Master’s thesis Entrepreneurship
Universiteit van Amsterdam / Vrije Universiteit
Supporting crowdfunding projects: the influence of
consumers’ negative experiences and individual
characteristics
Damir Wijdenes VU: 2507528 UVA: 10061819
Supervisor: Bram Kuijken Academic year: 2015-2016 Date: 23-12-2015
The influences of crowdfunding on consumers’
value perception
Master’s Thesis
University of Amsterdam / Vrije Universiteit Amsterdam
Vincent Sebastiaan Cieraad
10108874 / 2566712
Amsterdam, July 1st, 2015
Supervisor: Bram Kuijken Academic Year: 2014-2015 Semester 2, Block 3
Table of content Foreword……….…….. 3 Abstract..………..………. ……… 4 1. Introduction……….. ……… 5 2. Theoretical Framework………... ……… 9 2.1 Crowdfunding………. ……… 9
2.1.1 The concept of crowdfunding………. ……… 9
2.1.2 Crowdfunding categories……… ……… 10
2.1.3 The crowdfunding process……….. ……… 10
2.1.4 Different types of crowdfunding………. 11
2.2 Negative experiences with supporting crowdfunding projects……… 11
2.3 Risk-taking behavior……… 14
2.4 Personality traits and crowdfunding………... ………. 16
2.4.1 Altruism……….. ……… 17 2.4.2 Openness to experience………...……… 19 2.5 Conceptual model………....……… 22 3. Methodology………. ……… 23 3.1 Questionnaire………...……… 23 3.1.1 Pre-test……… ……… 23 3.1.1.1 Control variables………... 23 3.1.1.2 Independent variable……… 24 3.1.1.3 Dependent variables………...……… 24 3.1.2 Final questionnaire………..……… 25 3.1.2.1 Control variables……….……… 25 3.1.2.2 Independent variables……….. 25 3.1.2.3 Dependent variables………. 28 3.2 Procedure………. 29 3.3 Sample characteristics……….. 30 4. Results……….... 31 4.1 Correlations………. 31 4.2 Centralization……….……….. 32 4.3 Normal distribution……….. 32
4.4 Checking the assumptions……….……….. 32
4.5 Main results………. 33
4.5.1 The effect of negative experience with supporting crowdfunding projects… 34 4.5.2 Risk-taking and supporting crowdfunding projects………... 35
4.5.3 Altruistic behavior and supporting crowdfunding projects……..….……….. 36
4.5.4 Openness to experience and supporting crowdfunding projects……..….….. 37
5. Discussion……….. 40
5.1 Main results………. 40
5.2 Contributions to the study of crowdfunding……… 44
5.3 Managerial implications……….. 45
5.4 Limitations and future research………46
6. Conclusion………. 48
References………. 49
Foreword
This thesis is written for the Master’s degree in Entrepreneurship. This degree is a
collaboration between the Vrije Universiteit Amsterdam and the University of Amsterdam. I would like to thank my thesis supervisor Bram Kuijken for his support, effort and guidance during the writing of this research. I would also like to thank my girlfriend, brother and parents for their help and support during my study period. I hope that the reader will enjoy reading my thesis and will learn from its insights.
Abstract
Crowdfunding is a relatively new and evolving way of financing a project and is rapidly growing as an external funding alternative for entrepreneurs. However, crowdfunding projects are inherently risky, as there is a significant chance of failure. Moreover, a lot of crowdfunding projects are not successful. Therefore, it is likely that consumers can have a negative experience with crowdfunding. This research investigates the influence of consumers’ negative experiences with supporting crowdfunding projects and individual characteristics on the willingness to support future reward-based crowdfunding projects. Additionally, the possible moderating effect of these individual characteristics on this relationship was investigated. The results of this research show that previous negative experiences with supporting reward-based crowdfunding projects reduce the willingness to support future crowdfunding projects. Furthermore, altruistic behavior appears to have a negative effect on supporting crowdfunding projects, whereas risk-taking behavior and openness to experience have no effect. Finally, there appears to be no moderating effect of a person’s risk-taking behavior, altruistic behavior and openness to experience on the relationship between negative experiences with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects.
1.
Introduction
On June 2013, Pirate3D Inc. raised nearly 1.5 million dollars through crowdfunding platform Kickstarter – a record amount at that time – promising a 3D printer. A total of 3520 people invested money in the crowdfunding project. However, two years later, 60 percent of the investors were still waiting for their product. Moreover, it seems unlikely that the investors will get a refund or the 3D printer in which they invested (Techcrunch, 2015). This is a recent example of a negative experience that could happen to consumers when they support crowdfunding projects.
Crowdfunding can be a solution for entrepreneurs. To implement entrepreneurial ideas, financial resources are needed. Often entrepreneurs face difficulties in attracting this external finance from external parties (Belleflamme et al., 2010). Therefore, the concept of crowdfunding can be a useful means for entrepreneurs to gain access to the required financial resources. By an open call, mostly through the Internet, a producer can raise money in order to fund the development of a new product. In this way an entrepreneur is not longer dependent on one investor, but he can make an appeal on the contribution of many individuals contributing an amount of money. In this research, projects fit in the reward-based context of the nature of crowdfunding. Reward-based crowdfunding treats consumers as early customers and allows them to receive products produced by funded projects at an earlier date, better price, or some other special benefit (Mollick, 2014). Moreover, the concept of crowdfunding is interesting for other reasons. It is an uncomplicated way of sponsoring or getting a project funded. Furthermore, crowdfunding is a relatively new and evolving way of financing a project and is rapidly growing as an external funding alternative. The main incentives for the initiator to raise capital through crowdfunding are lower cost of capital and access to more information, like marketing research and input from their investors (Agrawal et al., 2013). Although many success stories are known, more often negative stories about crowdfunding become apparent.
Early-stage crowdfunding projects are inherently risky, because there is a significant chance of failure (Agrawal et al., 2013). Mollick (2014) showed that only twenty-five percent
of the crowdfunded projects delivers on time. After a delay of eight months, only seventy-five percent of the products are delivered. In addition, he showed that projects that are successfully funded, tend to do so by relatively small margins and projects that fail to raise funding, tend to fail by large margins. Furthermore, failure rates of crowdfunding projects are not easy to find on the page of Kickstarter Stats, the worlds largest crowdfunding platform. Lots of data from failing projects were missing (Mollick, 2014). Also, it is difficult for consumers to see which of the crowdfunding projects have more potential than others (Belleflamme et al, 2013). Sellers are better informed than consumers about the quality of the product and its specifications. Most sellers have the tendency to exaggerate the quality of their product, when advertising it, in order to ask for higher prices (Connelly et al., 2011). This problem is called ‘adverse selection’, and states that it is difficult for buyers to discriminate among sellers and to perceive which products are of better quality (Boulding & Kirmani, 1993; Connelly et al., 2011; Kirmani & Rao, 2000).
Crowdfunding is a relatively new phenomenon, so the related crowdfunding literature is limited (Belleflamme et al., 2013). Most research in the past has focused on the concept of crowdfunding and the crowdfunding initiators’ perspective (e.g. Mollick, 2014; Lambert & Schienbacher, 2010; Schienbacher & Larralde, 2010; Frydrych et al., 2014; Gerber & Hui, 2013, Li & Duan, 2014). However, little research has been done from the perspective of the consumer. Previous scientific research investigated consumers’ incentives to participate in crowdfunding (Mollick, 2014; Agrawal et al., 2013; Gerber et al., 2012; Gerber & Hui, 2013). Consumers may engage in crowdfunding for access to investment opportunities, early access to new products, community participation and the support of a product, service or idea (Agrawal et al., 2013). In recent years, crowdfunding has appeared negatively in the news, focusing on concerns such as unrealistic consumer expectations, the potential for fraud, inexperienced crowdfunding initiators and the opportunity cost of lost expert advice (Wired, 2011; The New York Times, 2012; The Financial Times, 2012; CNN Money, 2012). One of the aims of this research is to investigate to what extent consumers’ have had negative experiences with supporting reward-based crowdfunding and its effect on their willingness to
support future crowdfunding projects. This relationship has not been investigated yet and therefore contributes to the literature about crowdfunding from the consumer’s perspective. The main question of this research will be: what is the influence of consumers’ negative
experiences and individual characteristics on their willingness to support reward-based crowdfunding?
Previous research from Butler and Peppard (1998) showed that consumers evaluate their experiences in the post-purchase step, the phase after buying a product. Depending on this evaluation, consumers form intentions for future purchases. Several earlier studies on consumers experience showed that consumers tend to attach more value to a negative experience than a positive experience (Baumeister, Bratslavky, Finkenaur, & Vohs, 2001). Also a consumer’s negative experience will have a negative effect on the intention to repurchase a product (Khalifa & Liu, 2007). A consumer’s negative experience may lead to unwanted negative opinions about crowdfunding and will generate a bad reputation of crowdfunding platforms. Reputation consists of the collective expectations, both emotional and rational, which different groups have of an organization (Thörig, 2005). Thörig (2005) states that reputation is not created by the organization itself, but by its consumers. It is therefore likely that consumers determine the reputation of a crowdfunding platform like Kickstarter. A bad reputation of a crowdfunding platform may lead to a lower consumers’ willingness to support future crowdfunding projects. Hence, in this research, it is argued that consumers’ negative experiences with supporting crowdfunding projects has a negative effect on their willingness to support future crowdfunding projects.
However, a negative experience with supporting crowdfunding does not necessarily have a negative effect on supporting future crowdfunding projects, because person’s individual characteristics may influence this relation. Presumably, crowdfunding attracts certain type of people. It is expected that individuals who are more risk-taking, altruistic and open to experience are more attracted to crowdfunding platforms. These people are more likely to take a negative experience with supporting crowdfunding projects for granted, because of the possible moderating effect of these three individual characteristics. Therefore,
this research will investigate if the effect of a negative experience with supporting crowdfunding projects on the willingness to support future crowdfunding projects will vary at different levels of risk-taking, altruism and openness to experience.
Finally, this research investigates the effect of these individuals’ characteristics on the degree of supporting crowdfunding projects. Whereby, crowdfunding support measured in terms of financially backing products on crowdfunding platforms. By investigating this relationship, it can be determined whether a high degree of consumers’ risk taking, altruism and openness to experience is positively correlated with crowdfunding support.
To give a conclusion on the research question, this thesis has the following structure: first the theoretical framework is discussed along with the hypotheses that follow from the literature. Secondly the research method that is used will be described. After that, results of the statistical analysis will be discussed. And finally, the findings of the research will be discussed as well as the theoretical implications of the results. Also the strength and limitations of this research will be discussed.
Chapter 2: Theoretical Framework
In this chapter the concept of crowdfunding, crowdfunding categories, the crowdfunding process and different types of crowdfunding will be discussed. Next, the relationship between crowdfunding and a person’s behavior will be addressed by means of individual characteristics. Not only individual characteristics have an influence on a person’s behavior, also their previous experience plays an important role. Also, the effect of individual characteristics on the degree of crowdfunding support is discussed. Finally, the possible moderating effect of individual characteristics on the relationship between negative experiences with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects will be discussed.
2.1 Crowdfunding
2.1.1 The concept of crowdfunding
New ventures need resources to succeed. Several researches show that one of the most critical resources to succeed is financing (Gompers and Lerner, 2004; Gorman and Sahlman, 1989; Kortum and Lerner, 2000). In recent years, crowdfunding has become a valuable alternative source of funding for entrepreneurs who seek external financing (Belleflamme et al., 2013). The concept of crowdfunding however, is not new. Politicians have been collecting small campaign donations from the general public for generations. This in essence, is the concept of crowdfunding (Bradford, 2012). However, Internet-based crowdfunding is relatively new and the term did not appear before 2006 (Bradford, 2012). The term crowdfunding, first used in a blog post by Micheal Sullivan in 2006, has its roots in charitable donations (Castelluccio, 2012). Since then, it has been defined by Mollick as: the efforts by entrepreneurial individuals and groups – cultural, social, and for-profit – to fund their ventures by drawing on relatively small contributions from a relatively large number of individuals using the internet, without standard financial intermediaries (Mollick, 2014). By an open call, mostly through the Internet, a producer can raise money in order to fund the development of a new product. In
this way an entrepreneur make an appeal on the contribution of many individuals contributing an amount of money (Agrawal et al., 2013).
2.1.2 Crowdfunding categories
There are different crowdfunding categories a consumer needs to be aware of when deciding to support a crowdfunding project. This is of relevance, because the type of crowdfunding category determines the relationship between the crowdfunding project initiator and its consumers. Each crowdfunding category distinguishes what consumers are promised in return for their contributions (Bradford, 2012). It is important to make a clear distinction between the three actors, which are involved in the crowdfunding process. The crowdfunding initiator is the one who invents the product or service and needs external finance in order to develop his or her project. The crowd or people are the consumers who decide to support the project of the initiator. A final important element of crowdfunding is the online crowdfunding platform through which the consumers finance and support the initiator of the crowdfunding project.
2.1.3 The crowdfunding process
The crowdfunding process begins by publishing a request for funding on a crowdfunding website. The request describes the initiators’ proposed product and business plan. In this way, the consumer is aware of the initiators’ intentions. It also describes what, if anything, the returns for their contributions are. This in turn, depends on the type of crowdfunding category. After the request has been published, consumers can browse through initiators’ listings. If consumers discover an interesting project, they can decide to contribute an amount of money. This amount can vary from a dollar to the total amount the initiator is seeking. Research shows that crowdfunding offerings are rather relatively small (Bradford, 2012). The crowdfunding platform, on which the funding request is published, facilitates the exchange of the funds and, if applicable, the payments from the initiator back to the consumer.
2.1.4 Different types of crowdfunding
Bradford (2012) developed four ways of categorizing the different types of crowdfunding: equity, reward, lending and donation-based. The fist category is equity-based crowdfunding, which gives the consumer the right of a share, dividends or voting rights in the company or project that they are supporting (Bradford, 2012). Reward-based crowdfunding treats consumers as early customers and allows them to receive products produced by funded projects at an earlier date, better price, or some other special benefit. These varying levels of rewards correspond to the amount backed by the consumer (Mollick, 2014). This type of crowdfunding is the most common type of crowdfunding option and is used by crowdfunding platforms such as Kickstarter and Indiegogo (Bradford, 2012). As stated earlier, in this research projects fit in the reward-based contexts of the nature of crowdfunding.
Lending-based crowdfunding involves a loan where the project initiators are liable to return the money to the consumers. An advantage for the consumer is that they might receive interest on the given loan. A disadvantage is that the consumer may not receive anything for lending the money to the initiator, in case of a bad performance (Bradford, 2012). Finally, in donation-based crowdfunding, consumers give a donation to a project or company and do not expect anything in return. This type of crowdfunding mostly exists for non-profit institutions like a charity and is the least common type of crowdfunding option (Bradford, 2012).
A lot of crowdfunding projects are not successful. The amount of successfully funded crowdfunding projects on Kickstarter is currently 97,697 while the amount of unsuccessfully funded projects is 169,673 (Kickstarter Stats, 2015). This is an average crowdfunding project failure rate of 61,46%. Therefore, it is likely that people can have a negative experience with supporting crowdfunding projects.
2.2 Negative experiences with supporting crowdfunding projects
According to Chen & Chang (2003), an online shopper’s purchase experience is influenced by three components and their respective factors. The three main components are interactivity, transaction and fulfillment. A negative experience with crowdfunding may arise
when facing problems with the fulfillment component of the crowdfunded products. Order processing, delivery and post-sales services are important factors, which determine the quality of the fulfillment component. The post-sales services include the availability and quality of technical support. The processing and delivery includes delivery options, shipping and handling charges and warranties. It is possible that crowdfunded products may be delivered too late, or not at all (Mollick, 2014). Also, crowdfunding initiators may suddenly decide to cancel the project. These risks might have huge consequences for the fulfillment component of the crowdfunded products and consequently leads to a consumers’ negative experience with crowdfunding.
Previous research from Butler and Peppard (1998) showed that consumers evaluate their experiences in the post-purchase step, the phase after buying a product. Depending on this evaluation, consumers form intentions for future purchases. This post-purchase step in the consumer’s decision-making process is a very important step, because it determines if a relationship between the retailer and the consumers is strong and continued. Satisfactory purchase experiences influences customers’ future purchase intentions (Lu et al., 2012). Similarly, dissatisfactory purchase experiences will also affect customers’ future purchase decisions (Ferguson & Johnston 2011; Sheth 1973). Several studies argued that customers’ dissatisfaction could cause them to leave a seller (Athanassopoulos 2000; Keaveney & Parthasarathy 2001). As stated in the introduction, a consumer’s negative experience with supporting crowdfunding projects may lead to for example a negative reputation of crowdfunded platforms. So after a negative experience, the consumer’s opinions may become unfavorable towards crowdfunding platforms. Consequently, a bad reputation of a crowdfunding platform may lead to lower consumer’s willingness to support future crowdfunding projects (Kamakura & Russel, 1991; Thörig, 2005).
Several researches have investigated online customer retention in different contexts (Khalifa & Liu 2007; Mouakket, 2009; Koufaris, 2002; Li et al., 2006; Bhattacherjee, 2001; Hong et al., 2006). Online repurchase intention is influenced by the initial use and purchase experience (Wen et al., 2011). In online repurchase intention, the customer is not only a
website user, but also a consumer. Khalifa & Liu (2007) showed that a consumer’s positive shopping experience would have a positive effect on the intention to repurchase a product. On the other hand, a negative experience may have a negative effect on the intention to repurchase a product. Repurchase is defined as the re-usage of the online channel to buy from a particular retailer (Khalifa & Liu, 2007). Research showed that consumers tend to attach more value to a negative experience than a positive experience (Baumeister et al., 2001). Moreover, prospect theory argued that the negative effect of a dissatisfying shopping experience was about twice as great as the positive effect of a satisfying shopping experience (Swinyard & Whitlark, 1994).
Presumably, consumers have a preference for shopping in a particular channel. In the case of crowdfunding, online crowdfunding platforms are an important element through which consumer’s finance and support the initiator of the project (Bradford, 2012). Even though there is a lack of empirical evidence, it may be assumed that a satisfying shopping experience in a channel will influence preference for that channel in the future. Therefore, a negative shopping experience in a given channel is expected to have a negative impact on the possibility of shopping in the same channel in the future (Bandyopadhyay, 2008). Consequently, if the consumer is dissatisfied with its shopping experience through a crowdfunding platform, the likelihood of supporting the same crowdfunding platform again will decrease. It may even be possible that a dissatisfying shopping experience with crowdfunding also has a negative effect of supporting other crowdfunding platforms in the future. It is interestingly to investigate this relationship, because this relationship has not yet been investigated for crowdfunding platforms. In order to do so, the following hypothesis has been formulated:
H1: Consumers’ negative experiences with supporting crowdfunding projects has a negative effect on their willingness to support future crowdfunding projects
2.3 Risk-taking behavior
Many scholars in psychology or the social sciences study the behavior of individuals. Taylor (1974) argued that since the outcome of a choice in consumer behavior could only be known in the future, the consumer is forced to deal with risk and uncertainty. Any choice situation always involves uncertainty about the outcome of the choice and uncertainty about its consequences (Taylor, 1974). According to Kahneman and Tversky, people are different in the way they resolve decisions involving risk and uncertainty. These differences are often described as differences in a person’s risk attitude (Kahneman & Tversky, 1979; Tversky & Kahneman, 1992).
Risk attitude represents an individual’s positive or negative evaluation of controlled behavior with a perceived uncertain outcome (Lee et al., 2015). This risk attitude determines if an individual is more risk-taking or risk-averse. Risk-taking primarily arises from differences in a person’s risk perception and the evaluation of perceived risk in different domains. Perceived risk is being defined as people’s willingness to trade-off units of perceived risk, for units of perceived return (Weber, 1998). A person can be different in their level of risk-taking behavior across domains and situations. For example, individuals have different risk-attitudes when making decisions involving financial or recreational risk. Recreational risk occurs when a person for instance goes skydiving and financial risk occurs when a person faces risk, which involves personal money.
Weber (1998) considers risk attitude to be a person’s personality trait. A person’s personality trait determines the evaluation of perceived risk and determines whether the perceived risk is either desirable or undesirable. Previous research argued that risk-taking is a compound personality trait reflecting a specific combination of scores on all five main personality dimensions of the Big Five (Nicholson et al., 2005), whereas some personality researchers argue that it forms a separate sixth dimension of personality (Jackson, 1994; Paunonen & Jackson, 1996). Due to this mixed findings, it is relevant to examine risk-taking as a separate construct or personality trait. Personality traits were initially defined as stable personality characteristics (Allport & Allport, 1921) that were the result of biological
differences or early childhood experiences (Eysenck & Eysenck, 1985). However, empirical research has shown low correlation between trait-related behaviors in different situations (Mischel & Shoda, 1995). This means that a person’s behavior as a consequence of its personality trait varies among different situations. This research will investigate what the effect of a person’s risk-taking behavior has on the domain of crowdfunding. As stated, crowdfunding lies within the financial investment domain (Blais & Weber, 2006).
Risk-taking individuals have a behavioral intention with respect to participation in risky activities (Lee et al., 2015). In investment, risk-taking people prefer decisions that have
a chance for high return, yet risky (Masters, 1989). Because early-stage crowdfunding
projects are inherently risky (Agrawal et al., 2013) it is assumed that individuals, who are risk-taking, will support more products on crowdfunding platforms. Moreover, if this were the case, it would be beneficial for crowdfunding platforms to focus more on attracting risk-taking consumers. Furthermore, it is expected that risk-risk-taking behavior is positively correlated with the number of supported crowdfunding projects. Therefore, the following hypothesis has been formulated:
H2a: Consumers’ risk-taking behavior has a positive effect on supporting crowdfunding projects
Earlier in this chapter, it was argued that a negative experience with supporting crowdfunding projects has a negative effect on the willingness to support future crowdfunding projects. However, a negative experience with supporting crowdfunding projects does not necessarily have to lead to a negative willingness to support future crowdfunding projects, because this relationship may be moderated by a person’s risk-taking behavior. Early-stage crowdfunding projects are inherently risky because there is a significant chance of failure. Sources of potential failure include fraud and initiators’ incompetence (Agrawal et al., 2013). Like any investment, crowdfunding exposes consumers, whom invest in crowdfunding projects to risks of financial and psychological losses (Li et al., 2015). Presumably, a person who is risk-averse is more likely to stop supporting crowdfunding
projects after a negative previous experience than a seeking person. Assuming that a risk-averse person has less behavioral intention with respect to participation in risky activities than a risk-taking person (Lee et al., 2015). As crowdfunding faces several risks for consumers, it is beneficial for crowdfunding platforms to attract risk-seeking individuals. It is expected that the moderating effect of risk-taking behavior will weaken the relationship between a negative experience with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects. Therefore, the following hypothesis has been formulated:
H2b: Consumers’ risk-taking behavior has a negatively moderating effect on the relationship
between negative experiences with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects
2.4 Personality traits and crowdfunding
Robbins and Judge (2008) argued that a personality shapes a person’s attitudes, abilities and behavior. In the concept of crowdfunding, it is interesting to look at a person’s behavior and what kind of effect certain personality traits have on this behavior. Barrick & Mount (1991) stated that a person’s personality consists of five relatively independent dimensions, which provide a meaningful and orderly classification scheme for studying individual differences. The Big Five model has gained widespread acceptance among personality psychologists for identifying and labeling personality traits (Costa & McCrae, 1992). This model is a hierarchical organization of personality traits consisting of five dimensions: extraversion, agreeableness, conscientiousness, neuroticism and openness to experience. Each of the personality traits of the Big Five model consists of six sub traits. To test a person’s personality, these can be assessed independently of the trait, which they belong to. The sub traits of agreeableness are: altruism, trust, morality, cooperation, modesty and sympathy (Costa & McCrae, 1992). In this research, a person’s altruism and openness to experience will be addressed, in order to measure whether these two personality traits have a positively moderating effect on the relationship between a person’s negative experience with supporting
crowdfunding projects and their willingness to support future crowdfunding projects. It is chosen to address these two personality traits, because it is expected that altruism and openness to experience fit best in the context of crowdfunding and have the largest positively moderating effect.
2.4.1 Altruism
Emotion plays a very important role in pro-social behavior and provides a motivating force that explains when and for whom human beings care and help (Batson, 1990; Frank, 1988; Loewenstein & Small, 2007; Slovic, 2007; Smith, 1976; Darwin, 1965; Hume, 1777/1960; Davis, 1994; McDougall, 1908; de Waal, 1996). Helping others can get people out of a negative mood (Cialdini et al., 1987), sustain a positive mood (Forest et al., 1979) or even improve their subjective well being (Andreoni, 1990; Aknin, Sandstrom, Dunn, & Norton, 2011; Dunn, Aknin, & Norton, 2008; Harbaugh, 1998; Weinstein & Ryan, 2010; Lyubomirsky, Sheldon, & Schkade, 2005; Diener & Lucas, 2000). Altruism is characterized by a motivation to increase another person’s welfare and is driven by a selfless concern for others (Batson, 1990; Eisenberg & Miller, 1987; Darwin, 1871; Batson & Powell, 2003; MacIntyre, 1967). As stated; altruism is one of the six sub traits of agreeableness (Costa & McCrae, 1992)
However, considerable empirical support demonstrates that good deeds are driven by selfish desires and individuals may instead be motivated by intrapsychic rewards and not the desire to help others. Therefore, a debate in social psychology exists whether people are truly altruistic. This debate focuses on the direction of people’s emotions when helping others: are their feelings directed towards themselves or those whom they are trying to help? Contradictory previous research about this social psychological debate exists. Some researches argue that emotion driven pro-social behavior is conflicting with true altruism. These researches argue that helping when motivated by emotion is selfish, because it is triggered by intrapsychic rewards. (Cialdini et al., 1973; Cialdini et al., 1987). Examples of these intrapsychic rewards are feeling good for having done a good deed (Andreoni, 1990),
monetary rewards (e.g., Lin-Healy & Small, 2013) or a positive reputation (e.g., Yoon, Gürhan-Canli, & Schwarz, 2006).
Still, most research argues that emotion can genuinely be about other people welfare and claims that empathy triggers the altruistic motivation to help others in need (e.g., Batson, 1987; Batson, Fultz, & Schoenrade, 1987; Batson, Duncan, Ackerman, Buckley, & Birch, 1981; Batson, O’Quin, Fultz, Vanderplas, & Isen, 1983; Hoffman, 1976; Krebs, 1975). Altruism has been firmly conceptualized (Ashton & Lee, 2001, 2007; Ashton, Lee, & De Vries, 2014), operationalized (Lee & Ashton, 2006), and investigated (Zettler & Hilbig, 2010; Zettler, Hilbig, & Haubrich, 2011). However, the effect of a negative crowdfunding experience with supporting crowdfunding projects and its effect on the willingness to support future crowdfunding projects have not yet been investigated. A person’s altruism may influence this relationship. It is expected that individuals who are more altruistic, are more attracted to crowdfunding platforms.
Bretschneider et al., (2014) argued that intrinsic rewards, such as altruism, could motivate consumers to invest in a crowdfunding project. So, it can be argued that altruism in the reward-based context of crowdfunding may be a motive for consumers to financially back products on crowdfunding platforms. Therefore, it is expected that altruistic behavior has a positive effect on supporting crowdfunding projects measured in terms of financially backing products on crowdfunding platforms. Accordingly, the following hypothesis has been formulated:
H3a: Consumers’ altruistic behavior has a positive effect on supporting crowdfunding projects
People who are more altruistic are more likely to take a negative experience with supporting crowdfunding projects for granted, because of the moderating effect of this individual characteristic. Presumably, consumers with a high degree of altruism are more likely to help crowdfunding initiators because they feel fulfilled about their act, despite of a negative previous experience. Therefore, this research will investigate if the effect of a
negative experience with supporting crowdfunding projects on the willingness to support future crowdfunding projects will vary at different levels of altruism. It is expected that the moderating effect of altruistic behavior will weaken the relationship between a negative experience with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects. Therefore, in this research, the following hypothesis has been formulated:
H3b: Consumers’ altruistic behavior has a negatively moderating effect on the relationship
between negative experiences with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects
2.4.2 Openness to experience
Openness to experience is one of the five personality traits of the earlier discussed Big Five personality model (Costa & McCrae, 1992). Over the years, several definitions of the openness to experience dimension were defined. Feist (1998) stated that people, who are open to experience, have their own view and think out of the box because they see things from a new dimension. Furthermore, these people are more sensitive to variety seeking and open to change (McCrae, 1987). Barrick and Mount (1991) included the characteristics of curious, original, creative and open-minded. Caligiuri (2000) argued that openness to experience reflects the propensity of an individual for innovation. The latest definition comes from Zhao & Seibert (2006), who argued that openness to experiences characterizes someone who is intellectually curious and tends to seek new experiences and explore novel ideas. He also argued that people who are open to experience could be described as creative, innovative, imaginative, untraditional and reflective. On the contrary, someone who scores low on openness to experience can be characterized as narrow in interests and conventional (Zhao & Seibert, 2006). Based on previous research, it can be argued that people who are open to experience tend to be less conventional and less traditional in their behavior.
However, research argued that consumers are not solely motivated by extrinsic rewards, in the form of products or financial return. Intrinsic rewards, such as fun to invest,
altruism, identification and curiosity can motivate consumers as well to invest in a crowdfunding project (Bretschneider et al., 2014). Zhao & Seibert (2006) also found that entrepreneurial individuals have a strong desire to create something larger then themselves, by exploring novel ideas. Moreover, entrepreneurs obtain high scores on the openness to experience personality dimension (Zhao & Seibert, 2006). Thus, individuals who score high on openness to experience are therefore more likely to support crowdfunding projects.
It could be argued that openness to experience has a positive effect on financially backing products on crowdfunding platforms. Because people who like the novelty of new ideas and products funded by crowdfunding may be more inclined to financially invest in new crowdfunding projects as a result of their curiosity and innovativeness. Hence, openness to experience could be positively correlated with supporting crowdfunding projects. In order to investigate this relationship, the following hypothesis has been formulated:
H4a: Consumers’ openness to experience has a positive effect on supporting crowdfunding projects
As earlier discussed, it is assumed that crowdfunding attracts certain type of people, which are more likely to take a negative experience with supporting crowdfunding projects for granted. This may be partly explained because of the possible moderating effect of the openness to experience characteristic of individuals. Moreover, it is assumed that people who have a negative experience with supporting crowdfunding projects may still be willing to support future crowdfunding projects, because people like the novelty of new ideas and products funded by crowdfunding (Zhao & Seibert, 2006). Furthermore, people who are intellectually curious and tend to seek new experiences and explore novel ideas, which correspond to openness to experience (McCrae, 1987), may probably lead to consumers’ willingness to support future crowdfunding projects. Therefore, it is expected that a previous negative experience with supporting crowdfunding projects does not necessarily has to lead to a negative effect on the willingness to support future crowdfunding projects, because the level of a person’s openness to experience could have a moderating effect. It is thus expected that
openness to experience will weaken the relationship between a negative experience with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects. Accordingly, the following hypothesis has been formulated:
H4b: Consumers’ openness to experience has a negatively moderating effect on the relationship between negative experiences with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects
2.5 Conceptual model
H2a + H3a + H4a +
H2b - H3b - H4b -
H1 -
Figure 1: The effect of consumers’ risk-taking behavior, altruistic behavior and openness to experience on supporting crowdfunding projects measured in terms of financially backing products on crowdfunding platforms. And the negatively moderating effect (the moderating effect will weaken the relationship between a negative experience with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects) of consumers’ risk-taking behavior, altruistic behavior and openness to experience on the relationship between a negative experience with supporting crowdfunding projects and its negative effect on the willingness to support future crowdfunding projects Consumers’ altruistic behavior Willingness to support future crowdfunding projects Negative experiences supporting crowdfunding projects Consumers’ risk-taking behavior Consumers’ openness to experience Supporting crowdfunding projects
Chapter 3: Methodology
3.1 Questionnaire
The data is collected through a questionnaire. The questionnaire consisted of six parts, which measure: supporting crowdfunding projects, negative experiences with supporting crowdfunding projects, willingness to support future crowdfunding projects, risk-taking behavior, altruistic behavior and openness to experience. Thereafter, three questions were asked to measure respondents’ gender, age and education.
3.1.1 Pre-test
The questions, which measure negative experience with supporting crowdfunding projects and willingness to support future crowdfunding projects, are not from an existing questionnaire. Therefore, these new scales were pre-tested in order to test the internal consistency and identify possible questions that do not make sense to participants or may lead to unreliable answers. In the pre-test, 26 respondents participated. All the questions from the pre-test can be found in Appendix A. The control, independent and dependent variables of the pre-test will be further elaborated.
3.1.1.1 Control variables Gender, age and education
Three questions were asked to determine gender, age and highest level of education. Gender is being measured by the following question: ‘What is your gender’ with answer scales 1 = male and 2 = female. The respondent’s age is measured in years by means of the following question: ‘What is your gender?’ with a five point Likert scale: 1= younger than 25 years 2 = 25-35 years; 3 = 36-45 years; 46-55 years and 5 = older than 55 years. The highest level or current education is measured by means of the following question: ‘What is your highest level of education?’ with nine answer possibilities, whereby 1 = none; 2 = VMBO; 3 = HAVO; 4 = VWO; 5 = MBO; 6 = HBO; 7 = WO - bachelor; 8 = WO – master and 9 = different.
3.1.1.2 Independent variable
Negative experiences with supporting crowdfunding projects
The second variable measured consumers’ negative experience with supporting crowdfunding projects. The answers to these items were being scored on a scale from 1 to 5 whereby 1 = strongly disagree; 2 = disagree; 3 = neutral (not disagree nor agree); 4 = agree; 5 = totally agree. The total score was calculated by means of a sum score from all the statements per scale. Therefore, the following items were reverse coded: ‘In general, I have a positive experience concerning the timely delivery of the product/ products which I bought through crowdfunding’, ‘In general, I have a positive experience that the crowdfunding project(s) which I supported actually reached its/ their funding goal’, ‘I have positive experiences with the quality of the product/ products which I bought through crowdfunding’ and ‘In general, I have positive experiences with buying a product/ products through crowdfunding’. The second variable, which measured consumers’ negative experience with supporting crowdfunding projects, had a good internal consistency, with a Cronbach’s Alpha Coefficient of .79. Depending on the nature and purpose of the scale, Nunnaly (1978) recommends a minimum level of .70 in order for the scale to be reliable. Moreover, the scale had a good corrected item-total correlation, which indicates a reasonably high degree to which each item correlates with the total score (Pallant, 2013).
3.1.1.3 Dependent variables Supporting crowdfunding projects
The first variable of this survey measured the amount of supported crowdfunding projects, in terms of financially backing products on crowdfunding platforms: ‘How often have you financially backed products through crowdfunding platforms?’ In the pre-test, the answer was given based on a five-point Likert scale from 1 = no experience; 2 = 1 time; 3 = 2 times; 4 = 3 times and 5 = four times or more. If the respondent answered option 1 = no experience, then the questionnaire stopped and those participants were not used in the analysis.
Willingness to support future crowdfunding projects
The third variable measured consumers’ willingness to support future crowdfunding projects. Participants answered the statements by means of the same five-point Likert scale as the scale for negative crowdfunding experience. The total score was calculated by means of a sum score from all the statements per scale. Therefore, the following two items were reversed coded: ‘I would discourage my friends to buy a product through a crowdfunding platform’ and ‘I have little confidence in crowdfunding platforms’. The third variable, which measured willingness to support future crowdfunding projects, also had a good internal consistency, with a Cronbach Alpha Coefficient of .86. Moreover, the scale had a good corrected item-total correlation, which indicates a reasonably high degree to which each item correlates with the total score (Pallant, 2013).
3.1.2 Final questionnaire
All the questions from the final questionnaire can be found in Appendix A. The control, independent and dependent variables from the final questionnaire will now be further elaborated.
3.1.2.1 Control variables Gender, age and education
The same three questions and method were used as in the pre-test, in order to determine respondents’ gender, age and highest level of education (see Subsection 3.1.1.1)
3.1.2.2 Independent variables
Negative experiences with supporting crowdfunding projects
The second variable measured consumers’ negative experience with supporting crowdfunding projects. An example of a question is: ‘I have experienced that the product/ products which I bought through crowdfunding never arrived’. The answers to these items were being scored on the same scale as the pre-test, moreover the total score was calculated by means of the
same sum score as the pre-test. The questions had a good internal consistency, with a Cronbach Alpha Coefficient of .85. The corrected item-total correlation varied between .50 and .81, which suggests a high degree to which each item correlates with the total score (Pallant, 2013).
Risk-taking behavior
The fourth variable was selected from the Domain-Specific Risk-Taking (DOSPERT) Scale (Weber, Blais and Betz, 2002), to measure consumers’ risk-taking behavior. This scale allows researchers and practitioners to measure risk-taking, risk perception and risk behavior in five different domains: ethical, financial (further decomposed into gambling and investment), health/safety, social, and recreational decisions. In this research, the four statements of the financial investment domain were used, because crowdfunding concerns an investing decision. For each of the statements, the funder has to indicate his or her behavior if he or she were to find himself or herself in that situation. An example of an item, which measures risk-taking behavior, is: ‘Investing 10% of your annual income in a new business venture’. The five-point rating scale varies from 1 = extremely unlikely; 2 = unlikely; 3 = neutral; 4 = likely and 5 = extremely likely. The total score was calculated by means of a sum score from all the statements per scale. The DOSPERT scale has been used and validated in a wide range of settings and populations (Weber et al., 2002; Weber et al., 2006). Several studies showed that the DOSPERT scale is valid in many other cultures such as the Dutch, German and Spanish cultures (Johnson et al., 2004). Moreover in this research, the internal consistency of the items was reasonably high, with a Cronbach Alpha Coefficient of .79 (Nunnaly, 1978). The corrected item-total correlation of both scales varied between .56 and .62, which indicates a high degree to which each item correlates with the total score (Pallant, 2013).
Altruistic behavior
The fifth variable was selected from the Dutch version of the HEXACO Personality Inventory (de Vries et al., 2009; Lee & Ashton, 2004), to measure the degree of altruism.
This questionnaire assesses six major personality dimensions: honesty-humility, emotionality, extraversion, agreeableness, conscientiousness and openness to experience. In this research the third component of the questionnaire is used to measure the degree of altruism. The subscale altruism is measured by means of eight statements. In the HEXACO, each subscale can be used as a stand-alone scale. Participants answered the statements through a five-point Likert scale whereby 1 = strongly disagree; 2 = disagree; 3 = neutral (not disagree nor agree); 4 = agree; 5 = totally agree. The total score was calculated by means of a sum score from all the statements. Therefore the following three items were reverse coded: ‘The idea that only the strongest will survive attracts me’, ‘It wouldn’t bother me to harm someone I didn’t like’ and ‘People see me as a hard-hearted person’. The altruism scale had reasonably high internal consistency reliability with a Cronbach Alpha reported of .74 (Nunnaly, 1978). The corrected item-total correlation of the altruism scale varied between .28 and .66. Low values (less than .30) here indicate that the item is measuring something different from the scale as a whole (Pallant, 2013). Therefore, the following two statements were removed: ‘I am an emotional person’ (.29) and ‘People see me as a hard-hearted person’ (.28). As a result, the corrected item-total correlation improved to a value varying between .31 and .63, which indicates an acceptable degree to which each item correlates with the total score (Pallant, 2013). The Cronbach Alpha Coefficient remained the same (.74).
Openness to experience
The sixth variable was selected from the HEXACO-60, which is a shorter version of the HEXACO Personality Inventory (as discussed earlier) and is used when the available time for measuring personality traits is limited (Ashton & Lee, 2008; Lee & Ashton, 2004, 2006). These ten items measured the degree of a person’s openness to experience. Participants answered the statements by means of a five-point Likert scale where 1 = strongly disagree; 2 = disagree; 3 = neutral (not disagree nor agree); 4 = agree; 5 = totally agree. The total score was calculated by means of a sum score from all the statements. Therefore the following items were reversed coded: ‘I would be quite bored by a visit to an art gallery’, ‘I think that
paying attention to radical ideas is a waste of time’, ‘I have never really enjoyed looking through an encyclopedia, ‘I do not think of myself as the artistic or creative type’ and ‘I find it boring to discuss philosophy. The openness to experience scale had a very bad internal consistency, with a Cronbach Alpha of .06 and was thus unreliable (Nunnaly, 1978). Cronbach’s Alpha coefficients were also calculated for items deleted. In this research, deleting the following items will lead to a Cronbach’s Alpha Coefficient of .66: ‘I am interested in learning about the history and politics of other countries’, ‘I would like to do something crazy’, ‘I like people who have unconventional views’, ‘If I had the opportunity, I would like to attend a classical music concert’ and ‘People have often told me that I have a good imagination. This suggests a slightly low, but acceptable internal consistency and therefore these items were deleted. Deleting any more items will not improve the Cronbach Alpha any further. The corrected item-total correlation of the five-item openness to experience scale varied between .30 and .55, which suggests an acceptable degree to which each item correlates with the total score (Pallant, 2013).
3.1.2.3 Dependent variables Supporting crowdfunding projects
The first variable of this survey measured the amount of supported crowdfunding projects: ‘How often have you financially backed products through crowdfunding platforms?’. For the final questionnaire, it was decided to extend the five-point Likert scale to a seven-point Likert scale, whereby 1 = no experience; 2 = 1 time; 3 = 2 times; 4 = 3 times; 5 = 4 times; 6 = 5 times and 7 = 6 times or more. If the respondent answered option 1 = no experience, then the questionnaire stopped and those participants were not used in the analysis. By extending the answer scale, more specific conclusions can be drawn about the possible relation between the amount of crowdfunding support and the degree of a person’s risk-taking behavior, altruism and openness to experience.
Willingness to support future crowdfunding projects
The third variable measured consumers’ willingness to support future crowdfunding projects. Participants answered the statements by means of the same five-point Likert scale as the negative experience with supporting crowdfunding projects scale. An example of a question from this variable is: ‘I have little confidence in crowdfunding platforms’. The total score was calculated by means of the same sum score as the pre-test. The questions had a good internal consistency, with a Cronbach Alpha Coefficient of .84. The corrected item-total correlation varied between .50 and .81, which also suggests a high degree to which each item correlates with the total score (Pallant, 2013).
Table 1 provides an overview of the reliability measures for the negative experience with supporting crowdfunding projects, willingness to support future crowdfunding projects, risk-taking behavior, altruistic behavior and openness to experience scales.
3.2 Procedure
The data in this study were obtained by an online questionnaire developed by MWM2 software. Between 11 and 26 May it was possible to participate in this research. By clicking the link, the respondents were directly referred to MWM2 in order to fill in the questionnaire. The data was collected in several ways. First, the questionnaire was posted on Dutch crowdfunding forum ‘fok forum’. This is a Dutch website and online community with largely up-to-date content such as news, reviews, columns, polls etcetera. Second, the questionnaire was posted on the author’s Facebook and LinkedIn account and was shared in crowdfunding
Table 1
Reliability measures
Units Cronbach Alpha α Corrected item-total correlation Negative experience with supporting crowfunding projects .85 .50 - .81
Willingness to support future crowdfunding projects .84 .50 - .81 Risk-taking behavior .79 .56 - .62 Altruistic behavior .74 .31 - .63 Openness to experience .66 .30 - .55
groups on Facebook and Linkedin. Additionally, the questionnaire was shared by a few Dutch contacts on Linkedin, whom are specialized in crowdfunding. Among them, two authors of the ‘Nationaal Crowdfunding Onderzoek’ which was published in 2013. Moreover, a link to the questionnaire was directly sent to employees of Dutch crowdfunding platforms e.g. CrowdfundingHub, CrowdAboutNow, Oneplanetcrowd and Collin Crowdfund through Linkedin inmail. Lastly, data was collected through a collaboration with ‘Crowdfundmarkt’, a Dutch crowdfunding website which gives an overview of all crowdfunding projects in the Netherlands. In return for access to the customer database of Crowdfundmarkt, a customer satisfaction survey of five questions was developed in order to collect and measure the customer satisfaction of Crowdfundmarkt (see Appendix).
3.3 Sample characteristics
In this research, 157 respondents participated. Whereof 53 respondents have never financially supported products on crowdfunding platforms (see Appendix E). Therefore, 104 respondents were finally included in this research. Table 2 provides a summary of the sample characteristics.
Table 2
! !
Summary of the sample characteristics !! !!
Sample Frequencies Percentages
! ! !! !! Gender Male 74 71,2 ! ! Female 30 28,8 ! !
Age Under 25 years 20 19,2
! ! 25 - 35 years 27 26 ! ! 36 - 45 years 16 15,4 ! ! 46 - 55 years 30 28,8 ! !
Older than 55 years 11 10,6
! !
Highest educational level HAVO 3 2,9
! ! MBO 6 5,8 ! ! HBO 33 31,7 ! ! WO - bachelor 22 21,2 ! ! WO - master 40 38,5 ! ! !! !!
Note. N = 104. MBO = applied education, HBO = professional education, WO =
university
!
4.
Results
4.1 Correlations
In table 3, the correlation matrix of key variables is showed. An alpha level of .10 was used for all statistical analyses. There were seven significant relationships, which varied from hardly or no correlation to low correlation. However, one relationship was reasonably significant with a correlation of -.65. This is the negative correlation between consumers’ negative experience with supporting crowdfunding projects and the willingness to support future crowdfunding projects (p < 0.01). The lower correlated significant relationships were the following. First, there is a positive correlation between supporting crowdfunding projects and consumers’ willingness to support future crowdfunding (p < 0.05). Second, there is a negative correlation between risk-taking behavior and the willingness to support future crowdfunding projects (p < 0.05). Third, there is a negative correlation between altruistic behavior and supporting crowdfunding projects (p < 0.05). Furthermore, there is a positive correlation between risk-taking behavior and negative experience with supporting crowdfunding projects (p < 0.05) and finally, there is a negative significant correlation between altruistic behavior and negative experience with supporting crowdfunding projects (p < 0.10). Also, a negative correlation exists between age and gender. However, this last outcome is not useful for this research.
Table 3
Correlation Matrix of Key Variables
1 2 3 4 5 6 7 8 9
1. Gender 1
2. Age -.21** 1
3. Highest educational level -.06 -.07 1 4. Supporting crowdfunding projects -.12 .07 .03 1 5. Negative experience with supporting
crowdfunding projects -.01 -.01 .07 -.04 1 6. Willingness to support future crowdfunding
projects .11 .09 -.02 .23** -.65*** 1 7. Risk-taking behavior -.04 .02 .12 .09 .21** -.21** 1 8. Altruistic behavior .11 -.03 -.10 -.25** -.17* .15 -.07 1 9. Openness to experience .04 -.07 -.16 .12 .02 -.15 .16 -.04 1 Note. N = 104 for all variables. *p< 0.10 **p < 0.05 ***p < 0.01 (two-tailed significance)
! ! ! ! !
4.2 Centralization
The independent and moderator variables have to be centralized before doing the moderator analysis. Variables are centralized in order to interpret the values more easily. First, the mean of the independent variables were calculated. Thereafter, the mean was subtracted from the variables. With these new centralized values, the linear regression analysis has been done. Consequently the variables: supporting crowdfunding projects, negative experience with supporting crowdfunding projects, willingness to support future crowdfunding projects, risk-taking behavior, altruism and openness to experience were centralized.
4.3 Normal distribution
The distribution of the variable ‘negative experience with supporting crowdfunding projects’ is skewed to the right. To normalize this distribution in the best possible way, the data were transformed by means of a logarithm (Pallant, 2013). The variable ‘willingness to support future crowdfunding projects’ is skewed to the left. To normalize this distribution in the best possible way, the data required a reflected transformation (Pallant, 2013). This means that each data point must be reflected, and then transformed by taking the square root (Howell, 2007). Carrying out a square root transform will convert the data to a normal distribution. The distribution of scores for the variable ‘risk-taking behavior’ is normal. Finally, the distribution of scores for the variables ‘altruism’ and ‘openness to experience’ are reasonably ‘normal’. Therefore, for these three variables, no transformation has to be done.
4.4 Checking the assumptions
Before multiple regression analysis were carried out, the assumptions for multicollinearity, homoscedasticity, normality and linearity were checked. The independent variables showed at least (some) relationship with the dependent variables and the correlations between each of the independent variables were not too high (.7 <). Multicollinearity was not the case, as the Tolerance (.1 <) and VIF (10 <) assumptions were met. By inspecting the Normal Probability Plot (P-P) of the Regression Standardized Residual and the Scatterplot the assumptions of
normality, linearity and homoscedasticity were checked. In the Normal P-P plot, the points were laying in a (reasonably) straight diagonal line from the bottom left to top right. In the Scatterplot of the standardized residuals, the residuals were (roughly) rectangular distributed, with most of the scores concentrated in the centre. Thus, for all cases, it could be concluded that the assumptions were met according to the theory of Pallant (2013). Moreover, by passing the described assumptions, the robustness of the model was checked. Table 4.1, shows the Tolerance and VIF numbers of the first model, table 4.2 shows these numbers for the second model (see the conceptual model in 2.5).
Table 4.1
Multicollinearity model 1
Moderators Tolerance VIF
Risk-taking behavior 0.97 1.03
Altruistic behavior 1 1.01
Openness to experience 0.97 1.03
Note. Dependent variable = supporting crowdfunding projects
Table 4.2
Multicollinearity model 2
Units Tolerance VIF
Negative experience with supporting crowdfunding
projects* 0.93 1.07
Risk-taking behavior** 0.93 1.07
Altruistic behavior** 0.97 1.03
Openness to experience** 0.97 1.03
Note. Dependent variable = willingness to support future crowdfunding projects, *independent variable,
**moderators
4.5 Main results
In table 5, the descriptive statistics of key variables from this research are showed. As stated earlier, the independent and moderator variables were centralized before doing the moderator analysis. Consequently, the minimum values of the units can be negative. However, by
Table 5
Descriptive statistics of key variables
Units Minimum Maximum M SD
1. Age 1 5 2.86 1.32
2. Gender 1 2 1.29 0.46
3. Highest education level 3 8 6.84 1.18
4. Supporting crowdfunding projects -2.08 2.92 0 1.99 5. Negative experience with supporting crowdfunding projects -6.08 9.92 0 3.69 6. Willingness to support future crowdfunding projects -9 6 0 3.46
7. Risk-taking behavior -6.78 9.23 0 3.58
8. Altruistic behavior -9.05 6.95 0 3.29
9. Openness to experience -7.94 9.06 0 3.46
Note. N = 104 for all variables
Units 4,5,6,7,8,9 are centralized
4.5.1 The effect of a negative experience with supporting crowdfunding projects
Hierarchical multiple regression was used to test the first hypothesis, that consumers’ negative experience with supporting crowdfunding projects has a negative effect on their willingness to support future crowdfunding projects, after controlling for age, gender and education. Age, gender and education were entered at Step 1, explaining 3% of the variance in consumers’ willingness to support future crowdfunding projects. After entry of negative experience with supporting crowdfunding projects at Step 2 the total variance explained by the model as a whole was 45%, F (4,99) = 19,91, p < 0.001. Consumers’ negative experience with supporting crowdfunding projects explained an additional 42% of the variance in the willingness to support future crowdfunding projects, after controlling for age, gender and education, R squared change = .42, F change (1,99) = 74,90, p < 0.001. In the final model, only gender and the independent variable: negative experience with supporting crowdfunding projects were statistically significant, with the negative experience with supporting crowdfunding projects recording a higher beta value (beta = -.65, p < 0.001 than gender (beta = .13, p < 0.10). Hypothesis 1 is thus supported: negative experience with supporting crowdfunding projects does have a significant negative effect on the willingness to support future crowdfunding projects.
4.5.2 Risk-taking and supporting crowdfunding projects
Hierarchical linear regression was performed to test hypothesis 2a, that risk-taking behavior has a positive effect on supporting crowdfunding projects, after controlling for age, gender and education. Age, gender and education were entered at Step 1, explaining 2% of the variance in supporting crowdfunding projects. After entry of risk-taking behavior at Step 2 the total variance explained by the model as a whole was 2%, F (4,99) = .61, p = .66. Risk-taking behavior explained an additional 1% of the variance in supporting crowdfunding projects, after controlling for age, gender and education, R squared change = .01, F change (1,99) = 0,69, p = .41. Risk-taking behavior appears to have no significant effect on supporting crowdfunding projects, thus the results do not support hypothesis 2a.
Hypothesis 2b states that negative experience with supporting crowdfunding projects has a negative effect on the willingness to support future crowdfunding projects. Furthermore, it was expected that that this relationship would be less strong by a higher degree of consumers’ risk-taking behavior. Hierarchical multiple regression was used to test this hypothesis, after controlling for age, gender and education. Age, gender and education were entered at Step 1, explaining 3% of the variance in consumers’ willingness to support future crowdfunding projects. After entry of negative experience with supporting crowdfunding projects at Step 2 the total variance explained by the model as a whole was 45%, F (5,98) = 15,77 p < 0.001. Consumers’ negative experience with supporting crowdfunding projects, risk-taking behavior and the interaction between those two variables explained an additional 0% of the variance in the willingness to support future crowdfunding projects, after controlling for age, gender and education, R squared change = .00, F change (1,98) = .01, p = 0.91. In the final model, only gender and the independent variable: negative experience with supporting crowdfunding projects were statistically significant, with negative experience with supporting crowdfunding projects recording a higher beta value (beta = -.65, p < 0.001 than gender (beta = .13, p < 0.10). This means that the expectation was not confirmed and that there is no significant interaction effect of risk-taking behavior on the relationship between
