• No results found

In this research, the focus is on these challenges and new possibilities

N/A
N/A
Protected

Academic year: 2021

Share "In this research, the focus is on these challenges and new possibilities"

Copied!
55
0
0

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

Hele tekst

(1)

Relational trust-building in B2C e-Commerce

New Communication Methods as Driving Force

Master Thesis University of Groningen Faculty of Economics & Business

MSc International Business & Management

18 March 2016

W.W. (William) van den Hoorn

william@vandenhoorn.eu Student number 1769634

Supervisor: dr. A.A.J. (André) van Hoorn

(2)

Abstract

E-commerce is growing rapidly and is getting more important for (international) businesses all over the world. Borders are fading in a constantly growing online environment. Communication between companies and (potential) customers is also shifting from the “real world” to the online world. This shift in location brings new possibilities but also challenges. In this research, the focus is on these challenges and new possibilities. Previous research indicates that building trust in a personal way, by direct contact, is preferred by (potential) customers. Personal trust-building is the most common in regular commerce, when there is direct contact with employees of a

company. In e-commerce, trust-building is mainly based on the calculative view, when the contact between a company and a (potential) employee is more indirect and one- time. So, communication is shifting from relational trust to calculative trust, and this shift is not regarded positive. This is very interesting from the perspective of this study because this different way of communication and trust-building is a key characteristic of e-commerce. In this research, it is found that a remedy to this situation is the use of new communication channels or methods. This new insight serves as a further

expansion and an update of social presence theory in an e-commerce setting by adding the definition “Social Communication”. An example of social communication is via social media, in the form of webcare. Webcare gives companies new possibilities to engage with customers in a much more personal way. Also, communication via webcare could be much richer in comparison to more traditional means like telephone or a letter. This leads to more personal contact in e-commerce and is effectively raising the level of relational trust. To clarify this conclusion an existing model regarding trust and e-commerce was updated. To further support the conclusions of this research, a survey was conducted. This survey provides a provisional test of one specific aspect of this new model. In this survey only a weak link between webcare and different,

possibly higher, perceived benevolence was found.

Keywords E-commerce Trust Building Webcare Social Presence Social Communication

(3)

Acknowledgements

I would like to express my gratitude to my supervisor Dr. Van Hoorn for the useful comments and feedback, and the useful brainstorming throughout the learning process of this master thesis. Furthermore, I like to thank all participants in my survey and experiment, who have shared their precious time to help me finishing this master thesis. I would also like to thank my immediate family for supporting me throughout this journey and throughout the whole of my studies in all ways possible.

(4)

Table of Contents

Abbreviations p. 06

1. Introduction p. 07

1.1 Background & Motive p. 07

1.2 Research Objective & Problem Identification p. 08

1.3 Research Gap p. 09

1.4 Purpose p. 10

1.5 Research Questions p. 10

1.6 Implications of this Research p. 11

2. Research Method & Research Design p. 12

3. Theoretical Definitions p. 13

3.1 E-commerce p. 13

3.2 Trust p. 13

3.3 Social Presence p. 16

3.4 New Communication Channels: Webcare and social media p. 17

3.5 Traditional Communication Channels p. 18

3.6 Communication Methods in this Research p. 19

4. Theoretical Analysis p. 20

4.1 Scope of the Analysis p. 20

4.2 Dimensions of trust p. 20

4.3 Trust and e-commerce p. 21

4.4 Connecting the dots: “Social Communication” p. 24

(5)

5. First Try at Testing Findings: A Survey About Webcare p. 26

5.1 Survey Design p. 26

5.2 Survey Hypotheses p. 29

5.3 Survey Conceptual Model p. 30

5.4 Survey Findings p. 31

6. Concluding Discussion p. 33

6.1 Conclusions p. 33

6.2 Conclusions from the Survey p. 36

6.3 Reliability & Validity p. 37

6.4 Limitations and Further Research p. 39

References p. 41

Appendices p. 45

(6)

Abbreviations

B2C Business-to-consumer

E- Electronic [digital]

EEA European Economic Area

IB International Business

IT Information Technology

KLM Royal Dutch Airlines

NS Dutch Railways

NWOM Negative Word Of Mouth

TCE Transaction Cost Economies

TTIP Transatlantic Trade and Investment Partnership

(7)

1. Introduction

1.1 Background & Motive

E-commerce is a growing business, and it is growing fast. Where worldwide e- commerce sales had a worth of only 1233 trillion US dollars in 2013, it is expected that this amount will almost be doubled in 2018, to 2356 trillion US dollars

(EMarketer, 2014). Also, borders are fading because of for example the expansion of the European Economic Area (EEA) and trade agreements like the Transatlantic Trade and Investment Partnership (TTIP). This makes that companies engaging in (digital) cross-border business are getting more opportunities and are becoming more

important. Especially companies that start trading internationally from the start (born globals) (Knight, 2010).

Growth in global online sales and globalization (P. Williamson & Zeng, 2004) are resulting in a situation where companies are searching for new ways to be in contact with their customers. Distances between companies and (potential) customers are getting bigger because (potential) customers are normally not visiting companies in person in their brick-and-mortar shops anymore. The face-to-face contact an employee and a (potential) customer have in a physical store is mainly building personal trust, and this has to be substituted by other communication means (also known as channels or methods). Therefore, companies are not in the position to gain trust by interacting in-person with their customers (Reichheld & Schefter, 2000). This research wants to expand social presence theory by showing that new communication channels are a next step in trust-building in e-commerce.

(8)

1.2 Research Objective & Problem Identification

This research want to contribute to the literature by connecting previous research about e-commerce, trust, and communication. Doing so leads to an answer to the challenges in communication for e-companies, the biggest challenge being the diminishing personal interactions. There is already research on social presence, that is about personalizing communication and interaction within a website by giving users cues about the personality of the contact (Dennis & Kinney, 1998; David Gefen & Straub, 2004; David Gefen, 2000; Lee & Turban, 2001; Moon & Kim, 2001). This research wants, however, take a look at real communication between (potential) customers and the company.

The assumption in this research is that new communication channels like webcare have characteristics that really make personal contact posible. This means that they are a good way of replacing interpersonal interactions, and thus relational trust. On the other hand, more traditional ways a (potential) customer can contact an e-company are not great substitutes. This difference is mainly suspected because of richer

communication possibilities (pictures, videos) (Short, Williams, & Christie, 1976) that many new communication methods have. When combined with the fact that

communication like webcare is very informal (Dietz, 2015) it is expected that

communication is perceived personal. Therefore, it is expected to be generating more relational trust, or trust in the integrity and benevolence of a company (Daft & Lengel, 1986; Rains, 2007). Also, the fact that a communication channel like social media is synchronous in contrast to traditional communication that is asynchronous and mediated contact (Short et al., 1976) reinforces this concept. Research on these connections isvery relevant because a higher trust level leads to a higher level of purchase intentions. In general it is important because of the before mentioned

changing global business environment and the challenges and opportunities attached to this new IB reality.

(9)

1.3 Research Gap

Although commerce shifts to e-commerce, local business are becoming global

businesses and borders are fading there is not much research on how building trust in this new environment is achieved. Much research was done on how in real-life, face- to-face, interpersonal contact is influencing trust (Swan & Nolan, 1985), but only a minimal amount of research was carried out on digital contacts or social presence. The limited amount of research on social presence is focussing on this presence on the website of the companies itself (for example chat), rather than communication via new channels (Dennis & Kinney, 1998; David Gefen & Straub, 2004; David Gefen, 2000;

Lee & Turban, 2001; Moon & Kim, 2001). This research is being expanded in this thesis.

Researchers found that calculative trust and familiarity is the most important trust generated in e-commerce, while relational trust is more important (Schoder &

Haenlein, 2004). This clearly shows that there is a gap between the trust type that is important to customers (relational trust) and the trust type that is perceived and generated while doing business in e-commerce (calculative trust). On the other hand, research on webcare is mostly focusing on negative exchanges or Negative Word of Mouth (NWON) (Hong & Lee, 2005; Willemsen, Neijens, & Bronner, 2013). Another research about online trust is focussing on feedback from other consumers (Vermeulen

& Seegers, 2009). This raises the question of how to raise relational trust in this new environment, by combining research about trust and conducting research on contact via new communication methods like webcare.

(10)

1.4 Purpose

In this thesis the question raised in previous research will be answered, about how to raise relational trust in e-commerce by specifically looking at new communication channels like webcare as a mean to achieve this. This because relational trust is regarded the most important in doing business (Schoder & Haenlein, 2004), but is generated the least (David Gefen & Straub, 2004; David Gefen, 2000; Lee & Turban, 2001). Because of this importance, it is expected that higher relational trust is boosting overall trust more, compared to higher calculative trust.

1.5 Research Questions

To conclude the introduction above, the main question is:

Are new communication channels (like webcare) an expansion of social presence theory, and thus a good method for relational trust-building in an e-commerce setting?

To answer this questions the follow sub-questions have been formulated:

What is trust related to e-commerce?

What methods are there to communicate in an e-commerce setting, and how does this affect trust?

Can new communication methods like webcare be viewed as an expansion of theory on social presence in e-commerce and therefore update existing theory?

(11)

1.6 Implications of this Research

This report is interesting for scholars because this research tries to provide a further solution for the discrepancy between appreciated personal trust and generated

calculative trust in e-commerce. In the changing IB field, where globalization is one of the key factors for a shift from commerce to e-commerce this is getting more and more important. An interesting and meaningful subject. By adapting an existing model regarding trust and e-commerce, this thesis expands existing insights about social presence. From a managerial standpoint, this research is interesting because companies have a real reason to engage and invest in webcare if the connection between more relational trust and webcare is demonstrated. When they combine the knowledge from previous research about the high amount of calculative trust in e-commerce and this new insight on how to generate more relational trust, they can raise the overall trust level, resulting in more sales. This is especially important because earlier research found that relational trust is the more important than calculative trust, but less achieved (David Gefen & Straub, 2004; David Gefen, 2000; Lee & Turban, 2001; Schoder &

Haenlein, 2004).

(12)

2. Research Method & Research Design

To give an answer to the research question, a combination of desk research and literature analysis is conducted. Not by only looking at scientific papers and articles, but also by incorporating articles from renowned sources in the mainstream media.

The reason for this is the fact that e-commerce and communication methods are changing and expanding rapidly. This makes that the information that is used on the few recent scientific articles in this research area is sometimes obsolete when the paper is published after the peer-review process. To conclude this process an existing model about trust, e-commerce and social presence was adapted. A part of this model was tested by conducting a survey combined with a quasi-experiment. The research method and research design of this survey are mentioned in chapter 5.

(13)

3. Theoretical Definitions

3.1 E-commerce

Gartner Advisory Group, via (Damanpour & Damanpour, 2001) describes e-commerce as business activities that use the new, online, connected economy to create value. E- commerce or e-business is described in the terms of the quantity a company is

involved in this process, measuring the degree that a company is conducting business online, using electronic business channels. E-commerce is not a rigid concept, but can be implemented in many forms and many different degrees. However, the presence online on the internet is a crucial component of e-commerce. Companies that are using computer interactions to conduct business with other businesses or consumers are involved in e-commerce. In this research e-commerce between a consumer and a business (B2C) will be studied.

3.2 Trust

(Giffin, 1967) explains that trust typically consists out of three trust dimensions that were already described long ago in Aristotle's’ Rhetoric, namely: intelligence (or ability), good character (honesty or integrity), and good will (benevolence).

(14)

3.2.1 Calculative trust

Theories on calculative trust or trust in the ability of a partner (Mcallister, 1995) are based on research on implicit contracting, Transaction Cost Economics (TCE), and game theory (Axelrod, 1984; O. E. Williamson, 1993). This previous research

proposes that a structure that adjusts incentives with rewards can lead to a stable and predictable outcome. Calculative trust is building trust in a very rational way, by

deliberately assessing conditions in the future. This includes calculating future benefits and costs and possibilities, and enforcement costs (costs of cheating, for example, costs needed to terminate a relationship) (Bromiley & Harris, 2006; Lewicki, Tomlinson, & Gillespie, 2006). When calculative trust is at a high level, the parties involved do believe that the cooperation will be a success and goals will be achieved.

This is because if one of the parties is falling short or does not have the ability to hold his end of the deal, it will lead to penalties that can go as far as terminating the

relationship (Parkhe, 1993). So, calculative trust is the regulation and conservation of partnerships due to decreased opportunistic behavior that happens because of these possible penalties. Calculative trust is, therefore, a very rational assessments of the expected rewards and penalties, or, in other words, the idea that choices are based on a rational derivation of costs and benefits. It is needed to check continuously whether it pays to cooperate, or if it is better to cheat (Saparito et al., 2015).

(15)

3.2.2 Relational trust

Other scholars are arguing that a stable and predictable relationship is more based on a relational foundation (Granovetter, 1985; R Gulati, 1995). Getting to know each other and interacting with each other let the parties collect experiences and form

expectations of the other party. Also, the development of shared values between the parties involved and normative conventions that are found to be corresponding between partners will define how the partners will be working together (Bercovitz, Jap, & Nickerson, 2006). There is relational trust when there is no need for the partners to monitor constantly the situation where the partnership is in. Nor is there a need to be checking whether it pays to cheat or to cooperate or to constantly reassess the relationship (Dyer & Singh, 1998). In a situation with high relational trust, the partners develop a sort of mutual understanding. They “think like” each other, “feel like” each other and “respond like” each other (Lewicki & Bunker, 1995). This bond helps parties in making decisions, is reducing the risk of opportunistic behavior and results in outcomes of the partnership that are to the benefit of both partners.

Relational trust is thus based on a longer period, on repeated interactions and the overall perceived quality of the partnership, rather than on a single transaction within this partnership. It is more about feelings than about facts (Dyer & Singh, 1998).

(16)

3.2.3 Trusting Disposition

When measuring trust, trusting disposition, or the tendency to trust other parties is important (Mayer et al., 1995). Previous research conducted acknowledged that this also holds in the case of e-commerce (David Gefen & Straub, 2004). Therefore, to measure disposition to trust, the combined work of (Fukuyama, 1995; McKnight, Harrison Cummings & Chervany, 1998; Rotter, 1971) is used. This results in the following definition for disposition to trust: “Disposition to trust is a general inclination to display faith in other humans, and to adopt a trusting stance towards them (McKnight, Harrison Cummings & Chervany, 1998). This means that people believe that humans in general are trustworthy and they give people the credit of trusting them, regardless of the justification of it (McKnight, Harrison Cummings &

Chervany, 1998). This behaviour is the result of lifelong experiences (Rotter, 1971) and socialization” (McKnight, Harrison Cummings & Chervany, 1998). Trusting disposition is the most effective in the initiation phase (where people get to know each other), and they do not know about other trust-building beliefs like integrity,

benevolence and ability (McKnight, Harrison Cummings & Chervany, 1998).

3.3 Social Presence

The first research on social presence is about the amount of acoustic, visual and physical contact that can be achieved when being in contact with someone (Short et al., 1976). In more recent research about social presence in an e-commerce setting a slight different explanation is given to this theory. Here we see that social presence is the sense of the presence of another interaction partner when doing computerized task like online shopping. It is used when describing communication with a computer or with a website, which is normally a situation with no social presence. It could be as basal as adding a picture of a person to a website or give users a way to give reviews (David Gefen & Straub, 2004). In other words, social presence describes how online customers respond to technology that exhibits social behaviour (Moon & Kim, 2001).

Researchers also found that the amount of social presence predicts the success rate of a complex task like shopping online (e-commerce) (Dennis & Kinney, 1998). This latest view on social presence theory will be used, and updated, in this thesis.

(17)

3.4 New Communication Channels: Webcare and Social Media

Webcare is a way of communicating on social media. It is not technology that exhibits social behaviour or is a surrogate to personal contact, but is gives a real possibility to be in personal contact with another human being. This social media consists out of different rich digital communication channels that enable people to communicate with each other, but also with for example companies online. These communication

channels are also known as social media and are currently offered by companies like Facebook, Twitter, LinkedIn, Instagram, and Whatsapp (Hennig-Thurau et al., 2010) In recent history also other channels like MySpace and Hyves were present, but these disappeared, and new ones appeared. Social media is characterized by the possibility to use rich media (pictures, sound, photo’s) and by the fact that it is perceived much more synchronous when comparing it to e-mail. The engagement of companies on social media is what is called webcare (Kerkhof, Beukeboom, & Utz, 2009). Previous research conducted does suggest that people find it important to be able to

communicate with media richness (Short et al., 1976) and to communicate in a direct way that is perceived the most interpersonal (Daft & Lengel, 1986). New

communication channels like webcare have some of these special characteristics.

Scholars see that “Social media introduce[s] substantial and pervasive changes to communication between organizations, communities, and individuals” (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011). Social media, and with it webcare, tends to be very informal (Dietz, 2015), and by looking at the social media presence of Dutch companies that are most active on social media (ICT Magazine, 2015): NS, KPN, PostNL, Ziggo, and KLM; it is evident that webcare is also made personal, by using first names and pictures of employees. This personal touch is important because trust is only generated when the other party is not anonymous, but some information on this person is known (Rains, 2007). It also connects to the media richness and interpersonal contact desires that people tend to have. Many companies all over the world are

making use of these possibilities to stay in contact with their customers. For example, more than 50% of the small and medium businesses in the US were present on social media in 2011 (Zoomerang, 2011).

(18)

3.5 Traditional Communication Channels

More traditional communication channels to contact a company are contacting a company by telephone, writing a letter to a company, or sending a fax or an e-mail.

Traditional communication is, in contract to communication via webcare,

asynchronous and mediated and is therefore perceived as a not-so personal way to have contact (Short et al., 1976). This is also because there is not much possibility to enrich the communication, or in other words use for example gestures, intonation, pictures, video’s, emoticons or other ways to express emotions (Daft & Lengel, 1986;

Short et al., 1976).

(19)

3.6 Communication Methods in this Research

As mentioned before, webcare and traditional communication channels are used in this research. In order to make a clear distinction between several communication methods, one traditional communication method was chosen, and one new communication method. This is the best way to identify differences and is required to conduct a survey as described in chapter 5. As traditional communication method e-mail was used.

Mainly because this research is about e-commerce, and this is the only digital method.

For webcare or social media communication there are many channels eligible to be used in this research. The characteristics of these channels are highly comparable, so the channel that is used the most (figure 1) and is very known for webcare is used, Twitter.

Communication Methods in this Research Interpersonal

Contact

Traditional Communication

New communication channels (Webcare) Examples Personal,

face-to-face contact

Contact by telephone, letter, fax, e-mail

Twitter, Facebook, LinkedIn, WhatsApp, Instagram

Used in this research

Store visit E-mail Twitter

Characteristics Synchronous, personal, includes verbal and non- verbal

communication

Asynchronous, non- personal, scores low on richness

Synchronous, personal, scores high on richness so able to substitute verbal and non- verbal communication Table 1. Communication Methods in this Research

(20)

4. Theoretical Analysis

4.1 Scope of the Analysis

Analyzing the role trust and communication methods play in e-commerce is the scope in which the analysis of literature takes place. This is done by first looking at how trust is being built and perceived, and then looking at the role trust plays in e-commerce.

Lastly, the role of communication and trust is being looked upon, to be able to draw conclusions.

4.2 Dimensions of trust

After a literature analysis I found that most researchers, over the years, agree with Giffin that perceived trustworthiness is consisting out of trust in the integrity, ability and benevolence of the other party, as shown in Table 2 (Blau, 1964; Crosby, Evans,

& Cowles, 1990; Elangovan & Shapiro, 1998; D Gefen, Karahanna, & Straub, 2003;

Giffin, 1967; S L Jarvenpaa, Knoll, & Leidner, 1998; Luhmann, 1982; Mayer, Davis,

& Schoorman, 1995; Moorman, Deshpande, & Zaltman, 1993). Therefore, I will use these also in my assumptions. Here, we also see that the basic sociological work of Blau, dated 1964, still holds with research done over dozens of years.

The dimensions of trust in the literature

Blau (1964) Integrity, benevolence, and ability

Crosby et al. (1990) Integrity and benevolence Elangovan & Shapiro (1998) Benevolence and integrity

Gefen et al. (2003) Ability, integrity, and benevolence Giffin (1967) Integrity, benevolence, and ability Jarvanpaa et al. (1998) Ability, integrity, and benevolence

Luhmann (1982) Integrity, and benevolence

Mayer et al. (1995) Ability, benevolence, and integrity Moorman et al. (1993) Integrity and ability

Table 2. Dimensions of trust according to scholars throughout the years

(21)

The three dimensions of trust found can also be divided in the way of types of trust, namely in calculative trust and relational trust. Ability would fall under the trust type calculative trust, where integrity and benevolence would fall under relational trust (Mcallister, 1995).

4.3 Trust and e-commerce

Trust is very important online and when new technologies are introduced (Tanaka, 2005). This is even more important when looking at e-commerce. This is because the traditional web interface (e-mail, websites) does normally not allow the (potential) customers of an e-company to judge if a vendor is trustworthy (scores on ability, integrity, and benevolence). In situations where there is face-to-face two-way interaction on the other hand, trust is being gradually built (Reichheld & Schefter, 2000). A situation with this sort of direct contact would be for example happening in brick-and-mortar stores. Also, because of the fact that e-companies can take advantage of customers easily online, trust is a very important issue when doing business online (Sirrka L. Jarvenpaa & Todd, 1996). Without trust in a company people would not buy from this company.

An example of e-companies taking advantage of customers can be the sharing of individual customer data to third parties. This is clearly a case of opportunistic

behavior in a relationship where the company is taking advantage in this relationship.

In the past, Amazon, a big e-company that is doing business all over the world, lost customers because they introduced a policy that made it possible to share information on the activity of their individual customers (Rosencrance, 2000). However, having learned from this and being aware that an e-company needs to constantly rebuilds trust, they changed the way they treat customer data, and now promise never to sell data that is traceable to an individual (Amazon, 2014).

(22)

When looking at the different types of trust, research shows that disposition to trust and familiarity with an e-company is very important. Also, the familiarity with a company is important regarding trust generation (R Gulati, 1995; Ranjay Gulati &

Sytch, 2008). Research conducted by (David Gefen & Straub, 2004) confirmed that this subject is also important in the case of e-commerce. So, familiarity will be included by using the definition of familiarity by (Luhmann, 1982). He describes familiarity as “an understanding, often based on interactions from the past, past

experiences, and the learning of what, why, where, and when other people do that they do.”

When measuring trust, trusting disposition, or the tendency to trust other parties is important (David Gefen & Straub, 2004; Mayer et al., 1995). This is followed by a high level of perceived ability, or calculative trust (David Gefen & Straub, 2004;

David Gefen, 2000; Lee & Turban, 2001). However, customers tend to emphasize more on the importance of relational trust, or the perceived integrity and benevolence of a partner (Schoder & Haenlein, 2004), like in brick-and-mortar shops where

customers interacted with (employees of) a company. This shows that there is a gap between the type of trust that is important for customers, and the type of trust that is typically generated.

(23)

Previous research found that there are ways to let people experience some personal contact by embedding certain technologies into a website (David Gefen & Straub, 2004; Moon & Kim, 2001). This corresponds to the fact that people find it important to be able to communicate with media richness (Short et al., 1976) and to

communicate in a direct way that is perceived the most interpersonal (Daft & Lengel, 1986). New communication channels like webcare have these special characteristics.

Scholars see that “Social media introduce[s] substantial and pervasive changes to communication between organizations, communities, and individuals” (Kietzmann et al., 2011). Social media, and with it webcare, tends to be very informal (Dietz, 2015), and by looking at the social media presence of Dutch companies that are most active (ICT Magazine, 2015) on social media, NS, KPN, PostNL, Ziggo, and KLM, it is evident that webcare is also made personal, by using first names and pictures of employees. This personal touch is important because trust is only generated when the other party is not anonymous, but some information on this person is known (Rains, 2007). It also connects to the media richness and interpersonal contact desires that people tend to have.

(24)

4.4 Connecting the dots: “Social Communication”

Not just letting (potential) customers experience personal contact, but really making this personal contact happen. That is what new communication channels are all about.

Instead of creating a social environment within an e-commerce environment, new communication methods have the same characteristics as social presence but are a channel on top of the website of a company. I argue that by introducing new

communication technologies to be in contact with (potential) customers the idea of social presence in e-commerce is taken to a new level. Contact via social media services like Facebook, Twitter, LinkedIn, Instagram, and WhatsApp is a good example of this new contact. A good description of these new communication

channels would be “social communication”. This because it is based on a high social presence with all its characteristics but is mere a new way of communication. A new definition for this type of communication is warranted because of the big gap with traditional communication and the total different approach (real personal

communication versus give a party the feeling that there is personal communication).

One of the main reasons for the succes of using new communication channels like webcare is that media rich and synchronous contact is possible with webcare, that can substitute face-to-face interpersonal contact (Daft & Lengel, 1986; Short et al., 1976).

Finding communication methods that substitutes personal communication is thus possible. It is also very important because of the multiplicity of reasons indicated in this part. Companies that are in the business of e-commerce should embrace new communication technologies like social media to engage with their customers, because of the reasons mentioned before.

New definition: SOCIAL COMMUNICATION Communication between two or more physically divided participants that is rich, synchronous, and is substituting interpersonal contact, via new communication methods that is building trust in a B2C e-commerce environment.

(25)

A good way to visualize the addition of social communication to theory is a conceptual model. Because relatively recent a comprehensive model about social presence was drafted (David Gefen & Straub, 2004), this model will be used as a basis. This existing model is about social presence in e-commerce by looking at technologies to mimic personal communication. This model needs to be updated to show the attribution to the literature that new communication methods like webcare are in fact better than

mimicking personal contact, but are really making this contact possible.

Model on Social Presence and Social Communication in e-Commerce

Figure 2. Model on Social Presence and Social Communication in e-Commerce e-Trust

Integrity Benevolence

Ability

Control Variables

Trusting

disposition Familiarity

Purchase Intentions Social Presence

Social Communication

(26)

5. First Try at Testing Findings: A Survey About Webcare

After the theoretical analysis in this research a first try to test the new model is done with a survey including a quasi-experiment about this subject. To do this I will look at a very specific situation: a comparison between two contact methods between a

potential customer and a company. Because of the limited time and resources available this very specific situation was chosen to be queried.

5.1 Survey Design

A comparison between two groups have to be made to test the new social

communication definition. These are a group that experienced a traditional way of communicating with a company and a group that experienced social communication, in this case in the form of webcare. Besides analyzing literature and drawing

conclusions from other scholars, this chapter aims to provide supporting evidence in a quantitative way. This is done by looking at a specific situation using a survey with quasi-experiment.

Respondents in the survey have to be divided in two groups. In each group

respondents have to be scored on the three dimensions of trust (Table 3) and their familiarity with the company and their trusting disposition. To be able to really measure the trusting disposition of the respondents, a well-known company in The Netherlands was chosen to use, namely Royal Dutch Airlines (KLM). This is the flag- carrier of The Netherlands and therefore it is expected that participants have a trusting disposition towards this company.

Because of the nature of the research question a nested research design is appropriate.

Therefore, a survey with a limited form of a controlled experiment or quasi- experiment, in a nested design was conducted (Appendix 2).

(27)

5.1.2 Survey Treatment Groups

Two treatment groups were formed by automatically serve different content to respondents, randomly.

First, respondents did receive the same questions about their familiarity with KLM and their trusting disposition. For familiarity they could choose if they knew KLM and if they did ever fly with KLM. For trusting disposition two theorems were given to be scored on a scale from 1 (Completely Disagree) to 10 (Completely Agree). Then, respondents were randomly assigned to either a conversation with an employee via webcare (group 1) (Appendix 3) or a conversation with an employee via e-mail (group 2) (Appendix 4), this was the random controlled experiment. To make sure the

difference between the different communication methods is perceived, the webcare conversation was much more informal compared to the e-mail or traditional

conversation, while keeping the customer question the same. In the case of the e-mail conversation an extensive introduction about the training of the employee was given, and much more emphasis was laid on the professionalism of the company (KLM) and the employee. After one of these two randomly assigned conversations, respondents were asked to score KLM on the three dimensions of trust (ability, integrity,

benevolence), also on a scale from 1 (Completely Disagree) to 10 (Completely Agree) and ask them about their purchase intentions. Scales and questions used to do this were derived from the work of (David Gefen & Straub, 2004), see Appendix 1.

(28)

The survey and controlled experiment was conducted via Qualtrics and was spread with the help of Facebook. The choice of distributing the survey only via Facebook was a deliberate choice, because this way it was sure that the respondents have at least a basic knowledge about webcare, because they are active on it (in this case on

Facebook). The survey was not distributed on Twitter because of the lack of active users with regards to sharing and engaging with this kind of call for respondents.

Because the research is conducted in The Netherlands it is beneficiary for the number of respondents to offer the survey in both English and Dutch, so the survey and the conversations were translated in Dutch.

Treatment/Respondent Groups Group 1 – Social Communication Group 2 – Traditional

Webcare conversation Traditional (e-mail) conversation Expecting significant high(er) scores on

integrity, benevolence (relational trust)

Expecting significant high(er) score on ability (calculative trust)

Table 3. Treatment/Respondent groups in the nested design of this research

For both groups a sample size of 50-100 (group 1 and group 2) would be preferable as suggested in the literature (Thomas, 2004).

(29)

5.2 Survey Hypotheses

Because this survey will zoom in on a specific situation, being the contact between a potential consumer via two communication methods, namely Twitter and e-mail, the research question can be narrowed a bit. By doing so it is very evident that the scope of this survey is narrower, but the basic idea of the research question is still there.

Is webcare via Twitter a better method for relational trust-building in an e- commerce setting, compared to traditional communication via e-mail?

Research Question for this Chapter

To answer this research question two hypotheses are formed, particularly:

H1, that is about the assumption that webcare is influencing integrity and benevolence, and thus the level of relational trust in a higher degree than traditional communication.

This because of the earlier connection of research about being a more personal communication method (Rains, 2007), having richer communication possibilities (Short et al., 1976), and being able to communicate in a more direct way (Daft &

Lengel, 1986) because of informality (Dietz, 2015).

Hypothesis 1 (H1): Webcare via Twitter is influencing the perceived integrity and benevolence more than traditional communication

And H2, on the other hand, that is about confirming earlier research about calculative research (thus including ability), namely that e-commerce and the traditional means connected to it (for example telephone, letters, fax, and e-mail) is influencing the perceived ability of a company in the highest degree (David Gefen & Straub, 2004;

David Gefen, 2000; Lee & Turban, 2001).

Hypothesis 2 (H2): Traditional communication via e-mail is influencing the perceived ability more than webcare communication

(30)

Social Communication 5.3 Survey Conceptual Model

The new model created in the theoretical analysis part of this research will be used as a basis for this conceptual model. To count for the fact that this research is focussing on the differences between the two groups instead of the scores on the different

dimensions another dimension is added to the model. These facts combined leads to a conceptual model (figure 2).

Figure 3. Conceptual model Webcare (Twitter)

Traditional comm. (E-mail)

e-Trust Integrity

Benevolence

Ability

Control Variables

Trusting disposition

Familiarity with KLM

H1

H2

(31)

5.4 Survey Findings

77 respondents did finish the questionnaire and experiment, while 68 respondents answered all questions. Only surveys with all questions answered are included in this analysis of the results. Because of this, 33 of the respondents were in the group that was randomly selected for the e-mail (traditional) conversation, while the other 35 respondents were in the group that was randomly selected for the webcare

conversation.

Because two different groups have to be compared, with no overlap between them and because the answer possibilities were in the form of a Likert scale (ordinal), a Mann- Whitney U-Test was performed to first explore if there are differences between the two group, after which Hypotheses 1 and 2 could be tested on the three dimensions of trust: Ability (AB1), Integrity (IN1), and Benevolence (BE1). The Mann-Whitney U- Test is conducted with an alpha-level of 0.05, so an 95% confidence interval and is executed with help of statistical software SPSS (The full SPSS output is available in Appendix 6). This results in the following descriptive statistics:

Mann-Whitney U-Test - Descriptive Statistics

Group N Mean Std. Deviation

AB1 - Ability 68 7,72 1,244

IN1 - Integrity 68 7,44 0,920

BE1 - Benevolence 68 7,32 1,126

Table 4. Descriptive Statistics Mann-Whitney U-Test

When having a first look at the group statistics, one conditions immediately stands out.

Being that all means scored around 7,5 out of 10. This could be the result of a central tendency bias, meaning that respondents failed to differentiate enough between their answers (Oxford Reference, n.d.).

(32)

When assessing the results of the Mann-Whitney U-Test with an alpha-level of 0.05 there is no significant difference in neither one of the dimensions:

Mann-Whitney U-Test Asymptotic Significance (2-tailed) AB1 – Ability 0,826

IN1 – Integrity 0,771 BE1 – Benevolence 0,085

Table 5. Mann-Whitney U-Test

0,826 > 0,05, so there is no significant difference between respondents that were in the webcare group and respondents that were in the traditional communication group when looking at the trust dimension ability

0,771 > 0,05, so there is no significant difference between respondents that were in the webcare group and respondents that were in the traditional communication group when looking at the trust dimension integrity

0,085 > 0,05, so there is no significant difference between respondents that were in the webcare group and respondents that were in the traditional communication group when looking at the trust dimension benevolence

This means that since there is no significant difference, there sure is no significant prove of higher scoring on one of the dimensions. Because of this Hypothesis 1 and Hypothesis 2 have to be rejected.

(33)

6. Concluding Discussion

6.1 Conclusions

In this last part of this thesis the notion is to to draw conclusions from the analysis.

This is done by wrapping up the findings from the theoretical analysis, and ending with incorparating the results form the survey.

6.1.1 What is trust related to commerce?

Trust is basically the perceived integrity, benevolence, and ability that is perceived by one of the parties that are in a relationship. Such a relationship could be a personal relationship, but many times it is a formal relationship. Examples of these formal relationships are partners in a joint venture, or partners in a contract in commerce. In commerce trust can be build mainly by face-to-face, repeated, personal contact. Such contact is mainly important for the perceived integrity and benevolence of the other party. Also the familiarity with a company plays an important role. Companies that are very old and have a good reputation will keep more customers when there is a

situation when a customer and an employee can not get along, compared to this situation with a sole proprietor. This is more connected to ability, if the other party is able to fulfil his or her obligations. Lastly, people also have a trusting disposition. This means that some people will quicker trust someone than others. The most strong

answer to this first sub-question is: Trust is the keystone of ecommere. Without trust there would be no commerce.

(34)

6.1.2 What methods are there to communicate in an e-commerce setting, and how does this affect trust?

Since the start of e-commere is was apparent that there would be a need to have communicate with (potential) customers, because it was possible that a customer was located at a corner of the country or even at a random place somewhere in the world.

In the early days of e-commerce this communication was limited to communication that was already there to support face-to-face communication in commerce.

Communication via telephone, fax, e-mail or even by sending a letter was the norm.

This communication is not very personal, and misunderstandings could arise easily because of the lack of non-verbal communication. This was not a big problem in commerce because it was to support the in-person communication that was building trust. With the disappearance of that main communication method it turned out that people had to rely on the ability of the other party to fulfil an obligation. To cope with this situations companies started to build a social presence. They did so by using

techniques and tricks to make people feel that they were communicating with a person, or to enhance communication in a way that it felt more personal. Now there are new technologies to really communicate in a personal way. Social media is the biggest example of this. On this social media it is possible to communicate on a much more personal way. Adding pictures, linking to a movie, or tagging other people is done in mere seconds. Using the names of employees, combined with a funny picture is very normal, and communication is very down-to-earth and informal.

(35)

6.1.3 Can new communication methods like webcare be viewed as an expansion of theory on social presence in e-commerce and therefore update existing theory?

New communication methods are better and stronger than social presence, but have the same charasteristics. Therefore these channels are a step further than social presence and could be named social communication. This is basically the effect that social presence intended to provide on a superficial basis, completely fitted to replace interpersonal contact. This means that the question to the main research question is answered positively. To further clarify this new insight, the newly developed model and definition come in play.

Social Communication Communication between two or more physically divided participants that is rich, synchronous, and is substituting interpersonal contact, via new communication methods that is building trust in a B2C e-commerce environment.

Model on Social Presence and Social Communication in e-Commerce (as presented in chapter 4) e-Trust

Integrity Benevolence

Ability

Control Variables

Trusting

disposition Familiarity

Purchase Intentions Social Presence

Social Communication

(36)

6.2 Conclusions from the Survey

Based on the answer of my research question I did expect a significance difference between the two groups that were exposed to either the traditional way of

communicating and the new way of communicating via webcare. This significance, however, is not displayed. The most obvious reason for this is the fact that this survey is only covering a specific situation with two specific communication methods.

Nonetheless, a weak link between webcare and a different score on perceived benevolence is present. With an alpha-level of 0.10, this link would be proved significantly. Because of the scientific nature of this paper, an alpha-level of 0.10 would not be appropriate, but it gives an indication that there is indeed a link, but it was not adequately demonstrated in this research. When looking at the group statistics of an Independent Samples T-Test it appears that the difference on perceived

benevolence is indeed a higher score:

Benevolence statistics

Group N Mean Std. Deviation

BE1 – Benevolence Webcare

Traditional

68 35 33

7,32 7,57 7,06

1,126 0,979 1,223

Table 6. Group statistics Independent Samples T-Test and descriptive statistics Mann-Whitney U-Test for Benevolence

(37)

6.3 Reliability & Validity

When doing the theoretical analysis I attained to only use the most reliable sources. I did this by only using academic articles that used reliable and traceable sources. Also, only articles that were cited in other academic papers were used. In the case of articles from mainstream media or on websites, only publications that used sources and were from a reputable source were used. Only direct sources were used, except for one case where this original source was not accessable. In this case the indirect source was properly quoted. Another way to check reliability was to make sure that articles with contradictory information was also added to the report. Not all information that supported the sources used in this analyis were used, only the most prominent ones.

To check the validity of this thesis and the sources the following question was answerd for every source used: “Is the source/report answering the research question or sub- questions and/or does is help to answer this question, and is it in the scope of the research?” For every source, and search for articles and sources this question was in the back of the mind. After reading an article the question was answered and when the answer was “yes” the article was used in the report. This research leads to answer the research question and sub-questions and draws conclusions out of all information presented and cited in this thesis.

6.3.1 Survey Validity & Reliability

Because this survey with controlled experiment looks at a very specific situation and uses only two communication methods, there are some problems with the external validity, or the extent to which the results from this survey could be used to

corroborate the broader conclusions for the theoretical analysis (Thomas, 2004). A way to count for this effect is to have several different communication methods, and different situations for the quasi-experiment. However, because of the expected (and in hindsight received) respondents only two treatment groups were formed. This also is a result of not having access to a company with real user data.

(38)

One of the respondents identified after the survey told that he has strong experiences and expectations regarding KLM, drawing on memories on flying with them.

Consequently, he had hardship incorporating the information embedded in the survey in his scores. In other words, he had problems with history (Campbell & Stanley, 1967). So, the second possible explanation of this results is the fact that KLM is so familiar so respondents, that they have difficulties using the information from the survey. Choosing another company that is known to respondents, but without very stark expectations or preconceptions should have been a better idea to counter this.

Another possible explanation could be a gap in generations. Meaning that the age difference of the respondents could be one of the reasons that the informal way of communicating on the webcare communication channel stirred controversy for some of the respondents. After hearing about this from a respondent identified after the survey, my hypothesis on this subject would be that a higher degree of informal communication would be regarded out of place by older generations. This is however just a hypothesis since there is no solid evidence to corroborate this hypothesis in the literature and the age of respondents was not asked. This was an oversight.

Lastly, central tendency bias (Oxford Reference, n.d.) could play a role in answering the questions about the trust dimensions. The choice for a scale of 1 (Completely Disagree) to 10 (Completely Agree) was because it resembles the regular scale of grading in The Netherlands, so the significance of a score and the ratio of other scores would be clear to every respondent. This large scale, and the “middle” (5) choice could, however, make that respondents were not able to differentiate their answers.

Changing the response scale to a smaller and odd scale could helped to overcome this bias (Oxford Reference, n.d.). Unfortunately this is not possible after receiving the survey results.

(39)

Testing a part of the new “social communications” definiton should really be possible in future research when keeping in mind the limitations of this research and by

creating conditions to overcome this limitations by: (1) conducting research within a company among real (potential) customers by looking at much more datapoints regarding communication methods; (2) do exploratory research on the strength of familiarity with the company to be included in the research, before conducting the survey, to make sure enough safeguards are included to counter this effect; (3)

incorporate questions about the age of respondents, to be able to use this information to draw conclusions on a possible gap between generations; and (4) using smaller scales, preferably a scale from 1 (Completely disagree) to 4 (Completely Agree) to overcome central tendency bias.

6.4 Limitations & Further Research

As mentioned before, there is not much research done in this field, so sometimes it was neccesary to use information that is already a couple of years old. Normally this would not be a problem, but because of the rapid changing environment in this field, for some information there were indications that some facts could be changed.

Because these indications were not found in articles that met the requirements for validity and relability, this information was not used.

Referenties

GERELATEERDE DOCUMENTEN

clear limitation of this study and therefore we restrain ourselves from the discussion of the possible physiological models based on the selected markers or the attempts to

The influx of new research on Belgium and international law in the First World War era is therefore, to be applauded : Juliette Lafosse’s exploration of Belgian

HBSW08-II-S1 donkerbruin gracht Zand vermoedelijk Romeins HBSW08-II-S2 bleekgrijs gracht (?), vaag Zand vermoedelijk Romeins HBSW08-II-S3 donkerbruin paalspoor/kuil (?)

Through fieldwork and a detailed analysis of texts, the project aims to develop an understanding of how, and with what degree of success a form of Muslim

The project investigates innovative network architecture and trans- mission solutions for access and in-building networks based on many different sorts of optical fibers

Thirdly, this study expected a positive moderating effect of interdependence on the relationship between relational trust and relationship performance, based on

Mediator relationship: To test if relational trust mediates the relationship between the significant relational norms continuity expectation and information exchange

Over the whole period 47 publications are devoted to budgeting which are distributed in the research areas as follows; 20 for the cross-sectional research area, 6 for the