• No results found

The influence of extrinsic and intrinsic motivations on consumers’ participation in the sharing economy

N/A
N/A
Protected

Academic year: 2021

Share "The influence of extrinsic and intrinsic motivations on consumers’ participation in the sharing economy"

Copied!
64
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The influence of extrinsic and intrinsic

motivations on consumers’ participation in the

sharing economy

Andrea Popma

Student number: 6167446

MSc Business Administration (Marketing track)

Supervisor: Anouar El Haji

23 December 2014

Bot sm an & R og er s (2 01 1)

(2)

Abstract

This thesis studies motivators that drive consumers to participate in the sharing economy. Both intrinsic and extrinsic motivators are tested by means of an experimental auction. Intrinsic motivations include ‘sharing an experience’ and ethical consumption. Extrinsic motivations comprehend the prospect of social capital and cost savings. Furthermore, we test the moderating role of prior knowledge and perceived risk on the relationship between the various motivations and consumers’ willingness to pay. None of the hypotheses were confirmed. This suggests that the sharing economy is potentially not based on ideology, but depends on the principles of regular market exchange transactions.

(3)

Table of Contents

1. Introduction ... 4

2. Literature Review ... 6

2.1 Sharing ... 6

2.1.1 Intrinsic and extrinsic motivations ... 7

2.2 Sharing and the Internet ... 8

2.2.1. Motivations for online file sharing ... 9

2.2.2 Motivations to share knowledge online ... 10

2.3 The sharing economy and collaborative consumption ... 12

2.3.1 Participation in the sharing economy ... 13

2.4 Ethical consumption ... 17

2.5 Social capital ... 17

2.6 Perceived risk ... 18

2.7 Gaps in the literature ... 19

3. Conceptual framework ... 20

3.1 The sharing experience ... 20

3.2 Ethical consumption ... 21 3.3 Social capital ... 22 3.4 Cost savings ... 23 3.5 Perceived risk ... 24 3.6 Prior knowledge ... 24 4. Methodology ... 26 4.1. Research design ... 26 4.2 Auctioned product ... 27 4.3 Procedure ... 27 4.4 Treatments ... 28 5. Results ... 31 5.1 Sample characteristics ... 31 5.2 Descriptives ... 31 5.3 Main analyses ... 34

5.4 Moderator roles of perceived risk and prior knowledge ... 36

6. Discussion ... 43

6.1 Key findings ... 43

6.2 Managerial implications ... 46

6.3 Limitations & further research ... 47

7. References ... 49

8. Appendix ... 58

(4)

1. Introduction

The business landscape has been undergoing great shifts in power. The rise of the Internet and social media has led to increased consumer empowerment thereby eroding the influence that brands once had (Benkler 2004; Fournier & Avery 2011; Gansky 2010; Labreque 2013; Shipman 2001). This shift in power gives rise to a new type of business model, namely one in which the firm facilitates a form of exchange between consumers (Botsman & Rogers 2011; Gansky 2010; Labreque 2013). These new business models are found in a range of different industries. Zopa (a peer-to-peer money lending service) and Kickstarter (a funding platform for creative projects) co-exist next to the traditional banking industry, UberPOP (a peer-to-peer ridesharing platform) challenges conventional taxi companies and Airbnb provokes the hotel industry by enabling users to rent accommodation from each other (Guttentag 2013).

These examples show that a trend towards the sharing economy or collaborative consumption has surfaced, which encompasses a consumer to consumer (C2C) model that enables consumers to effectuate transactions that involve underutilized assets on one side in exchange for a monetary fee on the other side (Botsman 2013a; Zervas 2013). It seems that the sharing economy is becoming increasingly popular; it was estimated that the C2C market alone was worth $26 billion dollars in 2013 (Botsman 2013b).

There appears to be a research gap in the examination of consumer behavior in the sharing economy. Possibly this is due to the fact that it is a relative new phenomenon. This thesis contributes to the literature by studying what motivates consumers to engage in the sharing economy. We study to what extent various intrinsic and extrinsic motivations influence consumers’ valuation of sharing economy transactions.

The intrinsic motivations that are examined are ‘sharing an experience’ and ethical consumption. ‘Sharing an experience’ refers to the idea that the activity of sharing itself results in feelings of enjoyment. Ethical consumption entails the notion that consumers’

(5)

decision-making processes are impacted by ethical concerns about the environment, community, etc.

The extrinsic motivators that are studied involve bridging social capital and cost savings. Bridging social capital refers to “loose connections between individuals who might provide useful information” (Ellison et al. 2007, p. 1146). Cost savings refer to financial motivations to participate in the sharing economy.

There are indications that firms operating in the sharing economy are eroding turnover from conventional B2C companies (e.g. Sweers & De Graaff 2014). It is thus of high managerial value to examine why consumers choose to engage in C2C transactions. With this information, B2C marketing strategy can be adjusted accordingly to trigger the decisive motivators that consumers would otherwise only experience in C2C transactions.

This thesis is structured as follows: first a literature review is presented providing background information on the sharing economy and previous research that examined motivations to share. Second, the hypotheses are discussed and a conceptual model is visualized. Third, the results are provided. Lastly, a discussion section is presented followed by the limitations and directions for further research.

(6)

2. Literature Review

This chapter starts by providing an overview of the literature on the concept of sharing. Second, intrinsic and extrinsic motivations are discussed, followed by a section on sharing behavior online. Third, the concepts of the sharing economy and collaborative consumption are discussed. Fourth, an overview of previous research that examined motivations for consumers to take part in the sharing economy is provided. Lastly, the concepts of social capital, ethical consumption and perceived risk are explained.

2.1 Sharing

Sharing can be defined as “the act and process of distributing what is ours to others for their use as well as the act and process of receiving something from others for our use” (Belk 2007, p. 27). Often consumer behavior is mistakenly seen as either gift giving or economic exchange, thereby overlooking the potential influence of sharing (Belk 2009).

Sharing is different from gift giving and commodity exchange on two dimensions, namely ownership and reciprocity. Gift giving and commodity exchange emphasize private ownership, whereas sharing centers around joint or serial ownership (Albinsson & Perera 2012). Furthermore, in economic theory, a commodity exchange entails a reciprocal transaction (either through bartering or the medium of money) where neither of the parties is left with feelings of indebtedness (Belk 2007). Gift giving can also involve a form of reciprocity, albeit less explicit than the concrete reciprocity involved in a commodity exchange; it is mainly driven by a cycle of “give, receive and reciprocate” (Mauss, cited in Belk 2007, p. 128) and can contain a form of calculating self-interest (Belk 2009). Both sharing and gift giving have the capacity to build a connection between the giver and the recipient (Belk 2009). However, it should be noted that this can also occur in the case of commodity exchange, for example when customers form a relationship with their supplier or a brand (Belk 2007).

Belk (2013, p. 1596) draws a distinction between ‘sharing in’, relating to the

(7)

“inclusive act that is likely to make the recipient a part of a pseudo-family and our aggregate extended self” and ‘sharing out’ which “involves dividing something between relative strangers or when it is intended as a one-time act such as providing someone with spare change, directions, or the time of day”. ‘Sharing in’ is thus more related to the core concept of sharing, whereas ‘sharing out’ is often associated with gift giving of commodity exchange (Belk 2009).

2.1.1 Intrinsic and extrinsic motivations

The motivation to share can differ. Often a distinction is drawn between intrinsic and extrinsic motivations. Intrinsic motivations refer to doing activities “for its inherent satisfactions and enjoyment rather than for some separable consequence” (Ryan & Deci 2000, p. 56). To determine whether intrinsic motivations are present, often two measures are deployed.

Firstly, it is possible to detect intrinsic motivation through experimental research applying behavioral dependence measures (Ryan & Deci 2000; Deci 1971). In these types of experiments, respondents are asked to work on a task with the prospect of receiving an external award upon completion (Wiersma 1992). At some point during the experiment, participants are told that they do not have to complete the task anymore but will receive the reward nonetheless (Wiersma 1992). The experimenter leaves the respondent by himself in the room with the task and other distractor activities (Ryan & Deci 2000). When the respondent decides to allocate more time during this free choice period to complete the task, he is presumed to do this due to intrinsic motivation, as he already earned the external reward (Wiersma 1992).

The second measure to detect intrinsic motivation is “the use of self-reports of interest and enjoyment of the activity per se” (Ryan and Deci 2000, p. 57). An example of this can be found in Dewett (2007). This research aims to determine whether intrinsic motivation sparked creativity on the work floor (Dewett 2007). Respondents were asked to

(8)

self report their intrinsic motivation through questions such as “I feel driven to do my job because I genuinely like the tasks I work on” (Dewett 2007, p. 201). An example of sharing led by intrinsic motivation could be the pleasure experienced by sharing a meal with a loved one. The activity itself is where the satisfaction of enjoyment stems from; a concrete reward afterwards is thus not necessary (Deci et al. 1999).

Extrinsic motivation on the other hand is triggered by a separable outcome (Ryan and Deci 2000). It focuses on the instrumental value of an activity (Ryan and Deci 2000), and often results in receiving some type of reward (Eccles and Wigfield 2002). Extrinsic motivation can differ in terms of the degree of autonomy involved (Ryan and Deci 2000). For example, it could be that an employee completes a report out of fear for sanctions from his manager, thereby not possessing a high degree of autonomy. It could also be that the employee completes the report because he or she is hoping that it will help to get a promotion. In the latter case, the employee will experience more autonomy. However, the employee still completes the report for its instrumental value (to get the promotion) rather than the actual enjoyment of the activity itself. An example of extrinsic motivation for sharing could be the reception of praise after giving a colleague a ride to work.

2.2 Sharing and the Internet

The emergence of the Internet has enabled online users to share on a larger scale (Belk 2014a). One of the consequences of sharing online is the development of online piracy. Online or Internet piracy “refers to the unauthorized use or reproduction of copyrighted or patented (electronic) material, such as music or software files” (Choi & Perez 2007, p. 168). It encompasses online copyright infringements (Samuelson 2012).

Online piracy has had a significant impact on various industries. For example, it is estimated that online piracy costs the US economy between $200 and $250 billion a year (Raustiala and Sprigman 2012). Perhaps due to the major financial impact on for example the music and movie industry, a substantial amount of research has focused on motivations

(9)

for consumers to share online and how traditional industries should best deal with these developments.

In the remaining of this section, an overview of various articles that focus on motivational drivers for several sharing practices online (ranging from files to knowledge) is provided.

2.2.1. Motivations for online file sharing

Hennig-Thurau et al. (2007) design a utility framework explaining the drivers for consumers to illegally share motion pictures online. They found that collection utility (illegal copies allow consumers to have an extensive movie collection), knowledge about file sharing, and transaction costs (e.g. saving money) are the top three determinants of file sharing behavior. Furthermore, they also found that anti-industry utility proves to have a significant impact. Anti-industry utility relates to the concept of watching illegally downloaded movies to make a statement against the movie industry because it supposedly treats movies purely as commercial objects, rather than art. Thus, the anti-industry utility implies that sharing motion pictures online is not solely about extrinsic motives such as saving costs.

Plouffe (2008) too investigated peer-to-peer (P2P) sharing activities online. He administered a survey to college students to research their motivations to share files on P2P systems. He found that participation in these systems was driven by convenience and community connectedness amongst other things. Convenience is related to the ease of consumption. Through P2P sharing, music and movies are available whenever needed. Community connectedness refers to a sense of identification online among the P2P users based on “similar interests, styles and preferences” (Plouffe 2008, p. 1184). Convenience and community connectedness in turn had a positive effect on P2P satisfaction, which subsequently influenced future intentions for P2P systems usage. Interestingly, Plouffe (2008) also tested assortment (more choice in offer) but unlike the research conducted by Hennig-Thurau et al. (2007), Plouffe (2008) did not find a significant influence.

More recently, Argan et al. (2013) also conducted a research on what drives

(10)

consumers to engage in P2P file sharing (e.g. music, movies). Their results yield four categories that influence file sharing motivations. Firstly, similar to Hennig-Thurau et al. (2007) and Plouffe (2008), convenience showed to have a significant influence, due to easy and free access of files. Secondly, experience sharing is also a significant motivation to participate. Experience sharing refers to the idea that users feel part of an online community. This is thus similar to the community connectedness that Plouffe (2008) conceptualized. The third category that influences motivation in P2P file sharing is innovation adoption. When users are prone to adopt new innovations, they also display more behavioral intentions to share files online. Lastly, Argan et al. (2013) found evidence that altruism also functions as a motivator. This entails the concept of enjoyment experienced through helping and supporting others.

2.2.2 Motivations to share knowledge online

Kwok and Gao (2004) hypothesize an explanatory framework on what motivates consumers to share their knowledge online in a P2P network. They divide motivational factors into two categories: interpersonal motivations and individual factors. Interpersonal motivations include liking and affiliation. Liking is “an affection based on admiration, benevolence, or common interests” (Kwok & Gao 2004, p. 99). Affiliation in turn “refers to value derived from the connection to the group” (Kwok & Gao 2004 p. 99).

Individual factors are subdivided into extrinsic and intrinsic motivations. Extrinsic motivations include rewards and personal needs. Personal needs refer to compensations that will benefit the user in a way, for example through increasing the self-esteem. Intrinsic motivators to contribute knowledge to a P2P network are altruism and reputation. Altruism is “the behavior of someone that, although not beneficial or perhaps even harmful to oneself, benefits others” (Kwok & Gao 2004, p. 98). Reputation “refers to the overall quality or character as seen or judged by the community, or the recognition of some specific contribution to the community by other peers” (Kwok & Gao 2004, p. 98). A critical remark

(11)

here is that other research (e.g. Wasko Faraj 2005) labels reputation as an extrinsic rather than an intrinsic motivator. The instrumental value of the activity, namely gained reputation, is the driver of sharing knowledge and not so much the inherent pleasure of the activity itself.

Wasko and Faraj (2005) conducted a quantitative research on what motivates members of a legal professional organization to share their knowledge on an online member platform. Firstly, as conceptualized by Kwok and Gao (2004) as well, reputation showed to be of significant influence (Wasko and Faraj 2005). Secondly, the centrality of an individual (measured by the number of social ties in the network) also impacts the motivation to share knowledge; the higher the levels of network centrality, the more likely it is that the individual will contribute (Wasko and Faraj 2005). Interestingly, the enjoyment of helping others, self-rated expertise (displaying knowledge, skills and abilities (KSA’s)), and reciprocity were not found to be significant motivational drivers (Wasko and Faraj 2005).

Nov (2007) investigated what drives online users to contribute knowledge to Wikipedia. He conducted a survey among Wikipedia contributors (‘Wikipedians’) and found five significant drivers that trigger the motivation to share knowledge. Firstly, users contribute because the experience of contributing is seen as a fun activity. Secondly, the opportunity to express altruistic values also drives users to contribute. These values can for example resemble the importance to help others. Note that Wasko and Faraj (2005) could not provide evidence that the enjoying helping others had a significant positive impact. Thirdly, the opportunity to exercise knowledge, skills and abilities also drives Wikipedians to share their knowledge. This result is in conflict with the findings by Wasko and Faraj (2005) who also investigated whether self-rated expertise was of influence on the motivation to share, yet found no significant results. Fourth, the perception that contributing and exhibiting knowledge boosts the ego shows to be a significant driver for users, as conceptualized by Kwok and Gao (2004) as well. Lastly, Wikipedians also contribute for

(12)

career reasons by signaling their knowledge to potential employers.

Yang and Lai (2011) also use an online survey to study what influences knowledge sharing on Wikipedia. Their conceptual framework is not based on the division between intrinsic and extrinsic motivations, but rather on the internal self and external self-concept. The internal self-concept states that “individuals will engage in certain behavior that his/her capability can perform well” (Yang & Lai 2011, p. 134). The external self-concept relates to “a force that prompts individuals to behave in ways that is congruent with a reference group’s expectations” (Yang & Lai 2011, p. 134). Internal and external self-concepts thus both relate to extrinsic motivations. However, only internal self-concept had a significant impact on knowledge sharing (Yang & Lai 2011).

As online sharing is becoming increasingly popular, a new trend has emerged, namely that of the sharing economy and collaborative consumption. In the following section, this concept is further clarified and elaborated upon.

2.3 The sharing economy and collaborative consumption

Ownership no longer seems to be the ultimate goal of consumers (Bardhi & Eckhardt 2012; Chen 2009). Consumerism is becoming more about accessing goods or services at the times these are needed (Bardhi and Eckhardt 2012). Consumers are realizing that sharing what they own can be a more optimal way of reaching their goals (Botsman and Rogers 2011). This realization paved the way for the sharing economy. The sharing economy is defined as “an economic model based on sharing underutilized assets from spaces to skills to stuff for monetary or non-monetary benefits” (Botsman 2013). Collaborative consumption in turn, is what flows from the sharing economy. It comprehends “people coordinating the acquisition and distribution of a resource for a fee or other compensation” (Belk 2014a). However, the terms collaborative consumption and sharing economy are often used interchangeably (Bardhi & Eckhardt 2012).

There are two types of business models that have their roots in the sharing

(13)

economy; either the company facilitates the goods or resources that are accessed by consumers (e.g. a conventional car rental such as Europcar), or the company facilitates the (online) platform where consumers can connect with each other to offer their own skills, goods or resources (e.g. lodging through Airbnb) (Dervojeda 2013; Gansky 2010; Owyang 2013). The former has existed for many years, whereas the latter is becoming more popular (Dervojeda 2013).

At the early development stage of the consumer-to-consumer (C2C) platforms, most initiatives (e.g. Wikipedia, Couchsurfing) did not involve a monetary exchange (Sacks 2011). However, new C2C platforms seem to operate from a more commercial perspective, facilitating monetary transactions (Sacks 2011). This creates new business opportunities in the sharing economy (Sacks 2011) and consumers seem to want a piece of the pie. Nielsen (2014) found that the majority of online consumers are willing to participate in the sharing economy in exchange for monetary compensation. All in all, the sharing economy is becoming increasingly popular; it was estimated that the C2C market alone was worth $26 billion dollar in 2013 (Botsman 2013b).

2.3.1 Participation in the sharing economy

Since the sharing economy is a relatively new trend, there is no extensive collection of academic research yet that focuses on specific consumer motives to engage in collaborative consumption. To our knowledge, the only quantitative research conducted on attitudes and behavioral intentions towards collaborative consumption is a working paper by Hamari et al. (2013).

Hamari et al. (2013) test four potential motivations that predict behavioral intention for participation in collaborative consumption. Firstly, they test the influence of sustainability. This construct draws on the idea that participation in collaborative consumption is “highly ecologically sustainable” (Hamari et al. 2013, p. 12). Though sustainability has a significant influence on the consumers’ attitude towards collaborative

(14)

consumption, it does not impact behavioral intentions to actually participate.

Secondly, the enjoyment that stems from the sharing activity itself is examined. This enjoyment proved to have significant influence on both the attitudes as well as behavioral intentions towards collaborative consumption.

Thirdly, the influence of reputation enhancement is studied. Hamari et al. (2013) found that this did not significantly influence attitudes or behavioral intentions towards collaborative consumption. This is finding is interesting, considering that other studies found that reputation in fact was a significant motivator to share knowledge online (Wasko & Faraj 2005).

The last factor examined by Hamari et al. (2013) is the motivation to obtain economic benefits through collaborative consumption. They found that the prospect of economic benefits has a significant impact on behavioral intentions towards collaborative consumption, but not on attitudes.

Van der Glind (2013) conducted a qualitative study on three different Dutch C2C platforms about what motivates consumers to participate on these platforms. Five different types of motivations were revealed: curiousness related motives (the pleasure of trying and testing new concepts), practical motives (e.g. saving time), social motives (e.g. the enjoyment of helping someone), environmental motives (e.g. realization that sustainability is important) and financial motives (e.g. avoiding costs).

Campbell Mithun (2012) draws a distinction between rational and emotional benefits to take part in the sharing economy. Rational benefits to participate are financial benefits (“it saves me money”), environmental benefits (“it is good for the environment”), lifestyle benefits (“provides me flexibility” and “it’s practical”), and trial benefits (Campbell Mithun 2012). The top five of emotional benefits derived from participation in the sharing economy are generosity benefits (“I can help myself and others”), community benefits (“makes me feel a valued part of the community”), lifestyle benefits (“makes me feel smart”

(15)

and “makes me feel responsible”) and cultural benefits (“feel part of a larger cultural movement”).

Market research commissioned by Airbnb found similar results; key motivators for consumers to share belongings online are cost savings, motivations related to sustainability, the joy of helping others and gaining an opportunity to meet new people (Ipsos 2013).

In line with some of the previously mentioned motives, Owyang (2014) reports that participating in collaborative consumption is often a choice driven by convenience, cost considerations and product/service quality (e.g. “I couldn’t find it elsewhere”). Furthermore, sustainability motives, environmental concerns, and social benefits are also associations that respondents held towards the sharing economy (Owyang 2014).

John (2013) conceptualizes that participation in collaborative consumption can have various causes. Besides costs and environmental considerations, he also states that sharing is a natural human instinct; “it is said to be an integral part of what it means to be human” (John 2012, p. 121).

Surprisingly, Poynor Lamberton and Rose (2012) did not find that moral or social benefits seem to influence sharing propensity. They find that in line with rational models, emphasizing product benefits and lower costs result in a greater likelihood to take part in a car-sharing program.

Table 1 provides an overview of intrinsic and extrinsic motivations discussed in this literature review.

(16)

Table 1: Overview articles that studied intrinsic and extrinsic motivations in sharing

Authors Intrinsic motivation Extrinsic motivation

S ha rin g f ile s Hennig-Thureau et

al. (2007) Anti-industry utility Collection utility Cost savings Knowledge Plouffe (2008) Community

connectedness Convenience Argan et al. (2013) Enjoyment derived

from activity Altruism Convenience Innovation adoption Sh ari ng k no w le dg e

Kwok and Gao

(2004) Altruism Personal needs Rewards Liking

Affiliation Reputation Wasko and Faraj

(2005) None Reputation Centrality in network Nov (2007) Enjoyment derived

from activity Display values

Exercise KSA’s Ego boost

Career opportunities Yang and Lai (2010) None Display KSA’s

Ego boost

Authors Intrinsic motivation Extrinsic motivation

Pa rt ic ip at io n in th e s ha rin g e co no my Poynor Lamberton

and Rose (2012) None Cost savings Product benefits Campbell Mithun

(2012) Helping others Self-esteem (feel smart and responsible) Part of cultural movement Cost savings Environmental benefits Practicality Trial benefits Van der Glind

(2013) Helping others Curiosity Saving time Sustainability Cost savings Hamari et al.

(2013) Enjoyment derived from activity Economic benefits Ipsos (2013) Helping others

Expand network Cost savings Sustainability John (2013) Natural instinct to

share Cost savings Environmental benefits Owyang (2014) None Convenience

Cost savings Better quality Sustainability Environmental benefits Social benefits 16

(17)

2.4 Ethical consumption

Environmental and sustainability issues are hypothesized to have an impact on the growing popularity of collaborative consumption (Botsman 2010). Consumers are becoming more aware of the consequences of their behavior, which can result in ethical consumption.

Ethical consumption refers to “decision making, purchases, and other consumption experiences that are affected by the consumer's ethical concerns” (Cooper-Martin & Holbrook 1993, p. 113). It offers a tool for consumers to “articulate and promote their ideological interests to society, business, and government” (Bardhi & Eckhardt 2012, p. 885) and provides the opportunity to make a difference themselves. Ethical consumptions can result in boycotts or buycotts (Copeland 2014; Neilson & Paxton 2010) and the consumer power it carries has long been recognized (Kelley 1899). Boycotts are a tool for consumers to penalize companies for undesirable behavior, whereas buycotts function as the exact opposite; consumers decide to purchase products or services to recognize desirable behavior (Copeland 2014; Friedmann 1996; Neilson and Paxton 2010).

Some authors use variations of the term ‘political consumerism’ as a synonym for ethical consumption (Copeland 2014; Bardhi & Eckhardt 2012) but this might be confusing, as ethical consumption comprises much more motives than solely motives related to politics (Koos 2012; Neilson & Paxton 2010). Other examples of concerns potentially triggering ethical consumption are working conditions in development countries, the depletion of natural resources and the choice for organic food (Bray et al. 2011).

2.5 Social capital

Social capital is defined as an elastic construct to describe the benefits one receives from one's relationships with other people (Lin 1999; Neilson & Paxton 1999; Steinfield et al. 2008). Unlike other forms of capital (e.g. physical capital), social capital is less tangible as it evolves around the relationship between people (Coleman 1988). Social capital thus refers to the resources available through a person’s personal network (Adler & Kwon 2002).

(18)

Several authors draw a distinction between social capital at an individual level and social capital at a community level (for an overview see: Lin 1999). In general, social capital at the community level “reflects the affective nature and quality of relationships”, while at the individual level “it facilitates an actor’s actions and reflects their access to network resources” (Wasko & Faraj 2005, p. 39). Nonetheless, these two types of social capital are intertwined and thus usually hard to distinguish (Steinfield 2008). Often gains through individual social capital will be beneficial for the community as well (Lin 1999). According to Putnam (2000), community social capital has been decreasing over the last decades, which becomes visible in a declining number of members joining voluntary organizations (Neilson & Paxton 1999) and leading consumers to undertake more activities by themselves, rather than in groups (Steinfield 2008).

For the purpose of this thesis, the focus of this research is on individual social capital, which can be sub-divided in bonding and bridging capital (Putnam 2000). Bonding capital refers to strong ties (e.g. with family, friends) that provide strong emotional support and/or access to scarce resources (Putnam 2000; Steinfield et al. 2008). Bridging capital refers to weak ties, “loose connections between individuals who might provide useful information or new perspectives for one another, but typically no emotional support” (Ellison et al. 2007, p. 1146). Examples of bridging social capital are retrieving information (e.g. employment connections) and reinforcement and recognition of the self (Granovetter 1973; Lin 1999; Steinfield 2008). Though Adler and Kwon (2002) conceptualize social capital as a form of goodwill, Lin (1999) analyzes social capital from a more economic perspective and states that social capital is an “investment in social relationships with expected return” (p. 29).

2.6 Perceived risk

The concept of risk and its influence on consumer behavior has long been recognized (Bauer 1960 in Mitchell 1999). Objective risk refers to the probability of a loss based on “objective

(19)

facts in the physical world” (Hansson 2010, p. 232). However, consumers are often not able to assess objective risk due to “limited information, a reduced number of trials to consider and a semi-reliable memory” (Mitchell 1999, p. 164). Furthermore, Stone and Winter (1985) question the purpose of objective risk altogether. Therefore, this thesis focuses on perceived risk, which relates to “the subjective expectation of a loss” (Sweeney et al 1999, p. 81). It can be expected that consumer in the sharing economy are affected by perceived risk, possibly because the other party is a fellow consumer rather than a company.

2.7 Gaps in the literature

All in all, currently there is insufficient quantitative evidence on the underlying motivations that contribute to the engagement of consumers in C2C transactions. This thesis aims to enrich current knowledge by conducting an empirical study testing four potential motivators. Furthermore, we study the influence of perceived risk and prior knowledge on the sharing economy.

The managerial implications will provide insights in how existing B2C firms can adjust their marketing strategy to trigger the decisive motivators that consumers would otherwise only experience in C2C transactions.

(20)

3. Conceptual framework

In line with previous research, we expect that both intrinsic and extrinsic motivations influence consumers’ participation in collaborative consumption (Hamari et al. 2013; Wasko & Faraj 2005; Van de Glind 2013). The intrinsic motivators that are studied are the sharing experience and ethical consumption. The extrinsic motivators that are examined are social capital and cost savings. It is expected that these variables will motivate consumers to take part in collaborative consumption thereby increasing consumers’ willingness to pay.

3.1 The sharing experience

Pine and Gilmore (1998) state that the western economy evolved through four stages: agrarian economy, industrial economy, service economy and lastly the experience economy. While going through these economic phases, a progression of economic value arose. By creating an experience for consumers, companies can differentiate (Zomerdijk and Voss 2009) and charge a premium price (Pine and Gilmore 1999). Consumers can thus perceive an experience as valuable.

In addition, Hamari et al. (2013) found that participation in collaborative consumption is partly driven by the enjoyment of the sharing activity itself. Furthermore, Argan et al. (2013) and Nov (2007) showed evidence that the motivation to share knowledge and files online partly stems from the enjoyment experienced during the sharing activity. Lastly, John (2013) states that sharing is a natural instinct, and that it is part of who we are as human.

Thus, in line with what Argan et al. (2013), Hamari et al. (2013) and John (2013) found, in combination with the implications of the experience economy, we expect that enjoyment derived from the sharing activity itself can enhance consumers’ valuation of a C2C transaction in the sharing economy. In other words, we expect that the intrinsic motivation of enjoying the sharing activity can result in greater perceived value for consumers.

(21)

Concluding the above, this leads to the following hypothesis:

H1: Emphasizing ‘sharing’ as part of the experience increases consumers’ willingness to pay.

3.2 Ethical consumption

The concept of being part of a community has regained importance (Botsman & Rogers 2011). Furthermore, Owyang (2012) also states that people are searching for ways to (re)connect with their community and rely on each other, rather than companies. This reconnection with communities potentially implies mindful consumption. Mindful consumption encompasses the idea that consumers choose what and how much to consume without being forced to this because of market conditions (Sheth et al. 2011). In other words, the consumer makes more conscious consumption choices on the basis of what is important to him or her. This can comprehend economic, environmental and social dimensions (Sheth et al. 2011).

Though ethical consumption has gained popularity (The Co-Operative 2012), there might be a discrepancy between consumers’ attitudes and actual behavior. The attitude-behavior gap has long been recognized (LaPiere 1934), and several theories, such as the theory of planned behavior (Ajzen 1991; Ajzen et al. 2004; Armitage and Christian 2003), have conceptualized which factors predict actual behavior. However, in the case of ethical consumption, it might be that this discrepancy is even greater than is the case with regular consumption. Current literature does not seem to be concordant on this topic. According to some authors, consumers state to care about ethical concerns, yet when it comes down to it, do not follow up on these concerns with their purchasing behavior (Carrington et al. 2012; Cowe and Williams in Nicholls Lee 2007; Hamari 2013).

On the other hand, research conducted by Vermeir and Verbeke (2005) did find a strong correlation between behavioral intentions and actual ethical consumption behavior.

(22)

It is thus interesting to further explore what effect ethical consumption has on consumers’ value perceptions. It might be that consumers choose to participate in collaborative consumption because they do not want to contribute to conventional large scale B2C organizations, as was the case in the illegal sharing of motion pictures (Hennig-Thurau 2007). Emphasizing the opportunity to ethically consume, for example by helping a fellow consumer, might yield a sense of contribution to the local economy rather than large-scale corporations. As such, ethical consumption could be a motivator for consumers to participate in collaborative consumption. This thus leads to the following hypothesis:

H2: Emphasizing the opportunity to help a fellow consumer increases consumers’ willingness to pay.

3.3 Social capital

Social capital can be analyzed from an individual perspective or a community perspective (Lin 1999). As this thesis aims to test the influence of social capital as an extrinsic motivator, the individual view of social capital will be adopted.

Collaborative consumption often involves transactions between strangers. For this reason the focus of this research lies on (individual) bridging social capital, as this refers to “loose connections between individuals who might provide useful information or new perspectives for one another, but typically no emotional support” (Ellison et al. 2007). Participation in the sharing economy in the light of bridging social capital thus relates to ‘sharing out’ as this encompasses “sharing between relative strangers” (Belk 2014a, p. 1596).

Botsman and Rogers (2011) state that being part of a community is regaining importance, thereby propelling participation in the sharing economy. Collaborative consumption provides the opportunity to gain new contacts thereby expanding one’s

(23)

network, which might result in bridging social capital, such as information. Benkler (2004) and Nov (2007) have also recognized forms of social capital as a motivation for sharing goods. Thus, emphasizing the opportunity for consumers to gain more social capital through a participation in the sharing economy might create more value. This leads to the following hypothesis:

H3: Emphasizing the prospect to gain bridging social capital increases consumers’ willingness to pay.

3.4 Cost savings

This thesis focuses on both financial and non-financial motives that influence consumers to engage in collaborative consumption. Where hypothesis 1, 2 and 3 all focus on a non-financial motivator, the fourth hypothesis focuses on consumer motivation from a cost savings perspective. Hamari et al. (2013) for example, found that economic benefits had a significant impact on consumers’ behavioral intentions towards participation in collaborative consumption.

Besides the working paper of Hamari et al. (2013), other researched has also showed evidence that cost savings play a role in the participation in the sharing economy (Campbell Mithun 2014; Hamari et al. 2013; Van de Glind 2013; Owyang 2014; Poynor Lamberton & Rose 2012).

This thus leads to the following hypothesis:

H4: Emphasizing the opportunity to save costs increases consumers’ willingness to pay.

(24)

3.5 Perceived risk

Perceived risk can function as an important construct to analyze consumer behavior (Mitchell 1999). It can have an importance influence on customer value perception. For example, Sweeney et al. (1999) demonstrate that the greater risk consumers perceive, the lower the perceived value accredited to a product, which subsequently leads to a lower willingness to buy. In an e-retailing environment, similar results were found; perceived risk influences perceived value, which in turn impacts purchase intentions (Chen and Dubinsky 2003). As consumers expect a higher possibility of negative outcomes, perceptions of perceived risk increase (Gürhan-Canli & Batra 2004).

Trust plays an important role in collaborative consumption (Botsman & Rogers 2011). Within the sharing economy, transactions often take place between strangers, implying that trust is not a given. Trust and perceived risk are correlated (Mitchell 1999). This research thus expects that consumers participating in the sharing economy take perceived risk into consideration. More specifically, it is expected that it will influence the relationship between the defined independent variables (sharing experience, social capital, ethical consumption and cost savings) on the willingness to pay. Concluding the above, the following hypothesis is formulated:

H5: Perceived risk negatively moderates the relationship between motivators to engage in collaborative consumption (sharing experience, social capital, ethical consumption, cost savings) and the willingness to pay.

3.6 Prior knowledge

Knowledge that is related to consumer behavior consists of two components; familiarity and expertise (Alba and Hutchinson 1987). Familiarity refers to “the number of product-related experience that have been accumulated by the consumer” and is interpreted in a broad sense; it can range from advertising exposures to product use (Alba and Hutchinson 1987, p.

(25)

411). Expertise is defined as “the ability to perform product-related tasks successfully” (Alba and Hutchinson 1987, p. 411). It is the familiarity component that this research focuses on.

Previous research showed that ‘file-sharing knowledge’ reduces cognitive effort and search costs resulting in a higher number of downloaded movies (Hennig-Thurau et al. 2007). Plouffe (2008) found a similar outcome; the existent knowledge base positively impacts engagement in a P2P system. But also in the offline environment, prior knowledge appears to have a positive influence on engaging in collaborative consumption. Poynor Lamberton and Rose (2012) for example, found that familiarity with car sharing programs has a significant impact on consumers’ sharing propensity (Poynor Lamberton and & 2012). Thus, the following is hypothesized:

H6: Knowledge positively moderates the relationship between motivators to engage in collaborative consumption (sharing experience, social capital, ethical consumption, cost savings) and the willingness to pay.

Figure 1 depicts the stated hypothesis in a conceptual model.

Figure 1: Conceptual model

(26)

4. Methodology

4.1. Research design

This study aims to test the influence of four different independent variables and two moderators on the dependent variable. A between-subjects, experimental design offers a suitable way to investigate these effects (Saunders et al. 2012). A real-life experimental auction was held online to test the hypotheses.

The auction mechanism applied to this experiment is that of a Vickrey auction (Vickrey 1961). In this type of auction, participants bid once through a sealed bid. The winners pay an amount that is equal to the bid of the highest losing bidder (Vickrey 1961).

The benefits of conducting an experiment through a Vickrey auction are evident. Firstly, it provides the opportunity to measure consumers’ willingness to pay in a very precise matter as explained by Fox et al. (1996):

“In this auction it is in your best interest to bid the amount that you are truly willing to pay (…) If you bid more than your true willingness-to-pay you increase your chances of purchasing (…) but you may have to pay a price that is greater than what you are willing to pay. On the other hand, if you bid less than the amount that you are truly willing to pay then you may lose the chance to purchase (…) at a price that you would be willing to pay” Fox et al. 1996 p. 3).

In other words, it is expected that a Vickrey auction elicits true value assigned to the product by consumers (Hayes et al. 1995). Second, by conducting a real Vickrey auction, actual consumer behavior is measured, rather than attitudes or purchasing intentions, implying more predictive power (Lusk et al. 2004). Thus, this experiment bypasses issues related to the attitude-behavior gap (LaPiere 1934) and the behavioral intention-behavior gap (Armitage and Christian 2003). Previous research on participation in the sharing economy indeed found a discrepancy between attitudes and behavior (Hamari et al. 2013).

The auction took place online on the ‘Veylinx platform’. Veylinx is an online auction

(27)

platform set-up for research purposes only (Veylinx, n.d.). It offers the opportunity to conduct experiments online through a Vickrey auction, thereby measuring actual willingness to pay (El-Haji & Kuijken 2014).

4.2 Auctioned product

The products that were auctioned are 5 Airbnb vouchers worth € 100 each, paid for by the University of Amsterdam. Airbnb is a textbook example of a firm exploiting the opportunities that the sharing economy has to offer (Guttentag 2013). It challenges the hotel industry by claiming to be “a trusted community marketplace for people to list, discover, and book unique accommodations around the world — online or from a mobile phone” (Airbnb 2014). In other words, consumer A (referred to as ‘guest’) rents accommodation from consumer B (referred to ‘host’) through an online platform that is provided by Airbnb. Airbnb’s revenue stream flows from a fee that is charged on both the host and the guest side. Guests pay a service charge between 6-12%; this percentage depends on the subtotal of the reservation. Hosts are charged with a 3% service fee (Airbnb I n.d.; Airbnb II n.d.).

Airbnb is growing rapidly (Grémillon 2014) and it is expected that Airbnb will soon outpace the Intercontinental and Hilton group in terms of room capacity (Carr 2013; Cave 2013; Guttentag 2013).

4.3 Procedure

Five different treatments were designed, each treatment corresponding with an advertisement displaying an independent variable or the baseline treatment (see appendix A for the advertisements). In other words, five advertisements corresponding with ‘the sharing experience’, ‘ethical consumption’, ‘social capital’, ‘cost savings’, and ‘no information’ were designed. Furthermore, five questions were formulated to test for the moderator variables ‘prior knowledge’ and ‘perceived risk’.

4848 Veylinx members were invited to take part in the auction that lasted for 12 hours. Participants were randomly assigned to one of the five different treatments and

(28)

asked to complete the five questions after placing their bid on one of the vouchers. After the auction was closed, all participants received an e-mail communicating the winning bid and whether they had won or not.

4.4 Treatments

Table 2 provides an overview for the treatments that correspond with the hypotheses. The visuals of the advertisements are presented in appendix A.

Table 2: overview treatments and advertisements

Treatment Statement in advertisement

Sharing experience Airbnb, because sharing experiences during your travels is so important!

Ethical consumption Airbnb, enables you to support your fellow consumer!

Social capital Airbnb, stay with locals and discover places you would otherwise not have found! Cost savings Airbnb, often cheaper than a hotel room!

4.4.1 The sharing experience

Consumers demand a more experience driven approach in the hospitality industry (Oh et al. 2007; Sternberg 1997). Airbnb in turn emphasizes the ‘experience’ of using their services. An example of this can be found in a blog on the Airbnb website written by Chip Conley (head of global hospitality), who states:

“You inspire me. The stories and generous spirit of hosts like you are the reason so many travellers love staying on Airbnb. You are helping millions of people feel they can “belong anywhere” on our planet. As hosts, you’re turning strangers into friends through opening your homes and your hearts.” (Conley 2014).

In the ‘sharing experience’ treatment, the following statement was added to the advertisement: ‘Airbnb, because sharing experiences during your travels is so important!’

4.4.2 Ethical consumption

Consumers that ethically consume can have various goals (Carrington 2012; Shaw et al. 2005). Airbnb provides consumers with the opportunity to rent accommodation from a fellow consumer, thereby providing the host with an extra income. A motivation to

(29)

participate in this transaction might thus be of a social nature (Sheth et al. 2011). Therefore in the ethical consumption treatment, the following statement was added to the advertisement: ‘Airbnb, enables you to support your fellow consumer’.

4.4.3 Social capital

Airbnb can be categorized as a social network. Social network sites (SNS) are defined as:

“Web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system”(Boyd & Ellison 2008, p.211).

Previous research showed that SNS provide the possibility to create bridging capital among users (Ellison et al. 2007). Airbnb is a SNS that functions online to eventually facilitate an offline encounter thereby offering the possibility to gain access to social capital online and offline. Airbnb users value the possibility of interacting with their hosts, staying in less touristic areas resulting in a more ‘local experience’ (Guttentag 2013). The ‘local experience’ seems to be in line with the assumption that consumers seem to strive to be ‘travellers’ rather than ‘tourists’ thereby underlining the importance of having an authentic experience (MacCannell 1973; Sternberg 1997; Week 2012).

Therefore in the social capital treatment, the following statement was added to the advertisement: ‘Airbnb, stay with locals and discover places you would otherwise not have found!’

4.4.4 Cost savings

Cost savings are stated to have a potential influence on the decision for consumers to participate in the sharing economy (Campbell & Mithun 2014; Hamari et al. 2013; Van de Glind 2013; Owyang 2014; Poynor Lamberton & Rose 2012). Priceonomics (2013) found that renting a private room on Airbnb in the US is almost 50 percent cheaper than renting a hotel room. Other sources confirm that it is indeed cheaper to use Airbnb rather than a conventional hotel room (Geron 2013; Techonomy 2013). Therefore in the cost savings

(30)

treatment, the following statement was added to the advertisement: ‘Airbnb, often cheaper than a hotel room!’

4.4.5 Prior knowledge

To measure prior knowledge, two questions were asked after participants placed a bid (see appendix A). These questions aim to measure the familiarity component of knowledge; it refers to product-related experience (Alba and Hutchinson 1987). Consumers were asked the following questions:

“Are you familiar with Airbnb?” (1) Yes (2) No “Have you ever booked with Airbnb?” (1) Yes (2) No

4.4.6 Perceived risk

Perceived risk was measured through three separate items (see appendix A). The first item is a question adapted from Verhagen et al. (2006) and Leonard (2012):

“How high do you estimate the chance that the other party will not live up to the agreement when you place a booking with Airbnb?” Participants could answer on the following scale: (1) 0-20%, (2) 20-40%, (3) 40-60%, (4) 60-80%, (5) 80-100%.

The second and third items were statements to which participants could express whether they agreed with this statement on a five-point Likert scale ((1) completely agree, (2) agree, (3) neutral, (4) disagree, (5) completely disagree). The first statement was altered from Pavlou and Gefen (2004): “There is a substantial risk when I book with Airbnb”. The second statement was adapted from Li et al. (2010): “Booking an accommodation through Airbnb with a fellow consumer brings about more risk than more conventional alternatives such as a hotel”.

(31)

5. Results

5.1 Sample characteristics

A total of 4848 Veylinx members were invited to the auction of which 838 eventually participated, resulting in a response rate of 17.3%. Two participants placed bids of respectively € 178 and € 180. These cases were deleted from the dataset as the value of the auctioned product was only € 100. Of the participants 52% was male and 48% female. The average age was 41 years (SD = 15.0).

5.2 Descriptives

At first glance, the prospect of obtaining social capital seems to generate the highest willingness to pay (M=1090.42, SD=1880.58). The average lowest willingness to pay is found in the treatment that emphasizes the sharing experience (M=838.05, SD=1613.74). Table 3 depicts the descriptives for the raw data on willingness to pay per treatment. The data displays relatively large standard deviations in relation to the mean, indicating a wide spread of the willingness to pay.

Table 3: Descriptives willingness to pay in eurocents per treatment

Treatment Mean Std. deviation Minimum value Maximum value

1.No information (N=168) 997.43 1773.215 0 8600 2. Sharing experience (N=165) 838.05 1613.74 0 8513 3.Ethical consumption (N=177) 1005.54 1598.58 0 6350 4. Social capital (N=170) 1090.42 1880.58 0 8500 5.Cost saving (N=156) 1076.02 1775.92 0 8000

Test for normality

In order to see whether the various treatments have a significant influence on the willingness to pay, several parametrical statistical methods are applied. In order conduct these tests, the data should be normally distributed (Field 2013). The central theorem states that “when samples are large (above about 30), the sampling distribution will take the shape of a normal distribution regardless the shape of the population from which the sample was

(32)

drawn” (Field 2013, p. 871). Theoretically, the gathered data should thus be normally distributed as each of the treatments contains at least 156 participants. However, since participants were randomly invited to a real auction, it can be expected that a substantial amount of people placed a bid of zero euro due to the fact that they did not desire the product. This might result in a skewed distribution.

In order to assess normality, a Kolmogorov-Smirnov test and Shapiro-Wilk test were run (see table 4). As the p-value proved to be significant (p<0.001), the data for each treatment is significantly different from a normal distribution (Field 2013).

Table 4: Tests for normality for the willingness to pay variable (raw data)

Treatment Kolmogorov-Smirnov

a Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

W ill in gn es s t o p ay No information .287 168 .000 .629 168 .000 Sharing experience .302 165 .000 .588 165 .000 Ethical consumption .265 177 .000 .682 177 .000 Social capital .281 170 .000 .646 170 .000 Cost savings .300 156 .000 .663 156 .000 a. Lilliefors Significance Correction

However, due to the large sample size, it is advisable to analyze normality not just with the Kolmogorov-Smirnov and Shapiro-Wilk test, but to do this “in conjunction with histograms, P-P or Q-Q plots and values of skewness and kurtosis” (Field 2013, p. 188).

Firstly, when looking at the histograms of the four different treatments (see appendix B), asymmetrical curves leaning to the left are visible, thereby indicating that the data is not normally distributed. Secondly, the Q-Q plots for the different treatments (see appendix C) show that for none of the treatments the dots are aligned with the curve, also indicating that the data is approximately not normally distributed (Field 2013). Lastly, skewness and kurtosis values below -1 or above 1 indicate asymmetry of the curve. Additionally, ‘the further the value is away from 0, the more likely it is that the data is not normally distributed’ (Field 2012, p. 185). As shown in table 5, all treatments display a relative high degree of positive skewness and kurtosis. Furthermore, the skewness and

(33)

kurtosis measures are relatively large in comparison to the standard errors, also indicating a non-normal distribution (Field 2013).

Because the willingness to pay is positively skewed and thus not normally distributed, it was transformed by creating a new variable that contained the natural logarithm of the original willingness to pay. Furthermore, due to the high amount zero bids (N= 392) it was decided to only use the top fifty percentile of bids for each treatment. The data transformation has decreased the skewness and kurtosis values. For example, the skewness for the cost savings treatment for example is now -0.37 (SE=0.27) instead of 1.89 (SE=0.19). Table 6 displays the descriptives of the transformed variable.

Though the a Kolmogorov-Smirnov test and Shapiro-Wilk test still indicate that the data is not normally distributed (see table 7), the corresponding histograms (see appendix D) display a more bell shaped normally distributed curve. It was thus decided to continue the data analysis with the transformed variable for willingness to pay.

Table 5: Skewness and kurtosis values auction willingness to pay (raw data)

Treatment No transformation data Skewness Std. error Kurtosis Std. error

No information 2.39 0.19 5.92 0.37 Sharing experience 2.71 0.19 8.00 0.38 Ethical consumption 1.70 0.18 1.80 0.36 Social capital 2.09 0.19 3.90 0.37

Cost savings 1.89 0.19 2.88 0.39

Table 6: Descriptives natural logarithm of the willingness to pay in eurocents per treatment for the top fifty

percentile

Treatment Mean Std. deviation Skewness Kurtosis

No information (N=84) 6.90 1.37 -0.45 (SE=0.26) -0,93 (SE=0.52) Sharing experience (N=83) 6.54 1.64 -0.94 (SE=0.264) 1,11 (SE=0.52) Ethical consumption (N=88) 7.01 1.24 -0.65 (SE=0.26) -0,41 (SE=0.51) Social capital (N=85) 7.05 1.30 -0.47 (SE=0.26) -0,71 (SE=0.52) Cost savings (N=78) 7.12 1.18 -0.37 (SE=0.27) -0,82 (SE=0.54)

(34)

Table 7: Tests for normality transformed data for the willingness to pay variable

Treatment Kolmogorov-Smirnov

a Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

W ill in gn es s t o p ay No information 0.13 84 .002 0.91 84 .000 Sharing experience 0.14 83 .000 0.93 83 .000 Ethical consumption 0.13 88 .001 0.92 88 .000 Social capital 0.13 85 .002 0.93 85 .000 Cost savings 0.12 78 .005 0.94 78 .001 a. Lilliefors Significance Correction

Homogeneity of variances

For further analysis a one-way ANOVA test is applied. However, an assumption of this parametric test is equality of variances (Field 2013). In this case, Levene’s test shows that there is an equality of variances F(4.415)=2.23, p=0.07. This test thus confirms that this assumption for an ANOVA is met.

5.3 Main analyses

In order to test the first four hypotheses, a one-way between subject ANOVA was carried out and proved to be significant, F(4,413)=2.56, p=0.037, ω=0.12. This indicates that there is a significant difference in the mean willingness to pay for at least one of the treatments.

A Bonferroni post hoc analysis indicates that there is no significant difference between the mean of the no information condition (M=6.90, SD=1.37) and any of the manipulated conditions, as was hypothesized. The outcomes of the post hoc test are displayed in table 8. Hypotheses 1, 2, 3 and 4 are thus not confirmed.

(35)

Table 8: Post hoc analysis (Bonferroni)

Treatment 1 Treatment 2

Mean

difference (1-2) Std. Error Sig.

95% Confidence Interval Lower Bound Upper Bound

Sharing experience Ethical consumption -0.55 0.21 .084 -1.13 0.04

Social capital -0.50 0.21 .155 -1.09 0.08

Cost savings -0.57 0.21 .070 -1.17 0.02

No information -0.35 0.21 .964 -0.93 0.24

Ethical consumption Sharing experience 0.55 0.21 .084 -0.04 1.13

Social capital 0.04 0.20 1,00 -0.54 0.62

Cost savings -0.03 0.21 1,00 -0.62 0.56

No information 0.20 0.21 1,00 -0.38 0.78

Social capital Sharing experience 0.50 0.21 .16 -0.08 1.09

Ethical consumption -0.04 0.20 1.00 -0.62 0.54

Cost savings -0.07 0.21 1.00 -0.66 0.53

No information 0.16 0.21 1.00 -0.43 0.74

Cost savings Sharing experience 0.57 0.21 .07 -0.02 1.17

Ethical consumption -0.03 0.21 1.00 -0.56 0.62

Social capital 0.07 0.21 1.00 -0.53 0.66

No information 0.23 0.21 1.00 -0.37 0.82

No information Sharing experience 0.35 0.21 .96 -0.24 0.93

Ethical consumption -0.20 0.21 1.00 -0.77 0.38

Social capital -0.16 0.21 1.00 -0.74 0.43

Cost savings -0.23 0.21 1.00 -0.82 0.37

*. The mean difference is significant at the 0.05 level.

(36)

5.4 Moderator roles of perceived risk and prior knowledge

A first glimpse at the collected data reveals that willingness to pay, perceived risk and prior knowledge correlate with each other. For example, prior knowledge and log willingness display a medium effect (Field 2013), r=0.51, n=381, p < 0.01. Furthermore, willingness to pay and perceived risk appear to correlate with each other albeit only a small negative effect, r=-.18, N=379, p < 0.01. For an overview off the correlations, see table 9.

Table 9:Means, standard deviations and correlations

Variables M SD 1. 2. 3.

1. Log willingness to pay 6.94 1.36 -

2. Perceived risk 1.90 0.70 -0.18** -

3. Prior knowledge 0.95 0.76 0.51** -0.22** -

**. Correlation is significant at the 0.01 level (2-tailed).

5.4.1 Moderating role of perceived risk

To measure the construct of perceived risk three items were implemented in the question section at the end of the auction (see appendix A). Due to the scales of measurement, only two items were liable for a reliability analysis. These items proved to all have substantial reliability and internal consistency, Cronbach’s α=0,71 (Field 2013). Thus, a new variable was computed, based on the mean of the two items that measure perceived risk. This variable was then subdivided in three categories reflecting low perceived risk, medium perceived risk, high perceived risk. Figure 2 displays the log willingness to pay for various perceived risks levels and treatments.

(37)

Figure 2: The relationship between perceived risk and willingness to pay.

The graph displayed in figure 2 shows that for all treatments consumers that perceive low risk are willing to pay more than consumers that perceive high risk.

However, for the ethical consumption and cost savings treatment, we find counterintuitive results. It seems that for these two treatments, consumers who experience medium perceived risk are willing to pay more than consumer who experience low perceived risk. Furthermore, for the no information treatment and sharing experience treatment we find that consumers perceiving high risk are willing to pay more than consumers that experience medium perceived risk.

Another finding that stands out is that for the ethical consumption, social capital and cost savings treatments, differences in the willingness to pay for the low perceived risk group and medium perceived risk group are smaller than is the case for the sharing experience treatment. This might imply that once the ‘sharing’ component is emphasized, consumers become more sensitive to the perception of perceived risk. This increased sensitivity for perceived risk might thus lead to a greater difference in willingness to pay between the low

5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 No

information experienceSharing consumptionEthical Social capital Cost savings

M ean lo g w illin gn es s t o p ay in e uro ce nt s

Low perceived risk Medium perceived risk High perceived risk

(38)

perceived risk and medium perceived risk consumers.

In order to identify whether these differences in the willingness to pay are significant, a moderation analysis was conducted. Four dummy variables were created and the moderator analysis was conducted with a macro (PROCESS) within SPSS. From the data analysis it becomes clear that the differences in willingness to pay caused by perceived risk are not significant for none of the treatments (e.g. ‘sharing an experience’: Beta= 0.04, p=0.855). The outcomes of the moderator analysis are displayed in tables 10-13.

Table 10: Moderation analysis for perceived risk for ‘sharing an experience treatment’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 7.71 [7.29, 8.13] .22 36.39 p < 0.001 Perceived risk -.32 [-0.53, -0.11] .11 9.76 p = -0.003 Sharing an experience -.59 [-1.58, 0.40] .50 -1.18 p = 0.240 Interaction sharing an experience x perceived risk .04 [-0.15, 0.04] .24 .18 p = 0.855

Note. R-square=0.06 and N=379

Table 11: Moderation analysis for perceived risk for ‘ethical consumption treatment’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 7.47 [7.13, 8.00] 0.22 34.41 p < 0.001 Perceived risk -0.32 [-0.53, -0.11] 0.11 -2.95 p = 0.300 Ethical consumption 0.32 [-0.60, 1.23] 0.47 0.68 p = 0.494 Interaction ethical consumption x perceived risk -0.08 [-0.54, 0.39] 0.23 -0.32 p = 0.748

Note. R-square=0.04. and N=379

(39)

Table 12: Moderation analysis for perceived risk for ‘social capital treatment’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 7.62 [7.20, 8.05] 0.22 35.19 p < 0.001 Perceived risk -0.34 [-0.55, -0.13] 0.11 -3.19 p = 0.002 Social capital 0.11 [-0.84, 1.07] 0.49 0.23 p = 0.818 Interaction

social capital x perceived risk

0.01 [-0.46, 0.47]

0.24 0.03 p = 0.977 Note. R-square=0.03 and N=379

Table 13: Moderation analysis for perceived risk for ‘cost savings treatment’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 7.64 [7.22, 8.07] 0.22 35.20 p < 0.001 Perceived risk -0.36 [-0.56, -0.15] 0.11 -3.36 p = 0.001 Cost savings -0.09 [-1.04, 0.86] 0.49 -0.18 p = 0.853 Interaction

cost savings x perceived risk

0.14 [0.36, 0.63]

0.25 0.56 p = 0.579 Note. R-square=0.04 and N=379

Concluding, perceived risk does not function as a moderator in the relationship between the various treatments and willingness to pay. Thus, hypothesis 5 is rejected.

(40)

5.4.2 Moderating role of prior knowledge

To measure knowledge, a new variable was computed combining the outcome of the two questions that were asked after the auction (see appendix A). This new variable is categorical in nature (0=no knowledge, 1=low knowledge, 2=high knowledge).

The data analysis reveals when knowledge increases, the willingness to pay goes up. Thus, as consumers become more knowledgeable, their willingness to pay increases as well. This pattern seems to apply to all treatments, see figure 3.

Figure 3: The relationship between knowledge and willingness to pay.

In order to examine whether the differences in willingness to pay under various levels of prior knowledge were significant, a moderator analysis was conducted. As stated before, four dummy variables were created and the moderator analysis was conducted with a macro (PROCESS) within SPSS. Tables 14-17 provide an overview of the outcomes.

0 1 2 3 4 5 6 7 8 9 No

information experienceSharing consumptionEthical Social capital Cost savings

M ean lo g w illin gn es s t o p ay in e uro ce nt s No knowledge Low knowledge High knowledge 40

(41)

Table 14: Moderation analysis for prior knowledge for ‘sharing an experience’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 6.30 [6.09, 6.51] .10 60.17 p < 0.001 Prior knowledge .83 [0.66, 1.00] .09 9.76 p < 0.001 Sharing an experience -.67 [-1.10, -.23] .22 -3.03 p = 0.003 Interaction

sharing an experience x prior knowledge

.22

[-0,15; 0,04]

0.19 1.17 p = 0.244

Note. R-square=0.28 and N=381

Table 15: Moderation analysis for prior knowledge for ‘ethical consumption’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 6.12 [5.91, 6.32] 0.10 58.62 p < 0.001 Prior knowledge 0.88 [0.72, 1.05] 0.09 10.35 p < 0.001 Ethical consumption 0.16 [-0.31, 0.62] 0.23 0.66 p = 0.510 Interaction

ethical consumption x prior knowledge

0.01 [-0.37, 0.39]

0.19 0.06 p = 0.951

Note. R-square=0.26 and N=381

Table 16: Moderation analysis for prior knowledge for ‘social capital’

Unstandardized Coefficients

B Std. Error T-statistic P-value (constant) 6.12 [5.92, 6.33] 0.10 58.68 p < 0.001 Prior knowledge 0.88 [0.71, 1.05] 0.09 10.36 p < 0.001 Social capital 0.11 [-0.35, 0.57] 0.24 0.47 p = 0.641 Interaction

social capital x prior knowledge

0.02 [-0.36, 0,41]

0.20 0.11 p = 0.914

Note. R-square=0.26 and N=381

Referenties

GERELATEERDE DOCUMENTEN

In deze studie is gekeken naar het verband tussen expliciete en impliciete associaties bij zowel trait anxiety als wiskundeangst.. Expliciete associaties bij trait anxiety werden

In this study the application of the MCS with LHS technique for the probabilistic simulation of the pultrusion process was investigated based on the process induced variations

televisiereclames een positiever effect heeft op kijkers dan recente populaire muziek, omdat er met gedateerde populaire muziek meer nostalgische herinneringen worden opgeroepen

MalekGhaini, “Effect of friction stir welding speed on the microstructure and mechanical properties of a duplex stainless steel,” Materials Science and

“To what extent do health claims influence consumers’ willingness to buy soft drinks and how is this relationship influenced by product familiarity and brand trust?”... Hayes

“To what extent do health claims influence consumers’ willingness to buy soft drinks and how is this relationship influenced by product familiarity and brand trust?”. Based on

The aim of this research is to investigate the role of awe, a discrete positive emotion, on individuals’ levels of message reception and willingness to pay for consumer goods that

Need for Cognitive Closure (Webster &amp; Kruglanski, 1994; Roets &amp; Van Hiel, 2011) 15-items scale; 6-item Likert ranging from strongly disagree to strongly