The adoption of E-bikes

(1)

The adoption of E-bikes

The relationship of instrumental, hedonic and symbolic attributes and the type of

trip towards the adoption of E-bikes.

(2)

2

The adoption of E-bikes

The relationship of instrumental, hedonic and symbolic attributes and type of

trip towards the adoption of E-bikes.

DATE 2019

University of Groningen

MSc Marketing Management

Department of Marketing

Faculty of Economics and Business

Author:

First supervisor:

Rogier Schipper

Dr. J.I.M. de Groot

Verlengde Hereweg 20-1E

University of Groningen

9722 AC Groningen

(3)

3

ABSTRACT

The aim of this research is to identify the effect of the perceived instrumental, hedonic and symbolic attributes on the intention to adopt an E-bike. An online questionnaire was held under 112 citizens in the Netherlands to obtain data about the valuation of the attributes among potential E-bike adopters. In addition, this research further developed the theory of adopting innovation by creating new insights in the theory. Results revealed that the instrumental attributes are mediated through hedonic attributes, but this research could not find support for the mediation through symbolic attributes. Additional exploratory analysis indicated a more complex model; the symbolic attributes were mediated again through hedonic attributes. This thesis argues that the valuation of symbolic attributes has an effect on the valuation of hedonic attributes. At last, this research could not find any relationship between the type of trip and the effect on the instrumental, symbolic and hedonic attributes.

(4)

4

1. INTRODUCTION

Late 1980 in Japan electrically assisted peddle bikes, also referred to as E-bikes, where introduced to the market (Fishman & Cherry, 2016). However, the E-bike was, due to its technological limitations and price, not widely adopted (Fishman & Cherry, 2016). Around the year 2000, due to innovation, E-bikes had better technology and where affordable for the public. Therefore the sales of E-E-bikes where rapidly increasing (Fishman & Cherry, 2016). To this day, the sales and usage of E-bikes are still increasing across the globe (P. A. Plazier, Weitkamp, & Berg, 2018). The rapid growth resulted in E-bikes being the largest alternative to fuelled vehicles in the world (Fishman & Cherry, 2016). For this reason E-bikes became an important research topic (Fishman & Cherry, 2016). Especially the effect of E-bike adoption and the effect on other types of transportation are an important research topic because it effects various contents of other research topics (Fishman & Cherry, 2016; Jones, Harms, & Heinen, 2016; Kroesen, 2017).

The adoption of an E-bike has impact on the other types of transportation that are used by consumers. E-bikes can improve the accessibility of people who are not able to ride a conventional bike (Jones et al., 2016). Most importantly, it is suggested that the adoption of the E-bike can in some way substitute motorized transportation methods (Fishman & Cherry, 2016; P. A. Plazier, Weitkamp, & Berg, 2017). E-bikes can substitute journeys that are too long for a conventional bike, and that are therefore done by a car. E-bikes do not only substitute car usage, it is indicated that E-bikes also substitute conventional bikes (Kroesen, 2017). Indeed, research showed that people more often use an E-bike as substitution of a conventional bike in comparison to cars (Fyhri & Fearnley, 2015; Kroesen, 2017). For this reason, this research topic is not only interesting for the transportation research field, but it also has an impact on the health of users (Fishman & Cherry, 2016). E-bike usage can be a substitution of other types of transportation, the reason for substitution can be explained by the motivation for E-bike adoption (Jones et al., 2016; P. A. Plazier et al., 2017).

Research in relation to the motivation and adoption of technology often focuses on the instrumental reasons of E-bike usage (Noppers, Keizer, Bockarjova, & Steg, 2015; Vandecasteele & Geuens, 2010). One of the most important reasons for using E-bikes is that they can overcome longer or more complicated trips that would not be beneficial with a conventional bike. This is because journeys over 10 km with a conventional bike would take too much time and take a lot of psychical effort in

(6)

6 People do not adopt an E-bike for instrumental reasons only. For example: people buy an E-bike for the enjoyment that they gain from riding an E-bike (Jones et al., 2016). Vandecasteele & Geuens (2010) identified three important motivations for consumer innovativeness; instrumental, hedonic and symbolic motivations. Instrumental motivations refer to the functional buying motivations (Steg, Vlek, & Slotegraaf, 2001; Voss, Spangenberg, & Grohmann, 2003). Hedonic buying motivations refer to fantasy and emotional aspects, for example enjoyment or pleasure (Hirschman & Holbrook, 1982; Voss et al., 2003). Symbolic buying motivations are related to the social self; people tend to buy items that can be related to their self-image (Cialdini & Goldstein, 2004; Jans & Fielding, 2018; Sirgy, 1985). Most research on buying motivations focusses on the instrumental motivations, and if they include hedonic and symbolic motivations they only take into the account the direct effects of these motivations (Arnold & Reynolds, 2003; Cialdini & Goldstein, 2004; Steg et al., 2001; Vandecasteele & Geuens, 2010). They suggest that these motivations are not related to each other.

In addition to the theory of instrumental, hedonic and symbolic attributes, Schuitema, Anable, Skippon, and Kinnear (2013) indicated that the instrumental, hedonic and symbolic attributes interact with each other. To be more precise: they identified that the relationship between perceptions of instrumental motivations and the intention to adopt is mediated by the symbolic and hedonic motivations. Identifying the relationship between the instrumental, hedonic and symbolic buying motivations is important to better understand the adoption theory (Schuitema et al., 2013; Vandecasteele & Geuens, 2010). However, further research is needed to validate this theory.

This thesis also contributes to the E-bike literature. E-bike theory is a relatively new type of research, thus most research on the buying motivations are descriptive and quantitative. Based on many interviews the buying motivations have been indicated (Jones et al., 2016; P. A. Plazier et al., 2017). Also, the pros and cons of E-bike usage per type of bike trip have been indicated (P. A. Plazier et al., 2018). However, further research is needed to identify the importance of these buying motivations in the adoption process of an E-bike.

Finally, this research will also factorize several buying motivations in the field of E-bikes into instrumental, hedonic and symbolic attributes (Jones et al., 2016; P. A. Plazier et al., 2017, 2018). Further research on E-bikes can use the framework of this thesis in order to identify the instrumental, hedonic and symbolic attributes in the field of E-bikes.

(7)

7 Current research also initiated that the types of mode (work, leisure and shopping) are predictors of E-bike usage (Jones et al., 2016; P. A. Plazier et al., 2018). E-E-bikes that are used for commuting trips are ridden by people with different motivations than if they are used for leisure trips (P. A. Plazier et al., 2018). These different motivations also result in different buying motivations of E-bikes (Jones et al., 2016; P. A. Plazier et al., 2018). However, this is only based on qualitative research. To further understand the difference in buying motivations regarding the type of mode further research is needed. This thesis tries to further investigate this subject. In order to close all the research gaps discussed above the following research question of this thesis is stated:

“What is the importance of perceived hedonic, symbolic and instrumental buying motivations and their effect on the willingness to adopt E-bikes which are bought for different types of modes?”

2. THEORETICAL BACKGROUND

This section elaborates on the theoretical background of the research question stated in the introduction. First, the concepts of intention to adopt: instrumental, symbolic and hedonic buying motivations, are described. Second, the relationship between these buying motivations and E-bikes will be discussed. At last, the effect of the type of mode towards this relationship is discussed. 2.1 Intention to adopt and instrumental, symbolic and hedonic buying motivations

In this research the main concept is the intention to adopt new products. This relates to the theory of planned behaviour which refers to the intention to perform a behaviour (Ajzen, 1991). The intention to adopt is a proxy for actual behaviour and therefore the motivation for underlying intentions (Ajzen, 1991). Since this research is interested in the underlying process, for this reason measuring the intention to adopt is relevant.

The intention to adopt new products is linked to the innovativeness of a consumer. Consumers innovative buying behaviour (the purchase of new products/innovations) is a heavily studied topic within research (Hirschman & Holbrook, 1982; Midgley & Dowling, 2002; Vandecasteele & Geuens, 2010). The interest in research started in the early Seventies and started with connecting personality traits to product innovativeness (Midgley & Dowling, 2002). Most of this research has neglected the relationship consumers have with products (Goldsmith, 1992). More recent research acknowledges that consumer innovativeness is not just based on the personality traits of consumers (Anable & Gatersleben, 2005; Vandecasteele & Geuens, 2010). It is also based on the motivations related to the intention to innovate (Vandecasteele & Geuens, 2010). Three main motivations of consumer

(8)

8

Instrumental motivations

Many consumer innovativeness concepts include instrumental dimensions, this theory was first brought up by Hirschan (1984). It suggests that products are bought for the reason that they are new and the functionality that they have. A similar theory, the maximum utility theory (Stigler, 1950), suggests that consumers buy products mostly based on utilitarian reasons, this implies that consumers seek maximum utility (Szmigin, 2018; Steg, Vlek, & Slotegraaf, 2001; Voss et al., 2003).

An important instrumental reason to buy E-bikes is that it can transport you faster in comparison to a conventional bike (Jones et al., 2016). E-bikes are used as a way of transportation, the functional aspects of E-bikes are enabling that (Jones et al., 2016). In an interview a consumer said the following about E-bikes: “I just need to get to work on time” (P. A. Plazier et al., 2017). This is a good example of an instrumental shopping motivation to adopt an E-bike. The functional aspects of an E-bike can fulfil this need. It is also considered to be good for the health of a person. However, there are also major negative instrumental motivations not to buy an E-bike. The weight of an E-bike is perceived as a disadvantage (Jones et al., 2016). The concerns about battery range and performance is for some consumers a big issue (Jones et al., 2016). At last, the price is also perceived as high, especially in comparison to a conventional bike (P. A. Plazier et al., 2018).

Symbolic motivations

Besides the instrumental motivations for innovative adopting behaviour, there are also reasons beyond the instrumental benefits. Sometimes consumers buy products for their symbolic meaning. This symbolic meaning of products says something about the person who buys it: it can signal who they are or what they would like to be (Gaterseben & Van der Werff, 2018). For example, buying or

consuming an E-bike can impress others and signal social status (Noppers et al., 2015). These symbolic buying motivations are not just based on signalling, it is related to the social self-identity, people tend to buy products that can be related to their self-image (Cialdini & Goldstein, 2004; Jans & Fielding, 2018; Sirgy, 1985).

The social self-identity is how people describe themselves (Gaterseben & Van der Werff, 2018). People can have different identities, which can be explained by the social environment they are in. Every group has their own social rules (Keizer P.W., 2012) and standards that are accepted within the group without the force of the law (Cialdini & Goldstein, 2004). People are social species, they have a fundamental need for companionship. Being socially isolated is perceived as stressful and traumatic (Cialdini & Goldstein, 2004). For this reason we comply and act in line with the group norm in order to be liked and accepted. Thus we adopt behaviour resulting from real or perceived pressure to fall in line with group behaviour (Szmigin, 2018). As said, this also results in adopting behaviour of

(9)

9 compared to cars (Jones et al., 2016). A positive environmental self-image can derive from adopting an E-bike. Products can, besides strengthening the self-image, also repair the self-image when it is damaged (Ward & Dahl, 2014). For example, riding an E-bike might repair your image for not being sporty.

This thesis suggests that symbolic motivation also plays a major part in the adoption of E-bikes, especially since the usage of E-bikes can be perceived as negative. Using E-bikes has for example been perceived as cheating (Jones et al., 2016; P. Plazier, 2018; P. A. Plazier et al., 2017). People reported that they are teased for being a cheat or that E-bikes are only for old members of society (Jones et al., 2016). Society thinks that E-bikes are not suited for young and fit people, and the society is eager to tell young people that (Jones et al., 2016). The reason for this is that older people are the early adopters in the technology lifetime cycle. This is contradicting to the believe that mostly young and innovative people are the early adopters in the technology lifetime cycle. This has been termed the rejuvenation of E-bikes (P. A. Plazier et al., 2018). This negative image can damage the self-identity of consumers or it can lead to the rejection of the social group someone is in (Cialdini & Goldstein, 2004; Ward & Dahl, 2014). This can be a reason not to adopt an E-bike. This is also a reason why people camouflage their E-bike as a conventional bike (Jones et al., 2016). However, this negative image is expected to change rapidly, it becomes more and more accepted by society to ride an E-bike for younger people (Peine, van Cooten, & Neven, 2016). There are also positive symbolic effects of E-bike usage: E-E-bikes are identified as environmental friendly compared to cars. For this reason people might also buy E-bikes in order to be more environmental friendly. Jones et al. (2016) indicated that this only applies for a small part of the community, most people see it as just a side benefit.

Hedonic motivations

Hirschman and Holbrook, (1982) were one of the first to advocate more attention to the hedonic consumption and described it as follows: “Hedonic consumption are those facets of consumer

behaviour that relate to the multisensory, fantasy and emotive aspect of product usage experience”. In

other words, hedonic consumption motivations refer to the enjoyment, pleasure and experience people have from buying products (Alba & Williams, 2013; Arnold & Reynolds, 2003; Voss et al., 2003). Hedonic shopping motivations also play a role when buying an bike. A lot of consumers ride an E-bike because they want to get out of the car (Jones et al., 2016; P. A. Plazier et al., 2017). They enjoy riding E-bikes because of the exercise they get. “I thought, coming to work 4-days a week by bus, I

don't get enough exercise” (P. A. Plazier et al., 2017), this is a good example of hedonic reasons to

(10)

10 2.2 The interrelation between instrumental, symbolic, and hedonic attributes towards the

intention to adopt an E-bike

The instrumental, hedonic and symbolic motivational dimensions can be related to as the instrumental, hedonic and symbolic attributes (Schuitema et al., 2013). When consumers focus on instrumental motives when adopting a product it can be expected that they focus on the instrumental attributes of a product. The same goes with hedonic and symbolic attributes (Schuitema et al., 2013). Schuitema, Anable, Skippon & Kinnear, (2013) also identified that the relationship between perceptions of instrumental attributes and the intention to adopt is mediated by the symbolic and hedonic attributes. For example, a higher price (instrumental) of an E-bike might have a positive effect on the perceived symbolic (status) attributes, which can lead to a higher intention to adopt. This thesis suggests that this same concept is applicable to E-bikes. For example, the hedonic enjoyment may derive from the speed of the bike. Thus, this thesis proposes the same framework and for that reason the following hypothesis is stated:

H1.1: The effect of instrumental attributes to adopt E-bikes is mediated through hedonic attributes.

H1.2 The effect of instrumental attributes to adopt an E-bike is mediated through symbolic attributes.

The importance of perceived attributes is different per consumer. The type of trip, also referred to as the type of mode, in this thesis suggests that this could moderate the effect of the instrumental, symbolic, and hedonic attributes towards the adoption of E-bikes (Kroesen, 2017; P. A. Plazier et al., 2017).

2.3 Type of mode and the effect on instrumental, symbolic and hedonic attributes

The type of mode is an important indicator of E-bike usage (Kroesen, 2017; P. A. Plazier et al., 2017). Commuting and leisure trips are the most made types of trips on bikes in the Netherlands

(Fietsplatform, 2018). Based on the results of Plazier et al. (2017) this ratio also is applicable to E-bikes in the Netherlands. For this reason this thesis makes a distinction between those two modes. Leisure trips are free time rides which can be defined as recreational trips like touring and sightseeing. Almost 50 percent of the people in the Netherlands made a leisure trip, with a minimum of an hour, in 2018. 32 percent of these leisure trips were done with an E-bike. Leisure trips are mostly made in the weekend. At last, the age group of leisure riders are considered to be seniors (Fietsplatform, 2018). Commuting trips are rides between one’s place of residence and place of work or study. Using E-bikes for commuting trips is gaining in popularity (P. A. Plazier et al., 2017).

(11)

11 important issue (Heinen, van Wee, & Maat, 2010). Speed might, for this reason, be an important buying motive. Leisure riders, however, do not have this time pressure and, for this reason, they value this attribute differently. This is in line with Heinen et al. (2010) and Plazier et al. (2018) who suggest that commuting conventional trips and conventional leisure trips are used for different motivations. Schuitema et al., (2013) also indicated that the importance of symbolic, instrumental and hedonic attributes differ across different types of electrical vehicles. In conclusion, these papers suggest, based on qualitative research, that the shopping motivations are different depending on the type of trip. Based on these claims the following hypothesis is stated:

H2: The effect of the perceived instrumental, hedonic and symbolic attributes to adopt E-bikes

is moderated by different types of trips (leisure vs commuting).

To be more specific, this thesis also proposes the following specific effects of types of modes on the instrumental, hedonic and symbolic perceived attributes towards the adoption of E-bikes:

Different types of trips and instrumental attributes

Commuting trips are a method of transportation between a person’s residence and work/study (Heinen et al., 2010). Anable and Gatersleben (2005) identified that consumers overall attach more value to instrumental motives when making commuting trips in comparison to leisure trips. One of the reasons is that convenience is valued higher for commuting trips than leisure trips. The instrumental attributes are perceived more important because the main reason is to achieve the goal of getting from a person’s residence to work/study. Obviously, overall bike commuters value convenience lower in comparison to people who use cars for commuting trips, but it still plays a role (Anable & Gatersleben, 2005). Also, flexibility and combining activities are an important downside, which also suggests that the instrumental reasons are important for commuters (P. A. Plazier et al., 2018).

For this reason, this thesis suggests the following theory that the instrumental attributes are more important for commuting trips in comparison to leisure trips. Therefore, the following hypothesis is proposed:

H 2.1 : Instrumental attributes are more important for commuting trips than for leisure trips

when adopting an E-bike.

Different types of trips and symbolic attributes

(12)

12 (students) may suffer from the negative image of E-bikes (P. A. Plazier et al., 2018). One of the reasons for this is that E-bike commuters are confronted with the fact that it is thought to be for old people only (P. A. Plazier et al., 2017). In comparison, most leisure riders are older people

(Fietsplatform,2019). For this reason commuters who use E-bikes will potentially perceive more harm from having an E-bike than leisure riders. The reason for this is that buying an E-bike can hurt their self-image. However, the second reason is that commuting riders use E-bikes in another social environment than leisure riders. People who use it for commuting trips therefore use the E-bike, indirectly, in the work environment. When you ride your E-bike for leisure trips you either cycle alone or with other people who are also likely to have an E-bike. Having an E-bike in this social

environment is the norm, this is less likely the case in work environments. Based on this, the following hypothesis is stated:

H 2.2 : The symbolic attributes are more important for commuting trips than for leisure trips

Different types of trips and hedonic attributes

While the hedonic attributes are perceived as important for both leisure and commuting trips, there are differences. Leisure trips can typically be associated with affective dimensions such as relaxation and a sense of freedom (Alba & Williams, 2013; Heinen et al., 2010). These are values related to

experience, and thus hedonic attributes. These attributes also play a role in commuting trips (Anable & Gatersleben, 2005; Jones et al., 2016; P. A. Plazier et al., 2017). Leisure trips, however, are primarily focussed on the hedonic values while this is not the case for commuting trips (Anable & Gatersleben, 2005). For this reason the following hypothesis is proposed:

H 2.3 : The hedonic attributes are more important for leisure trips than for commuting trips

2.4 Conceptual model

The first conceptual framework (figure 1), which is derived from section 2.3 is stated. The

(13)

13 Figure 1. Conceptual framework 1

Figure 2. Conceptual framework 2

3. METHODOLOGY

This section describes the methodology of this paper in order to test the hypotheses. First, the variables in the research design are described. In the second part the research method is described. In the third part of this section the data will be stated, this is done by describing the descriptives and the data will be analysed for missing values and outliers. Finally, the plan of analysis is described.

3.1 Data collection

(14)

14

Power and Effect size calculations

In order to identify the least amount of respondents needed to investigate the conceptual model in Figure 1 prior calculations have been made with the use of an online calculator (Soper, 2019). This calculator is based on a variety of literature (Cohen, 1998; Cohen, Cohen, West, & Aiken, 2002; Howlett, Abramowitz, & Stegun, 2007). The formulas used in this calculator can be found in

Appendix 2. The calculations were based on the conceptual model shown in Figure 1, specifically the first hypothesis. In order to test hypothesis 1 both a between and within subject design were used. Therefore the minimum amount of respondents was multiplied. For this reason the calculations were based on the first hypothesis.

An overview of the parameters can be found in Table 1. The anticipated effect size was 0.27. This was based on the research of Schuitema et al., (2013) who performed research on the relationship between the instrumental, symbolic and hedonic attributes and their effect on the intention to buy a full

electrical car. This model is similar to the model in this research, for this reason this thesis used this R2 value as input for our parameter. Transforming this R2 _{into an effect size gives a parameter of 0.37} (Cohen, 1998). The desired statistical power used in the calculation was 0.8, which is a default setting in these types of calculations. The number of predictors were 5: instrumental-, hedonic-, symbolic attributes and both the mediations. The minimum probability level this thesis aims for was 0.05. This results in a minimum of 42 respondents. As said, this needs to be multiplied. This means that at least 84 respondents were needed based on the power and effect size calculations.

3.2 Questionnaire, procedure and variables

This research was based on two hypotheses. 1: The effect of the instrumental attributes to adopt an E-bike is mediated by symbolic and hedonic attributes (Figure 1). 2: The importance of these attributes and their effects in this model (Figure 2) are different based on the type of trip the E-bikes are bought for. For this reason the respondents were asked if they are most likely to buy an E-bike for either leisure or commuting trips. Based on this answer they either get to answer questions about leisure trips or commuting trips.

Table 1 Parameters effect size calculator

Parameter Value

(15)

15 Respondents were asked about multiple statements in order to identify the perceived importance of instrumental, hedonic and symbolic attributes. Respondents were asked to agree or disagree with multiple statements on a 5 points Likert scale in order to identify the hedonic, symbolic and

Table 2

Statements for measuring the hedonic, symbolic and instrumental attributes and the intention to adopt an E-bike.

Intention to adopt (Commuting α= .778 , Leisure α = .814) Based on - I would buy an E-bike for leisure trips in the next 5 years

- I would consider an E-bike when buying a (next) bike for leisure trips

1 1

Instrumental attributes (Commuting α = .649, Leisure α = .613) - An E-bike would offer convenience to me when it is used for leisure trips

- An E-bike has high maintenance costs when it is used for leisure trips(r) - An E-bike has to be charged frequently when it is used for leisure trips(r)

- An E-bike does not have to be charged frequently when used for leisure trips

- An E-bike cannot be charged fast enough when it is used for leisure trips(r)

- An E-bike can be charged fast enough when it is used for leisure trips

- An E-bike has an insufficient range when it is used for leisure trips(r)

- The range of an E-bike is sufficient to ride leisure trips with it

- An E-bike is cheap in use when bought for leisure trips - Riding an E-bike for leisure trips is comfortable

- An E-bike is relatively expensive to buy for leisure trips(r)

- An E-bike is relatively cheap to buy when used for leisure trips

1,2 1,2, 3 1,2 1 1 1 1 1

Hedonic attributes (Commuting α = .748, Leisure α =.782) - Riding an E-bike are pleasant to drive when used for leisure rides - Riding an E-bike is unpleasant when used for leisure trips(r) - Riding an E-bike for leisure would give me a sense of freedom - Riding an E-bike for leisure trips would give me enjoyment

- Riding an E-bike for leisure trips would make me feel good about myself - Riding an E-bike for leisure trips would make me feel bad about myself(r)

2 3 2,3 4

Symbolic attributes (Commuting α = .685 , Leisure α =.782) - Riding an E-bike for leisure trips would fit my lifestyle

- Riding an E-bike for leisure trips would not fit my lifestyle(r) - Riding an E-bike for leisure trips would give me no status(r) - Riding an E-bike for leisure trips would give me status - Riding an E-bike for leisure trips feels like cheating

- I would feel accepted if I had an E-bike for my leisure trips

- Riding an E-bike for leisure trips would make me feel environmental friendly

- Riding an E-bike for leisure trips would not make me feel environmental friendly(r) - I would be ashamed for having an E-bike when using it for leisure rides

2 1 2,3 1 2 2 (r)=reverse coded

*1= Noppers et al., (2015) 2= (Jones et al., 2016) 3= (P. A. Plazier et al., 2017) 4= (P. A. Plazier et al., 2018)

(16)

16 instrumental values. The statements can be found in Table 2. The ratings range from (1) strongly disagree to (5) strongly agree. The statements were based on several researches (Jones et al., 2016; Noppers et al., 2015; P. A. Plazier et al., 2017, 2018). A standard procedure in the development of Likert scales is to incorporate reverse questions to control for acquiescence response bias. This indicates that respondents have a tendency to agree with statements, also referred as “yeasaying” (Schriesheim & Hill, 1981). For this reason some questions were also asked in reverse. Another benefit of this is that it can help check for non-serious respondents.

A 5 point Likert scale was also used in order to measure the intention to adopt. At the end of the survey consumers were first asked how likely it is that they will adopt an E-bike for either leisure trips or commuting trips in the upcoming 2 years where they had to respond from (1) very unlikely and (5) very likely. Secondly respondents were asked if they will consider buying an E-bike when buying a next bike for leisure/commuting trips. Both these statements together were a good indicator of the willingness to adopt innovative products (Noppers et al., 2015). At last, respondents were asked about some demographics to get insight into the population. The demographics that have been asked were: age, gender and educational level.

Creating the constructs willingness to adopt, instrumental, symbolic and hedonic attributes

The statements that were used to create the three attributes and the willingness to adopt were analysed in order to test if the variables fit together. A Cronbach alpha analysis was used in order to identify the internal consistency between the variables. There is a lot of debate about the threshold value of an Cronbach’s alpha. Most of the time, the higher the amount of variables the lower the Cronbach’s alpha (Hair, 2006). While most researches use a threshold value of 0.7, this thesis uses the threshold value of 0.6. This is an acceptable value in exploratory research, which is applicable to this research (Hair, 2006; Nunnally, 1978).

The intention to adopt an E-bike had an Cronbach’s alpha value of α=.778 for leisure rides and a Cronbach’s alpha value of α=.814 for commuting trips (Table 2) . This indicates that these statements fit together. The instrumental attributes had a value of α=.649 for commuting trips and a value of α=.613 for leisure trips. The statement “Riding an E-bike for leisure trips is comfortable” was

removed because of the weak fit for both leisure and commuting trips. However, the statement “An

E-bike cannot be charged fast enough when it is used for leisure trips” was left in despite the fact that it

(17)

17 3.3 Sample, data and descriptives

There were 143 respondents that participated in this study. Twenty-three respondents were excluded for further participating because they owned an E-bike. 7 respondents were deleted because of their low amount of fill in time, which indicates non serious answers. Further analysis on these 7

respondents showed that a lot of the reverse questions did not equal to the response they gave to the non-reverse questions. This was another indication that these 7 respondents were not serious in there answering. Age was the only continuous factor, 1 value was deleted for being an outlier since the respondent filled in an age of 167. This respondent also had a low fill in time. This resulted in a dataset of 112 respondents that were used to test the hypotheses.

Table 3 Descriptives

Description Number of

respondents

Mean S.D Structure of the data

Age 112 33.76 16.49 Ratio

Gender** 112 1.55 0.50 Dichotomous

Leisure/commuting trips 112 0.58 0.50 Dichotomous Intention to adopt an E-bike

*Commuting *Leisure 112 65 47 2.81 2.78 2.85 1.10 1.01 1.23 Ordinal Instrumental attributes *Commuting *Leisure 112 65 47 3.00 3.08 2.899 0.39 0.40 0.38 Ordinal Hedonic attributes *Commuting *Leisure 112 65 47 3.49 3.44 3.57 0.63 0.59 0.68 Ordinal Symbolic attributes *Commuting *Leisure 112 65 47 3.10 3.14 3.02 0.60 0.55 0.66 Ordinal ** 1 = Male, 2 = Female

(18)

18 3.4 Plan of analysis and model specifications

In order to test the first hypothesis a mediation analysis was done. The mediation model was analysed with Process (Preacher, Rucker, & Hayes, 2007). First, the main effect of instrumental variables and their effect on the mediators (hedonic and symbolic attributes) were tested in the first 2 formulas. In the last formula the effect of both the independent and moderating variables were tested. After the last formula a bootstrapping method was used to test the indirect effect of the instrumental variables through the moderators, (Preacher et al., 2007). The data has been mean centred in order to make interpretation of the data more meaningful. This hypothesis was tested three times, first for leisure trips, then for commuting trips and at last with the data combined.

With the first three models H1 will be tested: The direct effect of perceived instrumental attributes to

adopt E-bikes is mediated by perceived symbolic and hedonic attributes.

MODEL 1(Leisure): YHEDONIC(L)= ß0 + ß1XINSTR(L) + R YSYMBOLIC(L)= ß0 + ß1XINSTR(L) + R YADOPT(L)= ß0 + ß1XINSTR(L) + R

YADOPT(L)= ß0 + ß1XHEDONIC(L)+ ß2XSYMBOLIC(L) +

ß3XINSTR(L) + R

MODEL 2(Commuting): YHEDONIC(C)= ß0 + ß1XINSTR(C) + R YSYMBOLIC(C)= ß0 + ß1XINSTR(C) + R

YADOPT(C)= ß0 + ß1XINSTR(C) + R

YADOPT(C)= ß0 + ß1XHEDONIC(C)+ ß2XSYMBOLIC(C) +

ß3XINSTR(C) + R

MODEL 3(Both): YHEDONIC= ß0 + ß1XINSTR + R YSYMBOLIC= ß0 + ß1XINSTR + R

YADOPT= ß0 + ß1XINSTR + R

YADOPTION= ß0 + ß1XHEDONIC+ ß2XSYMBOLIC + ß3XINSTR

+R

In order to test H2, 2.1 , 2.2 and 2.3 model 2 and 3 a moderation analyses was done with the use of Process (Preacher et al., 2007). Model 4 was used in order to test hypothesis 2. The moderator, type of trip, was dummy coded. 0 means leisure trips where 1 means commuting trips. The data has been mean centred in order to make an interpretation of the data more meaningful. The moderators were created by multiplying the attributes and the moderator (type of ride).

(19)

19 The assumptions of the regression analysis have been met and the analysis can be found in Appendix 3. There were some concerns about the autocorrelation of the leisure data subset, these were taken into account. However, as shown in Chapter 4, the results of the leisure data subsets was not different compared to the other data subsets which had no autocorrelation. For this reason this thesis assumed that the autocorrelation had no influence on the results.

4. RESULTS

In this chapter the results of the tests are presented that are used to test both the hypotheses. First, H1 is tested which suggests that the effect of instrumental attributes and its effect on the willingness to adopt an E-bike is mediated through hedonic and symbolic attributes. Secondly this thesis tests if the main effects of instrumental, symbolic and hedonic attributes are moderated by the type of ride (H2). 4.1 The mediation effect between instrumental attributes and symbolic and hedonic attributes The mediation effect is tested three times, first the hypothesis is tested with a subset of the dataset with leisure trips only. The second subset of the dataset consists of people who said that they would buy an E-bike for commuting purposes. The last subset consists of respondents with both the leisure

respondents and commuting trips respondents.

Mediation model 1: Leisure trips (Figure 3)

In the first step of the mediation model, the A1 path was tested. The regression of instrumental attributes and its effect on the symbolic attributes was significant (b=0.82, p<.01). The second part tested the A2 path. The second linear regression showed a significant effect of instrumental attributes on the hedonic attributes (b=1.11, p<.01). In the third part the C path was tested, this regression tested the effect of the instrumental attributes on the adoption of an E-bike without controlling for the symbolic and hedonic attributes. This effect was significant (b=1.12, p<.05). In the last regression analysis the C` path and the direct effect of symbolic and hedonic attributes on the adoption of an E-bike (B1 and B2 path) were tested. The effect of instrumental attributes was not significant anymore after controlling for the mediators (b=0.20, p=.67). The effect of symbolic attributes towards the adoption of an E-bike was not significant (b=0.15, p=0.63). However, the effect of the hedonic

(20)

20

Mediation model 2: commuting trips (Figure 4)

(21)

21 with controlling for the mediators was significant as well (b=0.61, p<.01). The overall model was significant (F= 18.79, P<.01) and has a R2_{value of .69. A bootstrapping method with 10.000 samples} showed a significant indirect effect of instrumental attributes through hedonic attributes on the willingness to adopt an E-bike (CI= 0.37-1.20). Since the instrumental attribute still had a significant direct effect on the willingness to adopt an E-bike the effect was only partially mediated through hedonic attributes. Thus, H1.1 is accepted. Lastly, the indirect effect through the symbolic ab path is insignificant (CI= -0.27–0.18). This is in line with the insignificant direct effect of symbolic attributes towards the adoption of an E-bike. For this reason H1.2 is rejected.

Mediation model 3: Leisure/commuting trips

In the third model the data of the first and second model were merged. In the first step a regression model showed a significant effect of the A1 path. The instrumental attributes had a significant positive effect on the valuation of the symbolic attributes (b=0.66, p<.01). The second part tested the A2 path, a regression analysis showed a significant positive effect of the instrumental attributes on the hedonic attributes (b=0.87, p<.01). The third part tested the C path, which tested the direct effect of

(22)

22 result showed a non-significant indirect effect of instrumental attributes through the symbolic

attributes. For this reason H1.1 is accepted while H1.2 is rejected.

Conclusion mediation analyses

In all three the models the instrumental attributes had a significant positive effect on the willingness to adopt an E-bike. This effect however was mediated through the hedonic attributes. In other words, the higher the valuation of the instrumental attributes, the higher the valuation of the symbolic attributes which results in a higher willingness to adopt an E-bike. For this reason hypothesis 1.1,“H1.1: The

effect of instrumental attributes to adopt E-bikes is mediated through hedonic attributes” is accepted.

However, there was a full mediation in the leisure trip and combined model while there was only a partial mediation of hedonic attributes in the commuting model. Thus, the valuation of instrumental attributes had no direct effect in the leisure trip and combined model while this was not the case in the commuting trip model. This is an indication that the effect of the models is different depending on the type of ride, this will be further tested in chapter 4.2.

In all the models the instrumental attributes had a positive significant effect on the symbolic attributes. However, in none of the models symbolic attributes had a significant effect on the willingness to adopt an E-bike. In addition, the bootstrap methods showed an insignificant indirect effect of the

instrumental attributes. For this reason hypothesis 1.2, “H1.2 The effect of symbolic attributes to adopt

an E-bike is mediated through symbolic attributes”, is rejected.

4.2 The moderation effect of type of trip

A regression analysis was done to test the relation of instrumental, symbolic and hedonic attributes to the intention to adopt an E-bike. More specifically, to test if this relationship is moderated by the type of ride that the E-bike is bought for (Table 4). The model contains the instrumental attributes,

symbolic attributes and hedonic attributes, in addition the three moderators have been added. An ANOVA test showed that the overall model was significant (F(105,6)=14.29, p<0.01). The model explains the variation in the dependent variable for 67% (R2_{= .67). The instrumental attributes have} negative effect on the intention to adopt an E-bike, however this effect is not significant (b=-0.08, p=.98). In addition, the type of ride (Type*instrumental) did not have a significant positive effect on the relationship between the instrumental attributes and the intention to adopt an E-bike (b=0.50, p=.20). Thus, the effect of instrumental attributes to the adoption of an E-bike became more positive when the E-bike would be bought for commuting trips. However, since the relation is insignificant hypothesis 2.1, “Instrumental attributes are more important for commuting trips in comparison to

leisure trips when adopting an E-bike”, is rejected.

(23)

23 H2.2, “The symbolic attributes are more important for commuting trips in comparison to leisure trips

when adopting an E-bike”, is rejected.

At last, the hedonic attributes had by far the largest influence on the intention to adopt an E-bike. This effect was also significant (b=1.35, p<0.01). The moderator in the model implied that commuting trips had a negative effect on the relation between hedonic attributes and the intention to adopt an E-bike, this effect however was not significant (b=-0.42, p=.29). For this reason hypothesis 2.3, “The perceived importance of hedonic attributes are more important for leisure trips in comparison to commuting trips”, is rejected.

Table 4: Regression analysis

Variable b (Constant) -1.56 Instrumental attributes -0.08 Symbolic attributes -0.13 Hedonic attributes 1.35* Type*Hedonic -0.42 Type*Symbolic 0.02 Type*Instrumental 0.50 F 14.29* R2 0.67

a. Dependent Variable: Intention to adopt an E-bike *, P < 0.01

4.3 Instrumental, symbolic and hedonic attributes and the intention to adopt E-bikes-exploratory analyses of alternative models

(24)

24 imply that the interrelationships might be even more complex than suggested in our conceptual model (Figure 1). To be more specific, in this part, I would like to explore whether the symbolic attributes mediate the relationship between instrumental and hedonic attributes. Thus, the instrumental attributes have an indirect effect on the intention to adopt through symbolic attributes and sequential through hedonic attributes (see Figure 7: Alternative conceptual model).

This alternative model theoretically makes sense as well. The literature suggests that symbolic attributes have some negative consequences for E-bike users (Jones et al., 2016; P. A. Plazier et al., 2018). These negative symbolic consequences might affect the valuation of the hedonic attributes. For example, a consumer might not feel comfortable using an E-bike because of jokes they receive from other people (Jones et al., 2016; P. A. Plazier et al., 2018). As indicated, negative symbolic aspects might lead to a negative self-perception (Szmigin & Piacentini, 2018). It is likely that these consumers would have a less positive experience when using an E-bike in comparison to consumers for who this is not the case. Thus, there seems to be some theoretical evidence for the alternative mediating model as shown in Figure 7.

In order to exploratory test this sequential mediation analysis was done, similar to the one in section 6.1. In the first step of the mediation analysis the A1 path was tested. There was a significant effect of the instrumental attributes on the symbolic attributes (b= 0.66, p<.01). The A2 path also showed a significant effect; the instrumental attributes had an effect on the hedonic attributes (b=0.51, p<.01). The third part tested the effect of the instrumental attributes on the intention to adopt an E-bike without controlling for the mediators. This effect was significant (b=1.21, p<.01). The effect of the symbolic attributes on the hedonic attributes turned out to be significant as well (b=0.53, p<.01). The symbolic attributes had no significant effect on the intention to adopt an E-bike (b= -0.09, p= 0.61). Thus, the B1 path was insignificant while the B2 path (the effect between the hedonic attributes and the intention to adopt an E-bike) was significant (b= 1.09, p<.01). The C` path, which is the effect of instrumental attributes while controlling for the hedonic and symbolic attributes was not significant (ß=0.32, p>0.10). The overall model was significant (F(3,114)=28.29, P<.01), the R2_{value was 0.66.} In order to test the indirect effect of the instrumental attributes a bootstrapping method with 10.000 samples was used. The hedonic ab path was significant (b=0.56, 95% CI=0.32 – 0.82), meaning that the instrumental attributes had an indirect effect on the intention to adopt an E-bike through the hedonic attributes. The indirect effect of instrumental attributes through symbolic attributes however was not significant (b=-0.06, CI= 0.29 – 0.15). The indirect effect of instrumental attributes through both symbolic and sequential hedonic attributes was significant (b=0.38, CI= 0.19 – 0.62).

(25)

25

5. DISCUSSION

This section discusses the findings of this study, the conclusions and its contribution to current literature. First the conclusions of the results are discribed. Second, the theoretical and practical implications are discussed. At last, the limitations and future research options will be given. 5.1 Conclusions

The aim of this study was to identify the role of instrumental, hedonic and symbolic attributes and their effect on the process to adopt an E-bike. The three attributes were created based on theories in the adoption of E-bikes/innovation (Jones et al., 2016; Noppers et al., 2015; P. A. Plazier et al., 2017, 2018). More specifically, this thesis hypothesized that the effect of instrumental attributes on the intention to adopt an E-bike is mediated through symbolic and hedonic attributes. In addition, this research hypothesized that the relation between the instrumental, hedonic and symbolic attributes towards the adoption of an E-bike would be moderated by the type of ride. An online questionnaire with 112 respondents was held in order to test these hypotheses.

Discussion mediation effect of symbolic and hedonic attributes

(26)

26 field of electrical cars (Schuitema et al., 2013). This mechanism works as follows, for example, an E-bike needs to be fast (instrumental attribute) in order to gain enjoyment (hedonic attribute). This thesis sided with the latter view and thus the following hypotheses were constructed.

H1.1: The effect of instrumental attributes to adopt E-bikes is mediated through hedonic attributes. H1.2 The effect of symbolic attributes to adopt an E-bike is mediated through symbolic attributes.

Results confirmed H1.1, none of the models found support for H1.2. The instrumental attributes had ,in all the models, a positive significant effect on the symbolic attributes, however the symbolic attributes had no significant effect on the intention to adopt an E-bike. This is not in line with prior research, research found for both of the mediations significant results (Schuitema et al., 2013). An explanation for the differences in the results of symbolic attributes can be that the predictive power of symbolic attributes depend on the type of innovation (Vandecasteele & Geuens, 2010). It might be the case that symbolic attributes play a more important role for cars in comparison to the adoption of an E-bike. However, this does not explain the insignificant result of the symbolic attributes, especially since previous research identified downsides of owning an E-bike (P. A. Plazier et al., 2018). Important to mention is that the symbolic and hedonic attributes had indications of correlation with each other. The research of Schuitema et al. (2013) showed similar multicollinearity issues between these two

attributes. As mentioned in Section 4.3, this could be an indication of another more complex model than the model that was hypothesized.

This research further explored the theory of the adoption of innovation. In addition, this research analysed the additional effects of symbolic attributes towards the hedonic attributes in the mediation model. This thesis found correlational support for this relation and the theory supports this model as well. E-bike owners can experience negative symbolic consequences, for example jokes from co-workers (Jones et al., 2016; P. A. Plazier et al., 2018). This can result into a negative self-image (Szmigin & Piacentini, 2018). It is very imaginable that this would lead to a less positive experience (hedonic attribute) of riding an E-bike. Based on this research I suggest that, in addition to current literature, the instrumental attributes are mediated through symbolic attributes, hedonic attributes and both sequential symbolic to hedonic attributes. However, future research should further validate this relationship (see Section 5.4). Based on this research we can suggest that the symbolic attributes indeed have an effect on the adoption of an E-bike, it is however different than hypothesized in the first place. In the adoption process of E-bikes the symbolic attributes have no direct effect on the intention to adopt an E-bike but they do have an indirect effect through hedonic attributes.

Conclusion moderation analyses.

(27)

27 differ based on the type of ride. In addition, Anable and Gatersleben (2005) identified for example that consumers overall attach more value to instrumental motives when making commuting trips in

comparison to leisure trips. This thesis hypothesized that the type of ride moderates the effect of the instrumental, symbolic and hedonic attributes to the intention to adopt an E-bike. This resulted in the following three hypotheses.

H 2.1 : Instrumental attributes are more important for commuting trips in comparison to

leisure trips when adopting an E-bike.

H 2.2 : The symbolic attributes are more important for commuting trips in comparison to

leisure trips when adopting an E-bike.

H 2.3 : The perceived importance of hedonic attributes are more important for leisure trips in

comparison to commuting trips.

However, this thesis could not find any support for these hypotheses. This was surprising because the mediation study for Hypothesis 1 indicated for some moderation. The instrumental attributes were fully mediated through hedonic attributes for leisure trips while they were only partly mediated for commuting trips. This indicated for a difference in perceived importance of instrumental attributes. In addition, it is not in line with the results with the literature which suggested, based on exploratory research, that the advantages were depending on the type of trip (Jones et al., 2016; P. A. Plazier et al., 2018). An alternative explanation of the theory can be that the type of trip does not moderate the effect of instrumental, symbolic and hedonic attributes, but that it does affect the willingness to adopt in another way. It could be possible that the valuation of these attributes differ between the groups, but that the effect size is the same.

5.2 Theoretical implications

This research has several theoretical implications. Since research about E-bikes is relatively new, not a lot of research has been done on the process to adopt an E-bike. This thesis created a first construct for the instrumental, hedonic and symbolic attributes for further research into this topic. These constructs can be used and developed in future research in the field of E-bikes.

Secondly, this thesis contributes to the theory of adopting innovation (Noppers et al., 2015; Schuitema et al., 2013; Vandecasteele & Geuens, 2010). This thesis further develops the theory, hereby providing more knowledge on the mechanism how the instrumental, symbolic and hedonic attributes are related to each other. More specifically, the effect of instrumental attributes to the adoption of technology is mediated through the symbolic and hedonic attributes. This thesis further validates that the

instrumental attributes are mediated through the hedonic attributes. In addition, this research could not find support for the mediation through symbolic attributes. However, as discussed, this thesis

(28)

28 5.3 Managerial implications

This research has some important managerial implications. This research identified that not the instrumental, often implied by marketeers, but mostly the hedonic attributes are important in the adoption process. This is important for E-bike marketeers, they should focus on the hedonic benefits of buying an E-bike. However, focussing on the hedonic attributes is not enough, since the valuation of instrumental attributes and symbolic attributes are an important indicator of the valuation of hedonic attributes. Thus, marketeers should inform potential adopters of an E-bike about the potential hedonic benefits they can gain, backed up by the instrumental attributes. For example, ease of use could be used as an argument, which will create more pleasure for the user (Jones et al., 2016). This is also important for E-bike designers, since they should focus on developing E-bikes that can create more hedonic benefits for the users.

5.4 Limitations and future research

The first limitation of this research paper is the development of the theoretical constructs of instrumental, hedonic and symbolic attributes. Since the E-bike literature is relatively new the constructs were created based on other fields of research. This is important to take into account, for example some statements for symbolic attributes in E-car literature might not be ideal to measure symbolic attributes in E-bike literature (Noppers et al., 2015; Schuitema et al., 2013). All the statements have been validated based on E-bike literature, however most of these articles are qualitative and explanatory (Jones et al., 2016; P. A. Plazier et al., 2018). Moreover, the factors that were created were explanatory in the field of E-bikes, which resulted in some relatively low Cronbach alpha values. These factors were sufficient in order to test the hypotheses, unique contributions of the mediators however should be interpreted with caution when simulated to a real life situation.

Another limitation of this research is the correlation between the mediators hedonic attributes and symbolic attributes. Although the correlation between the factors meets the requirements for no multicollinearity, there is a high amount of correlation between the factors. For this reason,

(29)

(30)

30

6. REFERENCES

Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behaviour and Human

Decision Processes, 50, 179–211.

Alba, J. W., & Williams, E. F. (2013). Pleasure principles : A review of research on hedonic consumption. Journal of Consumer Psychology, 23(1), 2–18.

Anable, J., & Gatersleben, B. (2005). All work and no play ? The role of instrumental and affective factors in work and leisure journeys by different travel modes. Transportation Research Part

A,39, 163–181.

Arnold, M. J., & Reynolds, K. E. (2003). Hedonic shopping motivations. Journal of Retailing, 79, 77– 95.

Blumberg, B., Cooper, D., & Pamela, S. S. (2014). Business research methods (4th edition). McGraw-Hill.

Cialdini, R. B., & Goldstein, N. J. (2004). Social Influence: Compliance and Conformity. Annual

Review of Psychology, 55(1), 591–621.

Cohen, J. (1998). Statistical power analysis for the behavioral sciences. Hillsdale: Lawrence Earlbaum Associates.

Cohen, J., Cohen, P., West, S., & Aiken, L. (2002). Applied Multiple Regression/Correlation for the

Behavioral Sciences. New York: Psychology Press.

Fietsplatform, 2018. Kerncijfers effecten van fietsen in Nederland 2012/2013/2014/2015. Available at: <http://www.fietsplatform. nl/fietsrecreatiemonitor/cijfers>

Fishman, E., & Cherry, C. (2016). E-bikes in the Mainstream: Reviewing a Decade of Research.

Transport Reviews, 36(1), 72–91.

Fyhri, A., & Fearnley, N. (2015). Effects of e-bikes on bicycle use and mode share. Transportation

Research Part D, 36, 45–52.

Gatersleben, B., & van der Werff, E. (2018). Symbolic Aspects of Environmental Behaviour. In

Environmental Psychology (pp. 198–206).

(31)

31 Heinen, E., van Wee, B., & Maat, K. (2010). Commuting by bicycle: An overview of the literature.

Transport Reviews, 30(1), 59–96.

Hirschman, E. C., & Holbrook, M. B. (1982). Hedonic consumption: Emerging Concepts, Methods, and propositions. Journal of Marketing, 46(3), 92–101.

Hirschman, E. C. (1984). Experience seeking: A subjectivist perspective of consumption. Journal of

Business Research, 12(1), 115–136.

Howlett, J., Abramowitz, M., & Stegun, I. A. (2007). Handbook of Mathematical Functions. The

Mathematical Gazette (Vol. 50). New York: NY Dover.

Jans, L., & Fielding, K. (2018). The Role of Group Processes in Environmental Issues, Attitudes, and Behaviours. In Environmental Psychology (pp. 228–237).

Jones, T., Harms, L., & Heinen, E. (2016). Motives , perceptions and experiences of electric bicycle owners and implications for health , wellbeing and mobility. Journal of Transport Geography,

53, 41–49.

Keizer P.W., K. S. (2012). Social Norms and Pro Environmental Behaviour. Environmental

Psychology: An Introduction, (2012), 153–163.

Kroesen, M. (2017). To what extent do e-bikes substitute travel by other modes? Evidence from the Netherlands. Transportation Research Part D: Transport and Environment, 53, 377–387. Midgley, D. F., & Dowling, G. R. (2002). Innovativeness: The Concept and Its Measurement. Journal

of Consumer Research, 4(4), 229.

Noppers, E. H., Keizer, K., Bockarjova, M., & Steg, L. (2015). The adoption of sustainable

innovations: The role of instrumental, environmental, and symbolic attributes for earlier and later adopters. Journal of Environmental Psychology, 44, 74–84.

Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.

Park, J. K., & John, D. R. (2014). I Think I Can, I Think I Can: Brand Use, Self-Efficacy, and Performance. Journal of Marketing Research, 51(2), 233–247.

Peine, A., van Cooten, V., & Neven, L. (2016). Rejuvenating Design. Science, Technology, & Human

Values, 42(3), 429–459.

(32)

32 Plazier, P. A., Weitkamp, G., & Berg, A. E. Van Den. (2017). “ Cycling was never so easy ! ” An

analysis of e-bike commuters ’ motives , travel behaviour and experiences using GPS-tracking and interviews ☆. Journal of Transport Geography, 65(June), 25–34.

Plazier, P. A., Weitkamp, G., & Berg, A. E. Van Den. (2018). Exploring the Adoption of E-Bikes by Different User Groups. Frontiers in Built Environment, 4(August).

Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42(1), 185–227.

Schriesheim, C. A., & Hill, K. D. (1981). Controlling Acquiescence Response Bias by Item Reversals: The Effect on Questionnaire Validity. Educational and Psychological Measurement, 41(4), 1101–1114.

Schuitema, G., Anable, J., Skippon, S., & Kinnear, N. (2013). The role of instrumental , hedonic and symbolic attributes in the intention to adopt electric vehicles. Transportation Research Part A,

48, 39–49.

Sirgy, M. J. (1997). Using Self-congruity and Ideal Congruity to Predict Purchase Intention. Journal

of Euromarketing, 6(1), 41–56.

Soper, D. S. (2019). A-priori Sample Size Calculator for Multiple Regression [Software]. Retrieved from http://www.danielsoper.com/statcalc

Steg, L., Vlek, C., & Slotegraaf, G. (2001). Instrumental-reasoned and symbolic-affective motives for using a motor car. Transportation Research Part F, 4, 151–169.

Stigler, G. J. . (1950). The Development of Utility Theory. Journal of Political Economy, 58(4), 307– 327.

Szmigin, I., & Piacentini, M. (2018). Consumer Behaviour (second edition). Oxford university. Vandecasteele, B., & Geuens, M. (2010). Motivated Consumer Innovativeness : Concept ,

measurement , and validation. International Journal of Research in Marketing, 27(4), 308–318. Voss, K. E., Spangenberg, E. R., & Grohmann, B. (2003). Measuring the Hedonic and Utilitarian

Dimensions of Consumer Attitude. Journal of Marketing Research, 40(3), 310–320.

(33)

33

7. APPENDIX

7.1 Appendix 1: Questionnaire

Start of Block: Introduction

Q16 Dear participant,

The following survey is part of my Master Thesis in the Department of Marketing at the university of Groningen. The purpose of this study is to identify the buying motivations for E-bikes among customers.

Your participation in this study will remain confidential, participation in this study is voluntary and you may withdraw at any given time by closing the browser. The study will take about 4-5 minutes to fill in.

If you have any questions about the survey, feel free to send me an email at : R.schipper.3@student.rug.nl

Thank you for your participation.

Rogier Schipper

Q1 Do you own an E-bike?

o

Yes (1)

o

No (2)

Skip To: End of Survey If Do you own an E-bike? = Yes

Q2 What is your gender?

o

Male (1)

o

Female (2)

(34)

34 Q17

What is your highest finished education?

o

Primary school (1)

o

Secondary school (2)

o

Informational Vocational Education (mbo) (3)

o

HBO Bachelor (4)

o

University Bachelor (5)

o

Master (6)

o

PhD (7)

o

Other namely: (8) ________________________________________________ Q3 What is your age?

________________________________________________________________ Q5 If I would buy an E-bike I would most likely use it for (choose 1)

o

Leisure trips (1)

o

Commuting trips (includes study trips) (2) End of Block: Introduction

Start of Block: Leisure

Display This Question:

If If I would buy an E-bike I would most likely use it for (choose 1) = Leisure trips

Q6 Please rate the following statements on agreeability

(35)

35

An E-bike would offer me convenience when used for

leisure trips (1)

o

An E-bike has high maintenance costs when it is

used for leisure trips (2)

o

Riding an E-bike is pleasant when used for leisure trips

(3)

o

Riding an E-bike for leisure trips would fit my lifestyle

(4)

o

An E-bike has to be charged frequently when it is used

for leisure trips (5)

o

Riding an E-bike for leisure trips would give me

enjoyment (6)

o

Riding an E-bike for leisure trips would give me no

status (7)

o

An E-bike has an insufficient range when it is

o

An E-bike is cheap in use when bought for leisure trips

(36)

36

Q8 Please rate the following statements

Riding an E-bike is unpleasant when used for

leisure trips (10)

o

Riding an E-bike for leisure

trips feels like cheating (11)

o

An E-bike is relatively expensive to buy for leisure

trips (12)

o

An E-bike does not have to be charged frequently when

o

An E-bike cannot be charged fast enough when it

is used for leisure trips (14)

o

Strongly disagree (1) Somewhat disagree (2) Neither agree nor disagree (3) Somewhat agree (4) Strongly agree (5) I would feel accepted if I had an E-bike for leisure trips (1)

o

trips is

(37)

37 An E-bike can be charged fast enough when it is used for leisure trips (3)

o

An E-bike is relatively cheap to buy when used for leisure

trips (4)

o

Riding an E-bike for leisure trips would give

me a sense of freedom (5)

o

trips would make me feel bad about myself (6)

o

trips would make me feel environmental friendly 7)

o

The range of an E-bike is sufficient to ride leisure trips with it (8)

o

trips would make me feel good about myself (9)

o

Riding an E-bike for leisure trips would not fit my lifestyle

(10)

o

Riding an E-bike for leisure trips would not

make me feel environmental

friendly (11)

(38)

38

Q9 I would buy an E-bike for leisure trips in the next 5 years

o

Strongly disagree (1)

o

Somewhat disagree (2)

o

Neither agree nor disagree (3)

o

Agree (4)

o

Strongly agree (5)

If If I would buy an E-bike I would most likely use it for (choose 1) = Leisure trips Riding an

E-bike for leisure trips would give

me status (12)

o

I would feel ashamed to ride

an E-bike for

(39)

39 Q11 I would consider an E-bike when buying a (next) bike for leisure trips

o

Strongly disagree (1)

o

Somewhat disagree (2)

o

Neither agree nor disagree (3)

o

Somewhat agree (4)

o

Strongly agree (5) End of Block: Leisure

Start of Block: Commuting trips

If If I would buy an E-bike I would most likely use it for (choose 1) = Commuting trips (includes study trips)

Q13 Please rate the following statements on agreeability

Strongly disagree (1) Somewhat disagree (2) Neither agree nor disagree (3) Somewhat agree (4) Strongly agree (5) An E-bike would offer convenience when used for

The adoption of E-bikes

The adoption of E-bikes

The relationship of instrumental, hedonic and symbolic attributes and the type of

trip towards the adoption of E-bikes.

The adoption of E-bikes

The relationship of instrumental, hedonic and symbolic attributes and type of

trip towards the adoption of E-bikes.

DATE 2019

University of Groningen

MSc Marketing Management

Department of Marketing

Faculty of Economics and Business

Author:

First supervisor:

Rogier Schipper

Dr. J.I.M. de Groot

Verlengde Hereweg 20-1E

University of Groningen

9722 AC Groningen

ABSTRACT

TABLE OF CONTENTS

1. INTRODUCTION

2. THEORETICAL BACKGROUND

3. METHODOLOGY

4. RESULTS

5. DISCUSSION

6. REFERENCES

7. APPENDIX

7.1 Appendix 1: Questionnaire

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o