Unwanted consumer behavior : the influence of observing others

(1)

Amsterdam Business School

Name: Monique Bastiaansen - 5876109 MSc Business Studies - Marketing Supervisor: Prof. Dr. Willemijn van Dolen Date: December 3, 2014

Unwanted consumer behavior:

the influence of observing others

(2)

Abstract

This research examines whether observing unwanted consumer behavior on YouTube increases the same unwanted behavior exercised by others, and whether this effect is stronger for adolescents who have an adult as service representative as compared to adolescents being served by peers. The data for this study was collected during an experiment at a Dutch high school, in which over 100 adolescents participated. The independent variables used in this research are YouTube clips showing adolescents who call a Child Helpline for fun; that is, prank callers. It is argued that watching prank clips may stimulate other adolescents to prank call. This effect is expected to be stronger when the adolescents talk to adults on the helpline compared to when adolescents talk to peers, because adolescents feel more empathy for peers. The research results shed light on how observing unwanted consumer behavior by others can create unwanted consumer behavior. In addition, this research also investigates whether certain personality types are more likely to prank chat or prank call than others. Finally, the managerial implications of the results will be discussed and suggestions for future research will be offered.

(3)

1 Introduction

Social culture exists as a result of individuals learning by observing others’ behavior. People learn how to behave within a certain group by observing the behavior of the people who are already part of that group (Bikchandani, S. et al., 1998, p. 153). People are, of course, influenced by offline behavior, but with the rise of the Internet there are more ways than ever before to observe others and be influenced by their behavior online as well as offline. In the online environment it is easy to observe other people’s behavior through blogs, recommendation systems and social networks.

A possible way to observe other people’s behavior in the online environment is YouTube. Observing others helps people make their own decisions and form their norms. Cialdini & Goldstein (2004, p. 598) describe a way how people create and maintain social relationships: “implicit in the concept of injunctive norms is the idea that if we engage in behaviors of which others approve, others will approve of us, too”. YouTube is an easy way to enrich knowledge about how other people behave.

Most of the existing research has focused on wanted consumer behavior and how wanted consumer behavior is created. Existing research is focused on positive behavior in life or at the workplace, because positivity leads to a more positive work environment and organizational culture (Luthans & Youssef, 2007, p. 322). Moreover, Chartrand (2005, p. 204) state that awareness of the consumer has an important role within consumer behavior. One of the three types of awareness is the outcome. This is the outcome of certain behavior. The consumer is most aware of the outcome of own exercised behavior.

(5)

Because the outcome is important for consumers within their consumer behavior and the outcome of unwanted consumer behavior is not positive, it is unknown how unwanted consumer is created. Furthermore, the aspect of social influence in unwanted consumer behavior is understudied. Observing other consumers misbehaving can possibly create unwanted consumer behavior.

In order to contribute to this research gap, literature about social influence, consumer behavior and unwanted consumer behavior is discussed. To adjust the scope of this study, this research is focused on the consumer behavior of adolescents.

In addition this study addresses the service relationship between service representative and consumer and if this relationship can influence the likeability of exercising unwanted consumer behavior.

Earlier research has shown that if service organizations focus on the relationship with the consumers, consumers will be more satisfied with the organization (Bolton, 1998, p. 62). Understanding this service relationship is important to create a view on the influence of the service relationship on consumer behavior. Therefore the focus in this study is on the service relationship with a service representative of a helpdesk, the Child Helpline. Adolescents can call or chat with the Child Helpline when there are problems or to have a normal conversation. In the literature about the service relationship there is another aspect at the base of a good relationship with your customer, namely trust. Moorman et al (1992, p.315) conclude that trust is an important determinant of the quality of a relationship. Besides factors as trust and satisfaction earlier research does not present other characteristics of a service relationship that can improve the relationship. In order to contribute to this research gap, this study will explore literature about the service relationship, characteristics of service representatives and the connection of the service relationship with behavior.

(6)

Besides the characteristics of service representatives, it will be explored whether the characteristics of consumers influence consumer behavior, in particular unwanted consumer behavior. Earlier research defined five personality labels (Barrick & Mount, 1991, p.2). In this study, the five labels will be used to investigate whether people with a certain personality, are more likely to exercise unwanted consumer behavior. It is tested which personality is linked with which characteristics, however there is no research about the possible link between personality and unwanted consumer behavior. Therefore the results could identify new research areas.

Problem definition

The general focus of this study is the impact of observing other consumers misbehaving on unwanted consumer behavior. It is unknown how observing unwanted consumer behavior by others can influence misbehavior in general. In addition, this study investigates the difference between engaging in unwanted consumer behavior by adolescents being served by peers and adolescents being served by adults. Finally this study researches a possible relation between personality types and the history of prank chats/calls of adolescents.

Therefore, this study has two main goals. First, to investigate a possible correlation between observing the misbehavior of other consumers and engaging in misbehavior in general. Second, to determine if this effect is stronger for adolescents being served by adults as compared to adolescents being served by peers.

(7)

The main research question is: Does observing unwanted consumer behavior by others create unwanted consumer behavior, and is this effect stronger for adolescents being served by adults than for adolescents being served by peers?

The current research aims to answer the following sub-questions:

- Does observing unwanted consumer behavior result in engaging in unwanted consumer behavior?

- Is this effect stronger for adolescents being served by adults in comparison with adolescents being served by peers?

- Are adolescents being served by adults more likely to prank-chat in comparison with adolescents being served by peers?

- Which characteristics of a service representative reduce the chance of unwanted consumer behavior?

- Are some personality types within the BFI-10 more likely to prank-chat?

Theoretical relevance

The main contribution of this study is to shed light on the creation of unwanted consumer behavior, whereas earlier studies focused on the creation of wanted consumer behavior (Luthans & Youssef, 2007, p. 322). This study aims to provide insight into the impact of social influence on unwanted consumer behavior and whether this effect is stronger for adolescents observing other adolescents prank calling/chatting adults as compared to adolescents observing other adolescents prank calling/chatting peers. By observing the connection between social influence and unwanted consumer behavior, a better understanding of the impact of social influence on consumer behavior can be achieved.

In addition, the impact of a service representative is investigated. This is measured by the difference in having a peer or an adult as service representative and engaging in unwanted consumer behavior. Bolton (1998, p.

(8)

62) showed that a good service relationship will lead to more consumer satisfaction. Moorman et al (1992, p.315) found that trust will lead to a better service relationship. This study explores which of the characteristics of a service representative as a peer or an adult will lead to more trust and satisfaction for consumers and in the end to less unwanted consumer behavior. This study contributes to a better understanding of the influence of the service representative.

The relevance of this study is two-fold, since the study first researches and tests the impact of observing unwanted consumer behavior on exercising unwanted consumer behavior and secondly, tests whether this effect is stronger for adolescents being served by adults than by peers.

Lastly, current research will be one of the first studies to investigate the influence of personality types on unwanted consumer behavior, where earlier research explored if characteristics can be divided in these personality types (Barrick & Mount, 1991, p.2). It can be concluded that current research aims to fill the gap in unwanted consumer behavior research.

Managerial relevance

In today’s world, where the behavior of others is constantly visible to all consumers, it would be interesting to know how such observation can create unwanted consumer behavior. Social media is a broad area where other consumers’ behavior can be observed. Clips, which include unwanted consumer behavior, are uploaded by consumers and not by companies. For companies it is good to know if there are certain clips about their company on the Internet, which behavior is exercised in these clips and how many people watch the clip, to inform their employees. These employees can respond more adequate to new misbehavior when serving a consumer and possibly reduce the amount of unwanted consumer behavior.

The influence of the characteristics of the service representative on engaging in unwanted consumer behavior is investigated in this current

(9)

research. Based on the results, management make decisions about which service representative can prevent unwanted consumer behavior for their company. This research is targeting which characteristics are important to maintain trust and satisfaction in the role of service representative to create a good service relationship with the consumer. These results can possibly be used by companies to select service representatives based on these characteristics.

Thesis overview

This research consists of a literature review and an empirical study. The literature review, summarized in Chapter 2, first describes how observing the behavior of others can influence people. Second, unwanted consumer behavior is explained and its characteristics regarding social influence are analyzed. Third, the difference between adolescents and adults and their influence on copying unwanted consumer behavior is discussed. Finally, the Big Five personality traits will be linked to the possible misbehavior of individuals. The possible explanations for unwanted consumer behavior will lead to hypotheses, presented in the literature section. In the methodology section the research setting and the methodology used for the research are described. Finally, the results will lead to a discussion, future implications for managers and future research possibilities. This research ends with a conclusion.

(10)

2 Literature Review

This chapter presents the literature review for the current study. The first part will be devoted to the social influence of other consumers, by reviewing theories of consumer behavior and social influence. Unwanted consumer behavior and the effect of social influence on unwanted consumer behavior is discussed. In addition, the influence o of the role of service providers is investigated which willlead to hypothesis development regarding the influence of observing unwanted consumer behavior on exercising consumers.

2.1 Responses to Social Influence

2.1.1 Definition of Social Influence

Social influence can be defined as a change in a person’s behavior as a result of the behavior of others (Dholakia, 2004, p. 242). According to Bakshy et al. (2009, p. 325), social influence determines when and to what extent people adopt behaviors.

The definition of social influence used in this research is the change in a person’s behavior when this person notices the behavior of others.

2.1.2 Online and Offline Social Influence

From birth people are able to observe other people and to learn from the behavior that is exhibited by others. Young children learn how to speak by observing their families. Observing others continues throughout life. People grow up and see which behavior is desirable for different occasions. Children learn what wanted consumer behavior is by observing others (Ward, S., 1974, p. 8).

There are multiple ways to observe the consumer behavior of others and to influence it. People can observe the behavior of others offline and online. In

(11)

each setting different observing strategies are used to see what other consumers are doing and what other consumers find important.

In the offline environment, consumers influence other consumers by communicating about their preferences and experiences with a business. Consumers use word-of-mouth communication from other consumers to form retail attitudes and to form purchase behavior intentions (Ortinau, 2013, p. 796).

While in an offline environment the social influence of other consumers on consumer decisions is exerted mainly through direct interaction between consumers, in the online environment the influence is mostly exerted through blogs, reviews and social media. Online reviews have a positive influence on product sales; with the impact of that influence depending on product and consumer characteristics (Zhu & Zhang, 2010, p. 144). In addition, blogs can evoke empathy and influence the reader’s buying intention (Hsiao & Lu, 2013, p. 173).

With the advent of social media, it became easy to observe the behavior of other consumers. On Facebook and Twitter consumers can easily see which products and companies their friends “like” or “follow”, and through YouTube people can post videos of their behavior. Moreover, all information conveyed through social media is easy accessible, in contrast to word-of-mouth communication in the offline environment. In particular, YouTube provides a way to observe the behavior of consumers and to observe the reactions of other consumers to that behavior, through comments and “likes”.

2.1.3 Responses to social influence

Existing academic research on social influence is almost all based on old theories. However, these theories have evolved. Researchers have found new methodologies within the traditional theories to clarify the operating model of

(12)

social influence (Cialdini & Goldstein, 2004, p. 614). Social influence exists, but how do targets of social influence react to social influence? Does the source of social influence make a difference as to how targets react to it?

One of the first operating models of the responses to social influence is the model of Willis (1963). Responses to social influences are the personal responses of a person who comes in contact with social influence. While earlier research focused on a single dimension of response , Willis (1963, pp. 499-500) states that there is an understanding of social influence having a distinction between two dimensions; the conformity response and the non-conformity response, which consist of two variables; namely dependence vs. independence and conformity vs. anti-conformity.

Nail et al. (2000, p. 455) elaborated on these two dimensions: “First the agreement or disagreement between a target and source of influence before the target is called upon to respond to influence (pre exposure), secondly, the agreement or disagreement between a target and source of influence after the target responds to influence (post exposure)”. Observing behavior in YouTube clips is also a way of social influence and these two dimensions are involved in the responses of the viewers. Current research is focused on the post exposure dimension of response. The question that should be asked is within this research: Is there agreement or disagreement between the viewer and the source of influence?

Nail et al. (2000, p.454) also adjust two widely recognized types of conformity to their model: conversion and compliance. By using all four types of the two dimensions of possible responses to social influence Nail et al. (2000, p.455) build a four-dimensional model of social response. This framework, shown in Figure 1, provides a complete model that includes and labels the various types of responses to social influence.

(13)

Nail et al. (2000, p. 455) explain the various types of responses to social influence thus: “Conformity is the implicit positive movement, by pre exposure disagreement between target and source, that is followed by post exposure agreement. Independence is defined by the absence of movement, pre disagreement/post disagreement. Anticonformity by negative movement, pre agreement/post disagreement. Congruence or uniformity is defined by the absence of movement, pre agreement/post agreement”.

Figure 1. A model derived from Willis’s (1963) scheme for symbolizing possible responses to social influences by Nail et al. (2000).

MacDonald et al. (2010, p. 89) developed the Social Response Context Model (SRCM). This is the first model in the social response literature to suggest that private beliefs that may exist, before social influence occurs, need to be considered. The main part of this model focuses on four questions, where the answers lead to 16 different possible sets of responses:

1. What is a person’s public position relative to an eventual influence source before exposure to influence (pre exposure, public

agreement/disagreement)?

2. What is that person’s private position before influence (pre exposure, private agreement/disagreement)?

(14)

3. What is that person’s public response relative to an influence source after exposure to influence (post exposure, public agreement/disagreement)? 4. What is that person’s private response after influence (post exposure,

private agreement/disagreement)?

These four questions provide insight into how customers think and feel in the pre exposure and post exposure phases.

From this model can derived that besides the post exposure phase, which is the focus of this research, also the pre exposure phase is an important part of how people respond to social influence. If the viewer of the YouTube clip will engage in the same observing behavior, MacDonald et al. (2010, p.89) suggest that this social response can also be made, before exposure of the YouTube clip. That is why current research also focuses on the personality types, and the influence of these personality types on engaging in unwanted consumer behavior after observing that behavior.

2.2 Unwanted consumer behavior

Most studies concerning consumer behavior are conducted to learn more about ways to create generally accepted behavior and how to speed up the creation of such wanted consumer behavior. For example, nowadays more people want to live healthily and feel fit. Research has shown that losing weight in groups is more effective than losing weight on your own, because others can see your improvement (Allan et al., 2010, p. 29). Companies are willing to contribute to this wanted consumer behavior and to the generally accepted norms. Fashion retail companies (for example, Nike), are investing their money in creating running applications for mobile phones to attract more consumers. With such a running application it is possible to share accomplished runs with friends, creating a better chance that people will feel as if they are part of a group. These two examples are ways how social influence and consumer behavior are connected with each other and show that social influence has a

(15)

great impact on the way people behave. Current research investigated if this effect of social influence will also have an impact on how people react to unwanted consumer behavior and if people are more likely to perform unwanted consumer behavior when people see a YouTube clip with the same behavior.

Although everyone knows which behavior is acceptable as a consumer, not everyone lives up to these standards. It is difficult to label unwanted consumer behavior, because consumers and organizations may view correct behavior from different perspectives (Fullerton & Punj, 1997, p. 338). Unwanted consumer behavior can do harm to both companies and consumers. Overall, it can be concluded that unwanted consumer behavior infringes on the generally accepted norms of consumer behavior (Fullerton & Punj, 1993, p. 570). Consumers who misbehave are just as diverse a group as consumers in general (Fullerton & Punj, 2004, p.1241). Therefore, it is still unknown which different characteristics consumers who misbehave have in comparison to consumers in general.

Misbehavior is seen in both the offline and the online environment. Online unwanted consumer behavior is a new way of misbehaving (Harris & Dumas, 2009, p. 381). Misbehaving is easier for consumers online than offline, because online there isa lower risk of getting caught.

Unwanted consumer behavior can be created by social influence, the same as wanted behavior can be created by social influence. On YouTube there are many videos of consumers engaging in unwanted consumer behavior. Observing that unwanted behavior can influence others to engage in unwanted consumer behavior in the same way how people are influenced when observing wanted consumer behavior.

(16)

2.2.1 Reasons for misbehavior

There are different reasons why people misbehave. Thrill seeking is one of the most common motivations for misbehaving (Fullerton & Punj, 2004, p.1244). There is a significant relationship between sensation seeking and risky driving (Jonah, 1997, p. 659). Another reason for consumer misbehavior is that consumers are more willing to victimize large companies than small companies, because of the impersonal nature of a large company (Fullerton & Punj, 2004, p. 1245).

Another reason for misbehaving is copying. For example, young people have different motivations for starting to drink alcohol. They may have, for example, an enhancement motive (people feel better when they are drinking) or a copying motive (copying the behavior of peers). Kuntsche et al. (2006, p. 1847) found that the percentage of adolescents reporting a copying motive for drinking alcohol decreased from 1984-1990, while the percentage of adolescents reporting an enhancement motive increased. Moreover, males scored higher on the copying motive. In today’s world with social media it is easier to copy behavior, because there is more behavior to watch than in the offline environment alone. Watching other people behave can lead to copying the same behavior. Adolescents are aware of YouTube and the different kind of clips collected on YouTube. Consumers can easily find clips where they can see how peers behave. When taking this in account, the copying motive can be an important motive for adolescents to engage in the same unwanted consumer behavior.

One of the clearest implications of our desire to affiliate with others is that the more we like and approve of them, the more likely we are to take actions to cultivate close relationships with them (Cialdini & Goldstein, 2004, p.598). This is one of the underlying arguments of the copying motive. Adolescents are aware that others would like and approve of them when copying the same behavior.

(17)

But, why do people misbehave if people know such behavior is not the generally accepted norm? A possible explanation lies with the theory of planned behavior. Azjen (1991, p. 182) stated that behavior follows from an intention that is formed by three determinants: attitude towards a behavior, a subjective norm, and perceived behavior control.

Within the theory of planned behavior, perceived behavioral control is positioned as the third determinant of intention; thus, the easier a behavior is, the more likely one will intend to engage in it (Armitage & Christian, 2003, p. 191). Control beliefs underpin behavioral control. Control beliefs are the perceived frequency of facilitating or inhibiting factors multiplied by the power of those factors to inhibit/facilitate the behavior in question (Armitage & Christian, 2003, p. 191). Behavioral intentions, which lead to an attitude towards a certain type of behavior, are a summary of motivation required to prompt engaging in a particular behavior. This reflects an individual’s decision-making process for exercising a type of behavior (Armitage & Christian, 2003, p. 190). The determinant of the theory of planned behavior that will lead to possible misbehavior is perceived behavioral control. If unwanted consumer behavior is easy to exercise, it is more likely an individual will intend to perform unwanted consumer behavior. This could be the underlying determinant for copying unwanted behavior.

This research investigates how observing unwanted consumer behavior can lead to unwanted consumer behavior within the context of the role of the service representative. Calling is an easy way to be anonymous and this behavior is easy to exercise because the chance that the adolescent is familiar with the service representative is low. In this case the perceived behavioral control is low because it is easy to exercises, an adolescent is more likely to engage in unwanted consumer behavior.

(18)

Therefore, in this research one of the main goals is to investigate if consumers copy misbehavior when consumer see other people misbehave, leading to the first hypothesis:

Hypothesis 1. Observing unwanted consumer behavior by others positively influences unwanted consumer behavior.

2.3 Role of service representatives

2.3.1 Service representatives

A commercial relationship between a service provider and a consumer is an important type of marketing relationship (Price & Arnould, 1999, p. 38). A commercial friendship involves private aspects, for example affection, loyalty, intimacy and support (Price & Arnould, 1999, p. 50).

The critical success factor in service relationships is trust. The consumer must perceive the service representative to be honest, someone who shows integrity and has high ethical standards (Coulter & Coulter, 2002, p. 37).

Therefore, the role of a service representative cannot be underestimated. If a service representative and a client feel connected and the client trusts the representative, ta friendship is possible. Affection, loyalty, intimacy, support and trust can lead to a strong relationship.

2.3.2. Service representatives as peers

Chein et al. (2011, p. 1) found a difference in behavior when adolescents in presence of a peer in comparison with an adult.

It is possible that observing the unwanted consumer behavior of others could positively influence unwanted consumer behavior in general. In this research the focus is on for which type of service representative, adolescent or adult, this effect is stronger. There could be multiple reasons why this effect

(19)

could be weaker for adolescents being served by peers as compared to adolescents being served by adults in a service organization.

The effect could possibliy be weaker among adolescents served by peers because children have more empathy and sympathy for peers than for adults and therefore do not want to behave in an unwanted manner in an interaction with a peer (Miller et al., 1996, p.211). ‘Empathy encompasses both affective and cognitive components, and is broadly defined as the experience and the understanding of other’s emotions’ (Kokkinos & Kipritsi, 2012, p. 44).

Another possible reason for the difference in behavior could be that adolescents are more loyal to peers than towards adults because of peer pressure. Peer pressure is the pressure to conform to the norms of the peer group, which creates commitment and loyalty (Clasen & Brown, 1985, p. 452). Furthermore Gardner & Steinberg (2005, p.629) found that adolescents take substantially more risk when surrounded by peers. Loyalty and empathy for peers can possibly lead to a difference in the amount of unwanted behavior when dealing with service representatives who are peers, rather than adults.

There is not a lot of research about the difference in behavior with an adolescent or a peer as service representative. However, similar research showed that there is a difference in behavior when adolescents are served by peers as compared to when adolescents are served by adults. Because adolescents are more loyal to, and have more empathy and sympathy for, peers, this research will investigate whether children who are served by peers are less likely to perform unwanted consumer behavior in comparison with children who are served by adults.

A stronger relationship with a client ensures that the client feels connected and is less willing to misbehave. When peers instead of adults serve adolescents, there is a greater possibility that the adolescent will have a stronger

(20)

relationship with the peer. Peers have more empathy and sympathy for each other, and because the interaction is between peers it is easier to trust each other, leading to a stronger service representative – client relationship. This leads to the following hypotheses:

Hypothesis 2. The effect of Hypothesis 1 is stronger for adolescents being served by adults as compared to adolescents being served by peers.

Hypothesis 3. Adolescents being served by adults, as compared to adolescents being served by peers, are more likely to engage in unwanted consumer behavior

2.4 Characteristics of service representatives

Besides the possible difference of social influence between a peer or adult as service representative, this research addresses the possible reasons for this difference in effect. In this research the focus is on the difference in the characteristics of the employees who serve the adolescents. Hypothesis 3 proposes that adolescents being served by adults are more likely to exercise unwanted consumer behavior than adolescents being served by adolescents. Lynch & Chiccetti (1997, p. 81) found that “middle school children report more positive perceptions of relationships with peers in comparison with adults”. Furthermore Bolton (1998, p. 62) showed that a good service relationship will lead to more satisfaction with the customers. Moorman et al (1992, p.315) found that trust will lead to a better service relationship. In this research is investigated whether the service relationship influences engaging in unwanted consumer behavior and whether peers are more likely to have a better service relationship with the consumer than adults.

(21)

There are possible explanations why adolescents have more belief in the genuine interest of adults. A child friendship is characterized by a lack of caring for other people (Parker et al., 2006, p. 453). Adolescents will feel that adults care, which can be a reason for adolescents to believe adults are genuine interested. Furthermore, adolescents feel safer with adults, because adolescents feel more protected surrounded by adults.

There are also multiple reasons for adolescents to have a stronger belief in the genuine interest of peers as compared to adults. For example, a possible friendship is an excellent way to measure if there is a connection with another person (Parker et al., 2006 p. 420). When you are a child there is a better chance that you will be friends with someone close to your own age and that you will feel more connected to a peer than to an adult. Also, when adolescents feel a connection with a person, they believe that they can rely on that person (Parker et al., 2006, p. 447).

Friends also enjoy the activities engaged in together and the conversations among friends. Shared experiences are an excellent way to increase empathy with one another (Parker et al., 2006, p. 441). Because peers are more likely to share the same experience, there is a greater chance that adolescents will empathize more with peers than with adults. Furthermore, when friends have a shared history together friends feel comfortable with each other (Parker et al., 2006, p. 447).

Earlier research concluded that peers always give each other honest and critical feedback (Parker et al., 2006, p. 435). This possibly explains why adolescents feel a peer’s honesty and this could be a reason why adolescents trust peers more than adults and feel safer among peers.

The literature review suggests that peers have a stronger belief in the genuine interest of peers in comparison to adults because adolescents feel that peers have a better understanding of the life the adolescents are living. This leads to the following hypothesis:

(22)

Hypothesis 4. Unwanted consumer behavior is less likely when services representatives are genuine interested in adolescents.

2.5 Personality Index

The Social Response Context Model (SRCM), which was discussed earlier in this research (MacDonald et al., 2010, p. 89) suggests that private beliefs that may exist before social influence occurs need to be considered to evaluate the response to social influence. To gain a complete impression of the influence of observing unwanted consumer behavior on engaging in unwanted consumer behavior, this research will also investigate the pre exposure beliefs of adolescents. Rallapalli et al. (1994, p. 494) found a significant relationship with personality and ethical beliefs. In this research the pre exposure beliefs will be investigated using a personality index.

2.5.1 Big five personalities

In 1963 Norman was the first to define the Big Five personality (BFI-44) traits (Barrick & Mount, 1991, p. 2); namely, extraversion, emotional stability, agreeableness, conscientiousness, and culture. Over the years researchers tried to define these big five personalities. The labels change over time, but overall the labels still have the same meaning.

Extraversion typically consists of sociability (Judge et al., 1999, p.624). People who are extraverted tend to be outgoing, but are also surgent and active. Extraverted people are more likely to be leaders because they are ambitious and assertive. People who score high on neuroticism are more likely to experience negative moods and exhibit a lack of emotional stability (Judge et al., 1999, p. 624). Conscientious people have a great belief in self control, achievement, and order, and are more persistent than less conscientious people

(23)

(Judge et al., 1999, p. 624). Openness refers to openness to experience; these people are intellectual and have an active imagination (Judge et al., 1999, p. 625). Individuals who are agreeable are cooperative and liked by other people, because agreeable people are gentle (Judge et al., 1999, p. 625).

Combining this big five personality types with the main reasons for misbehavior, namely the copying motive and sensation-seeking, there are two personality types who would be more likely to engage in unwanted consumer behavior. First, people with an extravert personality, because extraverted people are excitement seekers (Rammstedt & John, 2007, p. 208). Furthermore extraverted people are described as more active and impulsive (Judge et al., 1999, p. 624) Second, people with an openness personality, because they also have a positive relationship with excitement seeking, but they are also people open to experiences, to new ideas and actions (Rammstedt & John, 2007, p. 208).

People with these two of the five personalities are more likely to sensation seeking and as a result will be more likely to have an history of engaging in unwanted consumer behavior. This leads to the following hypothesis:

Hypothesis 5. Extravert people and people open to experience are more likely to have a history of engaging in unwanted consumer behavior.

(24)

The following conceptual model is derived from the hypotheses. The proposed hypotheses will be tested with the goal to retrieve answer to the research question.

Figure 3. Conceptual Framework

Hypothesis 1 Observing unwanted consumer behavior by others positively influences unwanted consumer

behavior.

Hypothesis 2 The effect of Hypothesis 1 is stronger for adolescents being served by adults as compared to

adolescents being served by peers.

Hypothesis 3 Adolescents being served by adults, as compared to adolescents being served by peers, are more

likely to engage in unwanted consumer behavior.

Hypothesis 4 Unwanted consumer behavior is less likely when services representatives are genuine interested in

adolescents.

Hypothesis 5 Extravert people and people open to experience are more likely to have a history of engaging in

unwanted consumer behavior.

Observing unwanted consumer behavior Characteristics of the service representative Service by an adult instead of a peer Unwanted consumer behavior History of unwanted consumer behavior Extravert people and

people open to experience H2: + H1: + H3: + H4: + H5: +

(25)

3 Methodology

This chapter presents the research method, the description of the sample, and the data collection procedure. In addition, the dependent and independent variables are described and the scales used to measure the variables are discussed.

3.1. Data collection procedure

The dataset used in this research was obtained through an experiment conducted on the 9th of December, 2013. A high school in Breda, het Onze Lieve Vrouwelyceum, was visited and four first-year classes participated in the experiment.

The experiment is constructed around the service of the Kindertelefoon (Child Helpline). The Child Helpline is a service organization that offers help to children with problems. Over the last 27 years more than 200,000 telephone conversations are conducted each year (Kindertelefoon, 2006, p. 13). The Child Helpline was founded in 2003, coinciding with the increased use of the internet by children. Since the start of the Child Helpline, there has been an upward trend in the number of serious conversations conducted. For example, in 2012 there were 67,538 serious chat conversations conducted (Kindertelefoon, 2012, p. 8). The Child Helpline offers the possibility to chat with peers or adults. The idea behind the peer-to-peer chat is that children are more comfortable talking to peers because peers are better able to empathize with a child’s problems. To analyze the difference in unwanted consumer behavior between adolescents served by peers and adolescents served by adults the difference in prank calls between peer-to-peer conversations and adolescents-adult conversations at the Child Helpline is measured.

The experiment consisted of two parts. First, each class was shown a YouTube clip in which a 15-year-old boy prank called the Child Helpline.

(26)

Two versions of the YouTube clip were shown. The first version showed a prank call with a 53-year-old service employee. The second version showed a prank call where the service employee of the Child Helpline was 17 years old. In total groups (equal to Ttwo classes) saw the first version of YouTube clip, in which the service employee was an adult, and the other groups saw the second YouTube clip, in which the service employee was a peer.

The second part of the experiment contained a questionnaire. The questionnaire included questions and additional information. Before the participants watched the clip, the participants received a questionnaire containing a short introduction. The participants were instructed to imagine a certain situation; namely, being bored and watching the clip. After watching the clip the participants answered the question whether they were willing to start a prank chat. whether the participants would share their opinion of the video on Social Media. Furthermore the participants were asked how the employee of the Child Helpline was perceived. The answers to these questions are measured using the concepts of earlier research, which were discussed in the literature review. The questionairre also contained ten questions about the participant’s personality. In addition, control variable questions regarding age, gender, and years in school were included. The questionnaire concluded with the question whether the participants had ever called or chatted the Child Helpline and whether the conducted calls and chats were serious or prank calls or chats.

The experiment included 110 participants from four different classes of the Onze Lieve Vrouwelyceum in Breda. The majority of the participants completed the whole questionnaire. The participants were all in the first-year class of VWO and all the questionnaires were completed on the 9th of December, 2013.

3.2 Measures

(27)

Table 1 the variables from the conceptual framework are defined. Table 1. Defining the variables

Dependent Variable

Unwanted consumer behavior

This variable is measured by the answer to the question of whether the participants are willing to start a prank chat.

Prank chat/call Child Helpline

This variable is measured by the answer to the question asking if participants ever started a prank chat or prank call in real life.

Independent Variable

Observing unwanted consumer behavior

This is the variable identifying which YouTube clip was watched.

Personality of participants This is the personality index of the participants as measured by the 10-question personality index (5-point Likert scale)

Moderating Variable

Service by adults/adolescents This variable consists of two options, depending on whether an adult or an adolescent served the participant

Characteristics of the service employee

How do the participants feel about the service employee? (7-point Likert scale)

Four classes participated in the experiment. Two classes saw the video in which an adolescent prank-called an adult employee of the Child Helpline. The two classes were seperted into four groups. One group was asked to imagine watching the clip alone at home and talking to a peer employee of the Child Helpline. The second group was asked to imagine watching the clip alone at home and talking to an adult employee of the Child Helpline. Another group was asked to imagine watching the clip with their friends and talking to a peer employee of the Child Helpline. The last group was also asked to imagine watching the clip with their friends, but this group had to imagine talking to an adult employee of the Child Helpline.

The other two classes who saw the YouTube clip in which a adolescent prank called a peer employee of the Child Helpline were divided into the same four groups as the groups who saw the YouTube clip in which a adolescent prank called an adult employee of the Child Helpline. The YouTube clip was made for this experiment.

(28)

3.2.1 A simplified method to measure the big five personalities

Rammstedt & John (2007, p. 204) believed it was time to develop a shorter version of the BFI-44, because there is now more limited assessment time for surveys and experiments. Based on the findings of previous studies Rammstedt & John (2007, p. 205) asked if it was possible to provide an item set with only ten questions, which meant including just two items per type, instead of 44 questions. Their conclusion was that it is possible to use the BFI-10 instead of the BFI-44, because it possesses acceptable psychometric properties. However, there were some losses in comparison to the BFI-44. The BFI-10 is recommended when there is limited testing time.

Figure 2. Big Five Inventory 10 (BFI-10) made by Rammstedt & John in 2007.

The scoring of the BFI-10 scales is different from the BFI-44, because there are only 10 questions instead of 44 (Rammstedt & John, 2007, p.211). The score of question 1 reversed and the score of question 6 measure extraversion. The score of question 2 and the score of question 7 reversed measure agreeableness. The score of question 3 reversed and question 8 measure conscientiousness. Furthermore the score of question 4 reversed and question 9 measure neuroticism. Finally the score of question 5 reversed and the score of question 10 measure openness.

(29)

4 Results

This chapter describes the results obtained from the statistical analysis of the experiment. First, an overview of the data is provided. Second, the descriptive statistics and correlations are described, and, finally, the results of the tested the hypotheses are provided.

4.1 Data Overview

In total, 110 children participated in the experiment. 41.7% of the participants were male and 58.3% were female. The sample consists of first-year high school students in the age between 11 and 14 years old.. The largest group of participants (79.6%) consists of 12 year old students. In total four classes participated. A vast majority of the participants (77.8%) had never called the Child Helpline, 19.4% had called the Child Helpline as a prank. The other 2.8% of the participants had friends who had prank called the child helpline In the experiment, participants saw a YouTube clip in which a adolescent is prank calling the Child Helpline. There are two versions of the clip, containing the exact same information, except that the age of the employee of the Child Helpline is different. 51.8% viewed the YouTube clip in which the employee was an adult, and 48.2% viewed the YouTube clip in which the employee was a peer. 47.3% of the respondents were asked to imagine that they were bored and alone sitting behind the computer watching the clip, and 52.7% of the respondents were asked to imagine that they were bored and with friends behind the computer watching the clip.

After watching the YouTube clip, participants were asked if they were willing to start prank chat when visiting the website of the Child Helpline. The possible answers consisted of a 7-point Likert scale, where 1 was absolutely start a chat with the employee and 7 was absolutely not start chat with the employee of the Child Helpline. 28.2 % of the parciticpants would start a prank chat with the Child Helpline, 8.2% of the participants did notknow

(30)

whether they would start a chat, and 63.6% of the participants would not start a prank chat conversation with the Child Helpline. The dummy variable, which consists of a yes or no answer to the same question instead of the 7-point Likert scale, had 27.3% of the participants that would start a prank chat conversation with an employee of the Child Helpline and 72.7% would not start a chat conversation.

In addition, the participants were asked about their willingness to share, “like”, or comment on this clip on social media. 50% of the participants would not share, like, or comment on this YouTube clip on social media.

Figure 4. Social Media Reaction after watching the YouTube Clip

The other 50% of the participants would want to share the You Tube clip on social media and what their feelings were about the clip. 26.4% of the participants would “like” the video and 13.6% of the participants would only place a comment on the YouTube clip. All the comments these participants provided were negative. 1.8% of the participants who would share the YouTube clip believed the video was funny and that their friends would like to see the clip. The participants who had several reactions after watching the

50,0% 26,4% 1,8% 13,6% 3,6% 1,8% 2,7%

What would you do after seeing this YouTube Clip?

nothing like share comment like and share

like, share and comment like and comment

(31)

clip also saw the YouTube clip as a funny clip that other people would like to see. In summary, half of the participants would want to share their opinion on Social Media, and the other half did not want to share their opinion about the clip.

4.2 Descriptive Statistics & Correlations

Table 2 presents the descriptive statistics for the independent, dependent, and moderating variables. The variables representing how the participants feel about the employee were measured on a 7-point Likert scale, where 1 represents completely agree and 7 completely disagree. The means of these variables are between 1.97 and 3.32, indicating that the participants in this experiment feel comfortable chatting with an employee of the Child Helpline. The variables representing the Big Five personality traits, reflecting how the participants see themselves as a person, are measured on a 5-point Likert scale, where 1 represents completely disagree and 5 completely agree.

The answers to the first question; whether participants are willing to prank chat with the Child Helpline after watching the clip are given on a 7-point Likert scale, where 1 represents completely agree and 7 completely disagree. The mean of 4.98, indicates that the participants are more likely not to start a prank chat. The second question about starting a prank chat is a dummy variable to control the outcome, where 1 represents prank-chatting with the Child Helpline and 2 represents not prank chatting with the Child Helpline. The results for the dummy variable question are in line with the results of the first question. The responses show that the participants are more likely not to prank chat after watching the clip (mean =1.72). For the variables measured on a 7-point Likert scale the standard deviations are between 1.04 and 2.31. A standard deviation between 1 and 1.5 is reasonable for a 7-point Likert scale. Some of the variables have a larger standard deviation due to the relatively small sample size. The standard deviation of the variables measured

(32)

on a 5-point Likert scale is between 0.82 and 1.33. This is within the, generally considered reasonable, range.

Table 2. Descriptive statistics

N Minimum Maximum Mean Std. Deviation

YouTube clip 110 1.00 2.00 1.4818 .50196

Service by adult/peer 110 1.00 2.00 1.4818 .50196

Start Chat? 110 1.00 7.00 4.9818 1.90596

Dummy Start Chat? 110 1.00 2.00 1.7273 .44740 Connection with employee 107 1.00 7.00 3.3271 1.27217 Emphasizing with employee 108 1.00 7.00 3.0093 1.45651 Politeness of employee 110 1.00 6.00 2.0909 1.04526 Friendliness of employee 110 1.00 5.00 1.9727 1.04451 Caring of employee 110 1.00 6.00 2.3455 1.10409 Honesty of employee 109 1.00 6.00 2.4037 1.42152 Reliability of employee 110 1.00 22.00 2.6182 2.31462 Understanding of employee 106 1.00 6.00 2.3113 1.20604 Feel at ease with employee 110 1.00 7.00 3.0182 1.47133 Feel safe with employee 110 1.00 7.00 2.8182 1.55145

Reserved -R 87 1.00 5.00 3.6552 1.30141

Trusting 109 1.00 5.00 4.0183 .82755

Lazy - R 87 1.00 5.00 2.9770 1.49400

Relaxed - R 76 1.00 5.00 2.4737 1.23799

Few artistic interests - R 85 1.00 5.00 3.0118 1.56216

Outgoing 107 1.00 5.00 3.9252 1.04341

Tends to find fault with others- R 74 1.00 5.00 3.6486 1.28673 Does a thorough job 109 1.00 5.00 3.4220 .87447

Nervous 109 1.00 5.00 2.9174 1.17164

Active imagination 108 1.00 5.00 3.4352 1.24009

Child Helpline 108 1.00 3.00 1.9259 .46772

Prank Child Helpline 103 1.00 2.00 1.7961 .40485

Table 3 shows the correlations between the most important variables used in this research. In the experiment, two different questions are used to determine whether participants would want to start a prank chat after watching the YouTube clip. The variables are defined as Start Chat and Dummy Start Chat

(33)

in Table 3. Dummy Start Chat is an extra variable to test the effect of observing unwanted consumer behavior on unwanted consumer behavior. The results presented in Table 3 show that watching the YouTube clip has a positive correlation with the dummy variable for starting a prank chat with the Child Helpline. This indicates that there is a correlation between watching a prank chat and not being willing to start a prank chat.

There is also a significant correlation between whether the service was provided by an adult or by a peer and whether the participants are willing to start the prank chat after observing unwanted consumer behavior. It is a weak downhill linear relationship, which indicates that adolescents being served by peers are more likely to prank chat or call than adolescents being served by adults.

Moreover there is a uphill positive linear relationship between the two variables that indicate the willingness for adolescents to start a prank chat or call with the Child Helpline. This means that the variable Start Chat and Dummy Start Chat are positively related to each other.

In addition, there is a positive correlation between the variable ever started a chat or a call with the Child Helpline and the variable ever started a prank chat or a prank call with the Child Helpline. This can be explained by that adolescents who ever contacted the Child Helpline have contact the Child Helpline more often to prank call or chat than adolescents who have not contacted the Child Helpline. Other correlations are shown in Table 3.

Table 3. Correlations 1 2 3 4 5 6 7 8 9 1. YouTube clip 1.000 2. Service by adult/peer -.068 1.000 3. With(out) friends .010 .029 1.000 4. Social Media -.090 -.102 -.030 1.000 5. Start Chat? -.080 -.201 -.030 .119 1.000 6. Dummy Start Chat? .231 -.188 .009 -.071 .172 1.000 7. Child Helpline -.044 -.425 -.120 .045 .152 .178 1.000 8. Prank Child Helpline .067 -.308 .000 .142 .418 .219 .601 1.000

(34)

9. Gender .110 .087 -.078 -.130 .114 .139 .032 .070 1.000 Note: Correlations greater than r=(.16) are significant at p<0.05 (one-tailed).

4.3 Hypotheses Testing

In this paragraph the results of the tested hypotheses are discussed. The hypotheses are accepted if p<0.05.

4.3.1 Hypothesis 1

In order to test the effect of observing unwanted consumer behavior and engaging in unwanted consumer behavior, a t-test is performed. The effect of observing unwanted consumer behavior on unwanted consumer behavior is significantly higher than a neutral reaction, t(109) = 5.403, p=0.00, which means that observing unwanted consumer behavior does not lead to perform unwanted consumer behavior. Therefore, Hypothesis 1 is rejected – observing unwanted consumer behavior does not result in unwanted consumer behavior.

Table 4. Hypothesis 1 - Statistics

N Mean Std. Deviation Std. Error Mean Start chat? 110 4,9818 1,90596 ,18173

Table 5. Hypothesis 1 - One-Sample Test Test Value = 4

t df Sig. (2-tailed) Mean Difference 95% Confidence Interval of the Difference Lower Upper Start chat? 5,403 109 ,000 ,98182 ,6216 1,3420

In order to confirm this effect a dummy variable is used, where the adolescents had to answer yes or no to the same question. A chi-square test is used to check if two variables are associated with each other. The Chi-Square test is significant, which confirms the finding that H1 is rejected.

(35)

Table 6. Hypothesis 1 - Chi-Square Test

Would you start a chat with the Child Helpline?

Observed N Expected N Residual

Yes 30 55,0 -25,0

No 80 55,0 25,0

Total 110

Dummy Start Chat? Chi-Square 22,727a

df 1

Asymp. Sig. ,000

4.3.2 Hypothesis 2

A Independent Sample Test is performed to test the moderating effect whether the service representative is a peer or adult on the relation of observing and exercising unwanted consumer behavior. The general conclusion is in line with the results of Hypothesis 1. Adolescents would not perform unwanted consumer behavior after observing unwanted consumer behavior. Nevertheless, there is a slight difference when adolescents are chatting with an adult or a peer. Adolescents are not interested to prank chat an adult (mean = 5,1228). On the other hand adolescents are not really interested to prank chat when talking to a peer (mean = 4,8302). The mean is slightly lower, however the difference is not significant. Therefore H2 is rejected; the effect of Hypothesis 1 is not significantly stronger for adolescents being served by adults than adolescents being served by peers.

Table 7. Hypothesis 2 – Group Statistics

YouTube Clip N Mean Std. Deviation Std. Error Mean

Start chat?

Adult 57 5,1228 1,68046 ,22258 Peer 53 4,8302 2,12798 ,29230

(36)

Levene's Test for Equality of Variances t-test for Equality of Means

F Sig. t df

Start chat?

Equal variances assumed 7,710 ,006 ,803 108

Equal variances not assumed ,796 98,908

4.3.3 Hypothesis 3

In order to test whether adolescents being served by adults are more likely to prank chat than adolescents being served by peers, a t-test is used. The results from this t-test are significant for p<0.05. However, the results contradict the hypothesis., Therefore H3 is rejected; adolescents being served by adults are not more likely to prank-chat than adolescents being served by peers. This outcome supports the result of the testing of Hypothesis 2. Moreover, this test confirms that adolescents being served by peers are more likely to prank chat than adolescents being served by adults for p<0.05.

Table 9. Hypothesis 3 - Statistics

Service by adult/peer? N Mean Std. Deviation Std. Error Mean

Start Chat?

Adult 57 5,3684 1,74879 ,23163 Peer 53 4,5660 1,99528 ,27407

Dummy Start Chat?

Adult 57 1,8246 ,38372 ,05083 Peer 53 1,6226 ,48936 ,06722

Table 10. Hypothesis 3 - Independent Samples Test Levene’s Test for Equality of Variance

F Sig.

Start chat Equal variance assumed 4,544 0,035 Equal variances not assumed

Dummy start chat

Equal variance assumed 22,905 0,000 Equal variances not assumed

(37)

4.3.4 Hypothesis 4

In order to test which characteristics of the service employee have an effect on the relationship between the age (adult or adolescent) of the service employee and engaging in unwanted consumer behavior, first a Cronbach’s Alpha test is performed. The Cronbach Alpha test, tests whether all characteristics of employees may form one scale together. This is the case when The Cronbach’s Alpha is higher than 0.7 The Cronbach’s Alpha is higher than 0.7, namely 0.808, therefore one scale can be used as an interpretation if the characteristic of service employees have an effect on engaging in unwanted consumer behavior.

Table 11 shows that most of the characteristics have a weak to moderate uphill positive linear relationship with each other.

Table 10. Hypothesis 4 – Reliability Statistics

Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items

,808 ,830 10

Table 11. Hypothesis 4 – Correlations characteristics services employees

1 2 3 4 5 6 7 8 9 10 1. Politeness of employee 1.000 2. Friendliness of employee .492 1.000 3. Caring of employee .462 .591 1.000 4. Honesty of employee .485 .373 .444 1.000 5. Reliability of employee .168 .195 .220 .338 1.000 6. Understanding of employee .364 .448 .515 .453 .260 1.000 7. Feel at ease with

employee .114 .436 .212 .360 .355 .360 1.000 8. Feel safe with employee .274 .346 .252 .306 .379 .347 .592 1.000 9. Connection with

employee .221 .175 .193 .186 .225 .101 .184 .136 1.000 10. Emphasizing of

employee .574 .416 .390 .401 .212 .376 .193 .277 .365 1.000 Note: Correlations greater than r=(.16) are significant at p<0.05 (one-tailed).

(38)

In order to test if the characteristic of service employees have an effect on the service relationship and engaging in unwanted consumer behavior, an Independent Samples Test is used.

The Independent Sample Tests in Table 13 shows that the Levene’s Test is not significant for the characteristic which means that this data is not complete enough to answer this question and therefore, H4 cannot be answered – because the Independent Samples Test is not significant.

Table 12. Hypothesis 4 – Statistics

Start Chat? N Mean Std. Deviation Std. Error Mean

Connection with employee

Yes 30 3.0000 1.25945 .22994 No 77 3.4545 1.26226 .14385 Emphasizing of employee Yes 29 2.8276 1.44096 .26758 No 79 3.0759 1.46561 .16489 Politeness of employee Yes 30 2.1333 1.04166 .19018 No 80 2.0750 1.05272 .11770 Friendliness of employee Yes 30 1.9667 1.09807 .20048 No 80 1.9750 1.03085 .11525 Caring of employee Yes 30 2.6667 1.32179 .24132 No 80 2.2250 .99333 .11106 Honesty of employee Yes 30 2.8000 1.39951 .25551 No 79 2.2532 1.40939 .15857 Reliability of employee Yes 30 2.3333 1.39786 .25521 No 80 2.7250 2.57532 .28793 Understanding of employee Yes 29 2.5862 1.45202 .26963 No 77 2.2078 1.09229 .12448 Feel at ease with employee

Yes 30 2.9667 1.71169 .31251 No 80 3.0375 1.38202 .15451 Feel safe with employee

Yes 30 2.9333 1.70057 .31048 No 80 2.7750 1.50084 .16780 EmplTotal

Yes 30 2.6248 .95908 .17510 No 80 2.5817 .83665 .09354

(39)

Table 13. Hypothesis 4 – Independent Samples Test Levene's Test for Equality of

Variances

t-test for Equality of Means

F Sig. t df

Connection with employee

Equal variances assumed .515 .474 -1.674 105 Equal variances not

assumed

-1.676 53.040

Emphasizing of employee

Equal variances assumed .133 .716 -.784 106 Equal variances not

assumed

-.790 50.681

Politeness of employee

Equal variances assumed 1.088 .299 .260 108 Equal variances not

assumed

.261 52.634

Friendliness of employee

Equal variances assumed 1.649 .202 -.037 108 Equal variances not

assumed

-.036 49.357

Caring of employee

Equal variances assumed 3.750 .055 1.890 108 Equal variances not

assumed

1.663 41.894

Honesty of employee

Equal variances assumed .844 .360 1.813 107 Equal variances not

assumed

1.818 52.731

Reliability of employee

assumed

-1.018 93.939

Understanding of employee

Equal variances assumed 7.173 .009 1.448 104 Equal variances not

assumed

1.274 40.529

Feel at ease with employee

assumed

-.203 43.948

Feel safe with employee

Equal variances assumed .016 .899 .475 108 Equal variances not

assumed

.449 46.945

EmplTotal

Equal variances assumed 1.286 .259 .231 108 Equal variances not

assumed

(40)

4.3.5 Hypothesis 5

To test whether the personality index of adolescents influences prank-calling, this study investigates whether participants with openness to experience and extraversion personalities are more likely to already have engaged unwanted consumer behavior in the past. First the Big-Five personalities were tested with the Cronbach’s Alpha test. The outcome of the Cronbach’s Alpha of the five personalities are not reliable.

Therefore the ten characteristics are tested independent with the Wald criteria The Wald Criteria tests whether only two of the characteristics had a logistic regression between the characteristics and if they ever had prank called the Child Helpline using the Wald criteria (Wald > 3,89). When adolescents are reserved there is a positive relation between being reserved and not prank called the Child Helpline before. When adolescents are lazy, there is a negative relation between being lazy and not prank called the Child Helpline before. Therefore, H5 is partly accepted – when adolescent are reserved, adolescents tend to call less often compared to other adolescents and when adolescents are lazy, they call the Child Helpline more often compared to other adolescents.

Table 14. Hypothesis 5 – Logistic Regression: Variables in the Equation B S.E. Wald df Sig. Exp(B)

Age -,273 ,680 ,162 1 ,688 ,761 Gender -,763 ,819 ,868 1 ,352 ,466 Reserved ,870 ,386 5,073 1 ,024 2,387 Trusting ,002 ,486 ,000 1 ,996 1,002 Lazy -1,274 ,458 7,746 1 ,005 ,280 Relaxed -,272 ,330 ,677 1 ,411 ,762 Few artistic interests ,134 ,241 ,309 1 ,578 1,143 Outgoing ,415 ,483 ,739 1 ,390 1,515 Tends to find faults by

others

(41)

Does a thorough job -,147 ,415 ,125 1 ,724 ,864 Nervous ,264 ,320 ,680 1 ,409 1,302 Active imagination ,274 ,323 ,721 1 ,396 1,316 Constant 4,778 9,412 ,258 1 ,612 118,852

Unwanted consumer behavior : the influence of observing others

Amsterdam Business School