This table clearly has many cells with Ej<5. Therefore all cells and columns are merged so that Ej>5 for each cell. Table 48 shows the same data with merged rows and columns. These numbers in table 48 are used to calculate X^2. Table 49 shows (Oj-Ej)^2/Ej per cell.

(34)

Table 47: Association Oj/Ej per cell Table 48: Association Oj/Ej per cell with merged rows and columns Q11\SUM 0,00-5,00 6,25-7,50 8,75-15,00 TOTAL DisagreeNEW 13/9,3 7/10,8 11/10,2 31 Neutal 5/5,4 10/5,9 2/5,6 17 AgreeNEW 4/6,6 7/7,2 10/6,9 21 Total 22 24 23 69 Table 49: (Oj-Ej)^2/Ej per cell Q11\SUM 0,00-5,00 6,25-7,50 8,75-15,00 DisagreeNEW 1,47 1,34 0,06 Neutral 0,03 2,85 2,31 AgreeNEW 1,02 0,01 1,39 This results in X^2 = 10,48 with DF = 4. X^2 < 13,28, while X^2 > 9,49 and therefore, the probability of finding this X^2 is in between 0,05 and 0,01, which can be derived from table 25. This means there is a chance of less than 5% to find this X^2 if there would be no association between the variables. This is less than 5%, so H0 is rejected and H1 is supported. Which means that loyalty intentions and willingness to pay are dependent variables. To check the degree and direction of the association a Kendall’s tau-c test is performed. This test is chosen because r ≠ c in table 47. SPSS gives the output shown in table 50. Table 50: Kendall’s tau-c

This output shows a Kendall’s tau-c value of 0,213. This indicates a weak positive association.

(35)

5. Discussion

This part discusses the research. To do this, the implications of the research and the results are firstly discussed. To do this, they are linked back to the theoretical framework. Then, the limitations of the research are discussed and the effect of this limitations on validity and reliability. Finally, some questions raised by the results are discussed.

5.1 implications

This research can be used by marketing departments of businesses offering transportation and delivery services. T-tests on single variables in general confirm the following statements:

- People believe that green consumption contributes as a solution to climate change

- People feel better when consuming green products

- People are more willing to buy from a green brand than from a regular brand with a similar product.

This means that the variables green consumption, emotional value and ecological value are generally confirmed to be existent in the population. However, the following statements are not significantly confirmed:

- People consume green products to signal to their social environment - People feel connected to a brand with a green image.

The statements that are not significantly confirmed can be explained in the following manner. Social value is indeed an existing variable, however only a little group of people uses green products to signal to their social environment. Most people do not use green products to make statements. Therefore, it cannot be stated that social value is existent in a generalized population. This means that this variable is not really applicable for implications. The same goes for feeling connected to a green image, which is used to indicate for loyalty intentions. Customer loyalty and feeling connected requires a high level of emotional involvement with the brand. For most people this is just not the case. Therefore, this variable is also not applicable for implication in the context of this research.

T-tests on the willingness to pay questions confirm that people are in general willing to pay more than 5% more for public transport, food delivery and taxi services. However, people are not willing to pay more than 5% more for moving car rental services. This aligns with the theory that willingness to pay decreases as the amount of additional charge increases. This is also confirmed by the paired sample t-test. Not all price increases resulted in significant decreased willingness to pay. Nonetheless, a price increase from €3,- to €10,- and a price increase from €3 to €100 did result in significant decrease of willingness to pay. This indicates that price is a moderating variable. This implicates that business should consider this while setting their prices.

The data analysis confirms the indirect relation between green consumption and willingness to pay. This is because all single relations between variables in the conceptual framework are significantly confirmed by the results of this research. Green consumption is related to creating ecological value, because green consumption influence emotional and social value. The results show that the effect of green consumption on emotional value is stronger than the effect on social value. This means that people would rather consume green products, because it makes them feel good then they would consume them to signal to their social environment. This is something that marketing departments could consider while creating marketing strategies. The effects

Business possibilities due to the creation of ecological value through the use of electric cars