UNIVERSITEIT VAN AMSTERDAM
Expression of electronic word-of-mouth (eWOM)
through Facebook and opinion platforms: The role of
eWOM motivation on consumers’ channel choice
Laura Dirksz - 5754461
8/15/2014Final submission Supervisor: Dr. Alfred Zerres Second supervisor: Joris Demmers, MSc
Master Business Studies Marketing track
Laura Dirksz, 5754461 3 Table of contents Abstract ... 5 1. Introduction ... 6 2. Literature review ... 10 2.1 Word of Mouth ... 10 2.2 Electronic Word-of-Mouth ... 11
2.3 Social networking Sites and eWOM ... 12
2.3.1 Facebook and eWOM ... 13
2.3.2 Conceptualizing eWOM ... 14
2.4 Consumer motives for engaging in WOM and eWOM ... 14
3. eWOM motives and eWOM channel choice ... 17
3.1 Uses and gratifications theory ... 17
3.2 Message involvement ... 18
3.3 Product involvement ... 19
3.4 Anxiety reduction ... 20
3.5 Social benefits ... 20
3.6 Helping the company ... 21
3.7 Advice seeking ... 22
3.8 Vengeance ... 22
3.9 Concern for others ... 23
3.10 Self-enhancement ... 24 3.11 Economic incentives ... 24 4. Methods... 25 4.1 Sample ... 26 4.2 Measures... 26 Dependant variable ... 26 Independent variables ... 27 4.4 Procedure ... 31 5. Results ... 32
Laura Dirksz, 5754461 4
5.2 Comparison of motivations between eWOM on Facebook and on opinion platforms ... 37
6. Discussion and conclusions ... 40
References ... 44
Appendix A – WOM and eWOM motives from previous research ... 52
Appendix B– Cover letter ... 53
Abstract
This study investigates the effect of electronic word of mouth (eWOM) motives on the likelihood of engaging in eWOM on Facebook or on a consumer opinion platform. Data was collected through an online survey and consisted of 96 respondents who engaged in eWOM on Facebook and 106 respondents who engaged in eWOM on opinion platforms. The results did not find a relationship between the eWOM motives and the number of posts on either Facebook or opinion platform. This was in contrary to prior research on eWOM through opinion platforms. However, a comparison of the eWOM motives for the posts on each channel did reveal that in general, eWOM Facebook was more motivated by message involvement, product involvement, self enhancement, social benefits and anxiety reduction than eWOM on opinion platforms. On the other hand, eWOM on opinion platforms was more motivated by a desire to get vengeance on the company, to help the company, to warn other consumers from a bad product or company and by a desire to help other consumers make the right purchase decision.
Laura Dirksz, 5754461 6
1. Introduction
People don’t trust brands, people trust people. Consumers frequently talk about their experiences and share their opinions with others. They might discuss in example a restaurant recently visited, their vacation or a purchased product. This type of interpersonal communication where consumption-related information is shared, known as word-of-mouth (WOM)
communication, has long been acknowledged as influential in a majority of consumer purchase decisions (Dichter, 1966). WOM is considered one of the most influential sources for
information for consumers (Buttle, 1998) and is seen as more consumer-oriented and reliable than market-generated information (Arndt, 1967).
In recent years, the subject of WOM has had increased academic and managerial attention due to the surge of WOM information shared in technology-enabled contexts (Dellarocas, 2003). The Internet has turned into one of the biggest interpersonal communication mediums for
consumers and has provided consumers with different channels through which they can express their experiences and exchange information concerning products and services to fellow
consumers (Trusov, Bucklin, & Pauwels, 2008). Every day we are confronted with stories about people’s experiences. In the past, a person had to wait to come across someone to share their experience. Nowadays, the Internet provides a way for people to express their opinions and share their experiences that is no longer dependent on an immediate receptor. The interpersonal
communication about a product, service or company in an online environment is known as electronic word-of-mouth (eWOM). eWOM has been defined as any positive or negative statement about a product or company made by former, actual or potential customers, which is made available through the internet to a multitude of institutions and people (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). Similar to WOM, research shows that communication through eWOM is considered more relevant and credible than information on the internet that is
Laura Dirksz, 5754461 7 marketer-generated (Bickart & Schindler, 2001). Furthermore, eWOM is also found to be helpful in reducing perceived risk in e-commerce (Ha, 2002).
Accordingly, it can be stated that traditional WOM has been extended to electronic media such as blogs, newsgroups, online discussion forums, review sites and social networking sites (Cheung & Lee, 2012; Chu & Kim, 2011). Particularly social networking sites (SNS) represent an ideal tool for eWOM, because through SNS consumers can create and disseminate brand or product related information within their online social network (Chu & Kim, 2011), thereby mimicking traditional WOM with offline social networks. SNS have been coined as one of the fastest growing arenas of the internet (Trusov et al., 2008) and includes for example Twitter and Facebook, which have attracted millions of users. In September 2013, Facebook had a reported 1.19 billion monthly active users, and Twitter had a reported 250 million active users
(Marketingfacts, 2013). The networking, collaboration and conversation found in SNS has provided great opportunities for consumers to engage in eWOM (Chu & Kim, 2011). As the contacts in SNS are entrenched in the personal network of the consumers, they are apt to be seen as more trustworthy and credible than unknown sources that might have hidden interests or marketers (Chu & Kim, 2011). This makes SNS a particularly interesting eWOM channel to study, especially why consumers engage in eWOM specifically on SNS.
Researchers acknowledge that by participating in eWOM, consumers acquire both economic and social value (Balasubramanian and Mahajan, 2001) and consequently might have different motivations for generating eWOM (Hennig-Thurau et al., 2004). This thesis proposes that this presumption can be extended to the channel through which the consumer chooses to engage in eWOM. In the online environment, a consumer has a variety of channels to choose from through which they can express their opinions, and thus engage in eWOM. The consumer’s
Laura Dirksz, 5754461 8 motive for eWOM might influence the channel choice for eWOM expression. Literature has focused on uncovering the motivational drivers for eWOM (Hennig-Thurau et al., 2004) but have not explored their implications for the consumer’s channel of expression choice.
As previously proposed, the consumer’s motive for eWOM might influence the channel choice for eWOM expression since each channel possesses a unique set of characteristics. In example, the study by Hennig-Thurau et al. (2004), one of the most cited articles on eWOM motivational drivers, is based on data from consumers who engage in eWOM on consumer opinion platforms. However, there are some crucial differences between the channel consumer opinion platforms and the channel SNS. These differences in characteristics include i.e. eWOM sender identity (anonymous - non-anonymous), audience breadth (public – closed social
network) and audience identity (unknown – personal social network). Some channel
characteristics might be better aligned with the consumer’s motivation for eWOM. In example, if the consumer is motivated by vengeance, the consumer might chose to express themselves on the channel that reaches the most other consumers whilst he/she can remain anonymous. In such a case, a consumer could choose to express themselves on an opinion platform which gives the most fulfilment by possibly affecting the most other consumers thereby having the biggest effect on the company. At the same time, choosing to post anonymously on an opinion platform might protect the image the consumer might want to preserve on a SNS if the comment is particularly negative. Therefore, different motivating factors driving eWOM might drive expression through a particular eWOM channel rather than another. More specifically, some motivational factors might drive expression through opinion platforms whilst other motivational factors might drive expression through SNS.
Laura Dirksz, 5754461 9 It is important to investigate whether the motives found by Hennig-Thurau et al. (2004) can be extended to other eWOM channels with different characteristics and if there are other factors driving eWOM behaviour through other channels than those proposed by Hennig-Thurau et al. (2004). Specifically, the focus of this thesis will be on uncovering possible differences in the motivational drivers that lead to eWOM expression within SNS in contrast with opinion platforms. Due to their differentiating characteristics, this thesis proposes that SNS can be distinguished from opinion platforms and that the motives driving eWOM have implications for the choice of eWOM through one of these channels rather than the other.
This research will contribute to the academic field by adding to the understanding of what motivates consumers to articulate themselves about services and products in the relatively novel area of SNS. Furthermore, it will also give insight into the influence of eWOM motives on eWOM expression. Specifically, the influence of eWOM motives on the channel through which eWOM is expressed. From a managerial perspective, further understanding of consumer motives for eWOM through different channels can be used as a foundation for further development of ways to communicate and influence consumers to engage in eWOM. If different motives underlie eWOM in SNS compared to eWOM through opinion platforms, one could argue that different approaches are necessary to stimulate eWOM through each channel.
The first section of this proposal consists of a literature review that examines a number of topics in order to understand consumers’ motivation for engaging in eWOM. First the literature on traditional WOM is discussed, followed by research on eWOM. The next section introduces SNS as a mean for consumer engagement in eWOM and explains the differences between this platform and opinion platforms. Then, the possible implications of these differences on the
Laura Dirksz, 5754461 10 motives for eWOM are discussed. Finally, the research method is proposed and the schedule for the thesis is presented.
2. Literature review
2.1 Word of Mouth
Word-of-mouth (WOM) is a communication concept which has been generally referred to as an informal method of communication between people about usage, ownership or
characteristics of product and services (Dichter, 1966; East, Hammond, & Wright, 2007;
Lazarsfeld & Katz, 1955; Westbrook, 1987; Wetzer, Zeelenberg, & Pieters, 2007). In addition, it has been stated that WOM is transmitted through oral face-to-face communication (Arndt, 1967) and that it is neither paid nor sponsored and thus independent from the company (Arndt, 1967; Mullen & Johnson, 1990; Silverman, 1997). Thus, the main principle of WOM is that
information about brands, companies or products can be shared from one consumer to another. This information is communicated in and informal and face-to-face setting and is not
commercially driven.
Throughout the years, research has been conducted on the impact of WOM on consumer attitudes and behaviour. Scholars posit that WOM has an effect on most consumer purchase decisions (Bansal & Voyer, 2000; Brooks, 1957; Dichter, 1966; Richins & Root-Shaffer, 1988) and that WOM can be more influential in consumer decision making than information that is marketer controlled (Buttle, 1998; Day, 1971; Metzger, Flanagin, & Medders, 2010; Sheth, 1971). When considering a purchase, consumers rely on brand and product information through WOM because it is created and delivered by a more trustworthy information source than
Laura Dirksz, 5754461 11 2006). Furthermore, research also shows that WOM can influence both long term and short term product judgments of consumers (Bone, 1995).
2.2 Electronic Word-of-Mouth
The emergence of internet based media has facilitated new communication environments which has allowed WOM to take new forms. For consumers, the Internet has expanded the possibility of attaining unbiased opinions on products and services (Hennig-Thurau et al., 2004) and has enhanced the chance to easily disseminate these opinions to a multitude of people (Dellarocas, 2003; Hennig-Thurau et al., 2004). WOM expressed in online environments is known as electronic Word-of-Mouth (eWOM) (Hennig-Thurau et al., 2004). Hennig-Thurau et al., (2004) define eWOM as any positive or negative statement about a product or company made by former, actual or potential customers, which is made available through the internet to a multitude of institutions and people. eWOM transpires on a broad scope of online channels, ranging from emails, blogs, virtual communities, consumer review websites and forums and social networking sites (Cheung & Lee, 2012; Dwyer, 2007; Hung & Li, 2007; Phelps et al., 2004; Thorson & Rodgers, 2006). eWOM is one of the important features of the Internet,
because it enables consumers to share their experiences and exchange information through many different channels (Trusov et al., 2008). The opportunity to share is no longer dependent on an immediate receiver.
The Internet has provided some unique characteristics to eWOM. First, the content of eWOM is in written format compared to verbal format in the case of traditional WOM, which makes the information transmitted easier to capture (Huang, Cai, Tsang, & Zhou, 2011). Because of eWOM’s written format and exposure to a multitude of people (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004), researchers have argued that eWOM diffuses faster than traditional
Laura Dirksz, 5754461 12 WOM (Cheung & Thadani, 2010; Prendergast, Ko, & Yuen, 2010). Additionally, most of the text based information communicated through eWOM is stored on the online platform for an indefinite period of time, making it more persistent and easily accessible than traditional WOM (Cheung & Thadani, 2010). Furthermore, the communication network of eWOM is larger than the network of traditional WOM. Therefore, consumers have multiple means of exchanging information (Jalilvand, Esfahani, & Samiei, 2011) making the reach of eWOM virtually
unlimited (Dellarocas, 2003). eWOM is also seen as an asynchronous process because the sender and the receiver are separated by both space and time (Steffes & Burgee, 2009). This is contrary to traditional WOM where the communication is face-to-face following a synchronous process where participants communicate concurrently. Additionally, eWOM is more likely to contain references to advertising than traditional WOM due to the ease of sharing links, information and pictures on the Internet (Keller & Fay, 2009).
2.3 Social networking Sites and eWOM
Social networking sites (SNS) allow users to create and maintain a network of friends for professional or social interaction thereby encouraging social interaction. In contrast to consumer opinion platforms platforms, SNS retain their users by relying on user-generated content and its core is based on personalized user profiles (Trusov, Bucklin, & Pauwels, 2008). Social
networking sites (SNS) are defined as computer mediated services that enables users to construct a personal user profile, to build and maintain network of friends for professional or social
purposes and to share information to their network (Ellison, 2007a; Trusov et al., 2008). Other scholars define SNS as web-based services that permits users to create a public or semi-public profile within a system, where they can build a user list with shared relationships and see the list of relationships of those people with other person within the system (Ellison, 2007b).
Laura Dirksz, 5754461 13 SNS lets the users reveal personal information such as feelings, thoughts, likes and
dislikes to a high degree (Kaplan & Haenlein, 2010) which suggests that through SNS users can project a desired image to others. Furthermore, the level of immediacy and intimacy of
communication within SNS suggest that through interaction with each other, users have influence to a certain degree on each other’s behaviour (Kaplan & Haenlein, 2010).Thus, SNS represent good method for distributing eWOM because it allows consumers to freely create and disseminate information about products and brands to their social networks (Vollmer, 2008).
This thesis proposes that SNS can be distinguished from other online platforms, specifically consumer opinion platforms, in a few key ways. First, in SNS users have the possibility of disclosing information about themselves, in other words, people are not
anonymous. Furthermore, previous research suggest that SNS function as a way to bring offline social networks to an online environment (Ellison, 2007a) and SNS are more used to retain relationships than to meet new people (Ellison, Steinfield, & Lampe, 2007). Finally, in SNS, the senders of eWOM are familiar with the people they are talking to because they can selectively add other users to their network. In the SNS context, the eWOM sender can choose who they expose their information to, in comparison to blogs or other platforms where the information published is public. The following section will elaborate on the choice of Facebook as SNS for this thesis.
2.3.1 Facebook and eWOM
Facebook is the most popular SNS with a reported 1.19 billion monthly active users (Marketingfacts, 2013). On Facebook users create a profile using their real identity and share information primarily with individuals who the user has approved as friends. The users are offered a variety of services that aide the sharing of information such as posting status updates,
Laura Dirksz, 5754461 14 uploading pictures, “like” pages and check-in on locations. These tools facilitate the distribution of eWOM on the newsfeed of all of the user’s Facebook friends. Facebook is an ideal channel for eWOM due to its built in information sharing tools and private social environment. Additionally, every time a user posts, likes, shares or checks-in on Facebook it is automatically displayed on the newsfeed of the user’s network of friends (Coulter & Roggeveen, 2012) thereby
automatically spreading the users potential eWOM.
2.3.2 Conceptualizing eWOM
The concept of traditional WOM, as well as of eWOM have been previously defined. But scholars have argued that SNS, specifically Facebook, present unique opportunities for eWOM. Shu-Chuan & Yoojung (2011) argue that any voluntary display of brand preference in the online context can stimulate eWOM communication. This implies that even “likes”, “shares” and “check-ins” can be considered as eWOM. However, in this study eWOM on Facebook will be limited to statements that are originally generated by the user. This ensures that the same concept is measured in both eWOM channels. Therefore, the definition for eWOM will remain as
proposed by Hennig-Thurau et al. (2004) as any positive or negative statement about a product or company made by former, actual or potential customers, which is made available through the internet to a multitude of institutions and people.
2.4 Consumer motives for engaging in WOM and eWOM
The amount of published studies explicitly addressing the underlying motives of WOM and eWOM behaviour is limited. Among others, Dellarocas (2006) and Tong, Wang and Teo (2007) indicate that the most dominant publications in this area are published by Dichter (1966), Engel et al. (1993), Sundaram et al. (1998) and more recently Hennig-Thurau et al. (2004). The next paragraphs will give insight into the findings of each of these publications.
Laura Dirksz, 5754461 15 Dichter (1966) published one of the first research on the motives for consumer
engagement in WOM. In-depth interviews were conducted where 225 consumers where inquired to recall their WOM experience. The findings conclude that the motivations for consumers to speak positively about a service or product is to gain satisfaction. These rewards are always purely psychological and not material (Dichter, 1966). The motivations for positive WOM are grouped in four categories; self-involvement, message-involvement, other-involvement and product involvement (Dichter, 1966). This publication is considered one of the most important studies regarding motives for WOM and is still used as a foundation within the field (see (Hennig-Thurau et al., 2004; Sundaram, Mitra, & Webster, 1998).
Engel et al. (1993) renamed these factors in their literature review on WOM in order to highlight the underlying meaning of these factors. Furthermore, the authors confirm the original findings by Dichter (1966) and explain the motives in a more modern and comprehensive way. They renamed the factors identified by Dichter (1966) to self-enhancement, message intrigue, concern for others and involvement (Engel, Blackwell, & Miniard, 1993).
Sundaram et al. (1998) published the most comprehensive study on motives for WOM communications up to date. The authors conducted critical incident interviews with 390
consumers and analysed using a content analysis procedure. Their results suggest eight motives for consumer engagement in WOM; product involvement, self-enhancement and altruistic motives underlie positive WOM and anxiety reduction, vengeance, advice seeking and also altruistic motives underlie negative WOM (Sundaram et al., 1998). The findings of Sundaram et al. (1998) correspond with three of the factors identified by Dichter (1966) and confirmed by Engel et al. (1993) except for message intrigue.
Laura Dirksz, 5754461 16 Hennig-Thureau et al. (2004) conducted one of the first studies on the motivational factors for eWOM. Their study examined the motives for consumer engagement in eWOM on online consumer opinion platforms. In order to capture the particular features of the online context, this study builds on the motives for WOM identified in previous research and also integrates social and economic activity from a model by Balasubramanian and Mahajan (2001). The authors propose 11 possible motives for eWOM and tested with motive structure with a principal component analysis. The results indicate eight motivational factors for eWOM; venting negative feelings, helping the company and platform assistance which were not significant. The other five factors; social benefits, concern for other consumers, self enhancement and economic incentives are indicated as the primary factors (Hennig-Thurau et al., 2004). Thus the authors are the first to confirm economic incentives and social benefits as motivational drivers of eWOM. The motives identified in the previous section are summarized in Appendix A.
Research on motivations for engaging in eWOM has focused on motives for eWOM within consumer opinion platforms (Hennig-Thurau et al., 2004). However, there is a lack of research in this field specifically for SNS. This thesis will address this gap by researching the motives for engaging in eWOM communications specifically through the channel of Facebook and comparing this with the motives for eWOM though the channel of opinion platforms. Particularly, this thesis will investigate whether different consumer eWOM motives lead to expression through Facebook rather than opinion platforms and vice versa.
The research question is;
What are the differences in consumer motives for engagement in electronic word-of-mouth within Facebook compared to consumer opinion platforms?
Laura Dirksz, 5754461 17 The following section will give an in depth description of a few of the consumer motives
identified in the WOM and eWOM literature and will hypothesize on the implications of these motives on the choice of expression through Facebook or through an opinion platform.
3. eWOM motives and eWOM channel choice
3.1 Uses and gratifications theory
What motivates consumers to engage in eWOM through a particular channel can possibly be explained by the uses and gratifications theory (UGT) proposed by Katz, Blumber and
Gurevitch (1973). The main question UGT scholars tries to answer is; why do individuals participate in one specific type of mediated communication compared to another, and what are the gratifications that they receive from it? The three principles underlying the UGT are: an individual’s behaviour is goal directed; the individual is an active media user; and the individual is aware of his/her needs and chooses media in order to gratify these needs (Blumer, 1979).
UGT has been applied to traditional media and recently also to understand internet usage motives (Stafford, Stafford & Schkade, 2004). The UGT examines how individuals use mass media to satisfy their needs, the underlying motives that drive this and the consequences of this media usage. Engaging eWOM can be perceived as an activity that fulfils a particular need of the consumer, which is preceded by a particular motive. UGT distinguishes four motives that drive media usage. These motives are; cognitive, personal, hedonic and social motives. Cognitive motives relate to the acquisition of information and knowledge. An individual with personal motives is seeking to increase their status, credibility, confidence and stability. Hedonic motives relate to an individual’s desire for relaxation, pleasant experiences and escape from daily life.
Laura Dirksz, 5754461 18 And an individual searching for social interaction or feedback from others is driven by social motives (Mendes-Filho & Tan, 2009; Rubin, 2009).
As such, individuals with different goals select and use media based on the needs they expect to satisfy through use of that media (Katz, Gurevith, & Haas, 1973; Perse & Courtright, 1993).Thus it can be argued, that a consumer chooses a channel to express eWOM based on the needs they desire to fulfil by expressing themselves though that particular channel. The
following section builds upon this assumption to hypothesize on the preferred eWOM channel for each eWOM motive.
3.2 Message involvement
The message of the ad, rather than the product itself, can serve as a motive for engaging in WOM (Dichter, 1966; Engel et al., 1993). The WOM is driven by the entertainment value of the ad or its originality and is not necessarily based on the speakers experience with the product (Dichter, 1966). The quotation of advertising slogans or judgements of the effectiveness of the advertisement can become the topic of conversation. Thus the focus of the WOM is not the product, but the message.
In an online context, specifically on Facebook, the advertisement itself can be shared and accompanied by a statement by the consumer regarding the ad. This can also be considered as a eWOM, which is motivated by message intrigue. However, in the context of opinion platforms, it is less likely that message intrigue motivates eWOM since expressions on opinion platforms is mostly about the (usage of the) product itself. Quan-Haase and Young (2010) found
entertainment value key gratification obtained by the use of Facebook. Therefore, it is hypothesized that;
Laura Dirksz, 5754461 19 H1: Consumers motivated by message involvement are more likely to engage in electronic Word of Mouth on Facebook rather than on an opinion platform.
3.3 Product involvement
Product involvement is one of the oldest acknowledged motives for engaging in
traditional WOM, originally proposed by Dichter (1966). This factor regards situations where the consumption experience of a product that is perceived as highly important or relevant leads to strong feelings of excitement that causes tension within the consumer. The consumer feels a need to relieve this tension by expressing their enthusiasm about the product (Dichter, 1966; Engel et al., 1993; Sundaram et al., 1998). WOM presents a way for the consumer to vent positive feelings and thereby relieve and dispose of the pleasure obtained. This motive has not yet been explored as a motive for eWOM.
Studies, studies on the motives for so called “homepages”, or user profiles, suggest that self-expression and impression management are the most important motives for hosting
homepages (Papacharissi, 2002; Jung, Youn,& McClung, 2007). Consumers can manage their self-presentation more strategically in their online user profiles than in face to face situations (Ellison, Heino, & Gibbs, 2006). Since a personal interest in the product also underlies the product involvement motive (Sundaram et al., 1998), expression through Facebook can signal preference or can add to the desired self-image of the consumer on Facebook. Accordingly, the hypothesis is;
H2: Consumers motivated by product involvement are more likely to engage in electronic Word of Mouth on Facebook rather than on an opinion platform.
Laura Dirksz, 5754461 20 3.4 Anxiety reduction
A negative consumption experience can lead to feelings of anxiety, anger and frustration. In such cases, eWOM can be a way for consumers to vent their negative feelings (Sundaram et al., 1998; Hennig-Thurau et al., 2004). The consumer hopes that sharing their negative
experience will help reduce the anxiety within. This need is better fulfilled by a message that is emotional and displays the consumer’s feelings rather than a logical message (Wetzer et al., 2007; McColl-Kennedy et al., 2009). Recent experiments have shown that individuals turn to Facebook when feeling distressed (Toma and Hancock, 2011) and that that college students use Facebook to receive and seek social support when feeling upset (Park, Kee, & Valenzuela, 2009; Wright, Craig, Cunningham, & Igiel, 2007). Therefore;
H3: Consumers motivated by anxiety reduction are more likely to engage in electronic Word of Mouth on Facebook rather than on an opinion platform.
3.5 Social benefits
Consumers can also engage in eWOM for social integration purposes. Engaging in eWOM serves as a way for the consumer to actively participate and to become part of the online community. Hennig-Thurau et al., (2004) derived this factor from Balasurbramanian and
Mahajan’s (2001) focus-related utility, which relates to the utility the consumer receives by adding value to the community through their own contribution. This motive can manifest in example by a consumer initiating eWOM in order to signify their community presence. As described by Ho and Dempsey (2010), this motive relates to a desire to be part of a group by communicating and participating in that group. Since social interaction is an integral part of Facebook, one could propose that a consumer can gain social benefits by engaging in eWOM through this channel. eWOM can illicit comments and thus interaction with other users and can
Laura Dirksz, 5754461 21 possibly serve to strengthen social ties. By posting, the consumer is actively contributing to Facebook and thereby adds value to this community. On the other hand, opinion platforms do not tend to focus on social interaction between the users. Therefore;
H4: Users motivated by social benefits are more likely to engage in eWOM on Facebook rather than on an opinion platform.
3.6 Helping the company
A satisfactory experience with a product or service can lead to a subsequent desire of the consumer to help the company (Sundaram et al., 1998). The company is considered as a social institution that is worthy of support, in the form of eWOM. The motivation for eWOM in such a case is rooted in the consumer’s desire to “give something in return” to the company for a good experience. This motivation has been found as a motivation to engage in eWOM in numerous studies (Bronner & de Hoog, 2011; Jeong & Jang, 2011; Yap, Soetarto & Sweeney, 2013). The consumer hopes that spreading a good word about the company will help the company become or remain successful.
Deriving from this logic, the more people read about the positive experience, the more potential (new) customers the company will receive that can help the company. Therefore, it can be argued that the consumer fulfils his/her role the best by reaching the biggest possible
audience. Given that public consumer opinion platforms reach bigger crowds of people than the average private personal Facebook page, the hypothesis is:
H5: Consumers motivated by helping the company are more likely to engage in electronic Word of Mouth on an opinion platform rather than on Facebook.
Laura Dirksz, 5754461 22 3.7 Advice seeking
Some consumers share a dissatisfying consumption experience with others in the hope of getting post-purchase help. This motive is known as advice seeking, and is meant to prompt information or advice from other consumers to aid the understanding and/or use of a service or product (Hennig-Thurau et al., 2004; Sundaram et al., 1998).
On an opinion platform the consumer has a higher chance of receiving helpful advice from other consumers since it reaches a broader audience and thereby makes it more likely that a consumer with experience with the product/service is active on that particular platform. They might possess specific knowledge about the product/service that can help the consumer to resolve their post-purchase problem. In comparison, on Facebook the number of users within the consumer’s social network that have experience with the product/service and can give accurate advice might be limited. Therefore;
H6: Consumers seeking post-purchase advice are more likely to engage in eWOM on an opinion platform rather than on Facebook.
3.8 Vengeance
Another consequence of a negative consumption experience is a desire to avenge the company (Sundaram et al., 1998). The consumer feels that the company should not be allowed to operate and thus wants to deter other consumers from patronizing the company and thereby punishing the company (Ward and Ostrom, 2006). This motive has been found in traditional WOM (Sundaram et al., 1998) but has been grouped with the anxiety reduction motive into the factor “venting negative feelings” in research on motives driving eWOM (Hennig-Thurau et al., 2004). However, it can be argued that a consumer motivated by revenge, will choose the channel that can reach the broadest audience and thus better suit the consumers goal of harming the
Laura Dirksz, 5754461 23 company. Therefore, this motive is considered separately from “anxiety” reduction. The
consumer explicitly wants to advice other consumers not to patronize the company, and therefore;
H7: Consumers motivated by revenge against the company are more likely to engage in electronic Word of Mouth on an opinion platform rather than on Facebook.
3.9 Concern for others
Either negative or positive consumer experience can lead to initiation of communication about a product/service. This communication is motivated by a desire to prevent other consumers from having a negative experience or to help other consumers with their buying decision
(Dichter, 1966; Engel et al., 1993). In both cases, the consumer is motivated to engage in eWOM to help other consumers. This motive is related to the concept of altruism, which Sundaram et al. (1998) defines as doing something for other people without expecting a reward back. In order to satisfy this need, it is essential that the consumer believes that others will read and actually benefit from the consumer’s post. Therefore, the audience plays a crucial role in the satisfaction of consumers driven by this motivation. The consumer can reach the broadest audience by
posting on an opinion platform which is public, compared to their personal Facebook page which is usually limited to their “friends” on the network. Therefore, it is likely that the consumer’s post will have a bigger impact on an opinion platform because it is presumably viewed by a larger audience. Thus;
H8: Users motivated by concern for other consumers are more likely to engage in eWOM on an opinion platform rather than on Facebook.
Laura Dirksz, 5754461 24 3.10 Self-enhancement
Self-enhancement reflects the need of the consumer to share their consumption
experience in order to present themselves as intelligent shoppers, to seek appreciation, to show connoisseurship and to seek recognition from others (Sundaram et al, 1998; Hennig-Thurau et al., 2004). This motivation reflects a consumer’s desire to augment their sense of self, thus to improve how others perceive them or how they see themselves. On Facebook, this recognition can be obtained by “likes”. But opinion platforms can offer the feature that readers can assess the “helpfulness” of the post and the posters can gain status as “trusted/expert reviewer”. A
consumer with motivated to enhance their sense of self is more likely to achieve this when they receive recognition from those who find the subject relevant, and thus are aware of the value of the eWOM post. Thus;
H9: Users motivated by self-enhancement are more likely to engage in eWOM on an opinion platform rather than on Facebook.
3.11 Economic incentives
One of the most important drivers of human behaviour are economic incentives, which are perceived by the receiver as a sign of gratitude for his/her contribution and efforts (Lawler, 1984). Hennig-Thurau et al., (2004) proposed economic incentives as a factor driving eWOM on opinion platforms. Opinion platform operators can apply this by offering remuneration to users in order to stimulate posts. It is important to note that opinion platforms do not allow users to post about a product or company if they receive an incentive from the company. So through this channel, economic incentives only apply to situations where the incentive is given by the
platform operators and not when given by the company. The opinion platforms found that offer economic incentives to its users are Ciao.com and Couverts.nl in the form of “Couvert Coins”
Laura Dirksz, 5754461 25 that can be exchanged for prizes. On the other hand, on Facebook companies can stimulate consumers to post on Facebook by offering prizes to those who post about the company or product. In such cases, the incentive comes specifically from the company. However, on Facebook the consumer has a chance of gaining an economic incentive, since most of the campaigns for prizes are raffles where other users also participate to win. In contrast, when posting on an opinion platform the incentive is guaranteed. Therefore, if the consumer’s goal is to receive an incentive for engaging in eWOM, posting on an opinion platform is a way to assure that tis goal is achieved. Building hereupon, it can be hypothesized that:
H10: Consumers motivated by economic incentives are more likely to engage in electronic Word of Mouth on an opinion platform rather than on Facebook.
4. Methods
The objective of this study was to explore the motives that lead to eWOM communication and to investigate the relationship between these motives and the choice of eWOM channel for expression. This study investigated the application of a previously validated theory in a new context, specifically the motives for engaging in eWOM communications on Facebook. Factors that have been previously validated as motives for WOM and eWOM communication in other settings served as a foundation for this study. These included findings from studies on motives for WOM (Dichter, 1966; Engel et al., 1993; Sundaram et al., 1998) and for eWOM in the setting of consumer opinion platforms (Hennig-Thurau et al., 2004). The following section outlines the research methodology for this thesis. The goal of this research is to analyse whether motivational factors influence the channel choice for eWOM engagement. A quantitative approach is applied to analyse the research questions. First, the sample and choice of research
Laura Dirksz, 5754461 26 methods and its characteristics are briefly explained. The next section addresses the chosen research instruments and their reliability. Finally, the procedure used for the collection of the data is presented.
4.1 Sample
The target population of this study was comprised of consumers who engage in eWOM on Facebook and/or on consumer opinion platforms. There were no demographic limitations that applied to this study. Convenience sampling was used because of time and resource constraints. The sample of this study comprised of a total of 233 respondents, all of which completed the online survey (N = 233). The respondents’ self-reported age ranged from 18 to 28 years, with an average respondent age of 23 years (SD = 2.46). The sample consisted of 125 female respondents (53.7 per cent) and 108 male respondents (46.4 per cent). Participation was voluntary and the respondents did not receive any incentive to complete the survey.
4.2 Measures
Dependant variable
eWOM communication. Hennig-Thurau et al. (2004) conceptualized eWOM behaviour as
the frequency of consumer visits to the opinion platforms and as the number of comments the consumer had written on the opinion platforms. However, it can be argued that in the case of Facebook, the number of visits cannot be considered as an indicator of eWOM since Facebook can also be used for other activities that cannot be classified as eWOM. Such other activities include in example playing online games and sharing pictures. Therefore, only the number of actual eWOM comments posted by the consumer was used to conceptualize eWOM. This allowed an accurate comparison of the Facebook posters and the Opinion platform posters since this conceptualization could be applied in both eWOM channels. Regarding Facebook,
Laura Dirksz, 5754461 27 respondents were asked to indicate how frequently they posted status updates on their profile that mentioned a product or company. For opinion platforms, respondents were asked to indicate their frequency of postings that mention a product or company on opinion platforms. Thus the dependent variable, eWOM communication, which was used to measure and operationalize eWOM, was the frequency of eWOM communication the consumer has posted. In order to include all frequency possibilities, the respondents could choose to indicate the frequency of posting either per day, per month, per week or per year. The frequency of posting was converted into the yearly number of posts per respondent for each respective channel. This yearly estimate of eWOM postings for either channel was used as a measure for eWOM communication.
Independent variables
The 10 proposed eWOM motives were measured by presenting the respondents with 28 statements on which they had to indicate their level of agreement on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree”. A 7-point Likert scale was chosen so a good discrimination among agreement levels can be secured, since many respondents apt to avert from selecting choices at the ends of the scale (Saunders, Lewis & Thornhill, 2009). Twenty-one items were retrieved from previously validated scales and seven items were formulated based on descriptions form previous research. The items corresponding to each scale were subjected to a reliability analysis to assess the internal consistency of the scales.
Message involvement. This construct assessed the degree in which the consumer engaged
in eWOM communication because of the entertainment value of the advertisement. Since there were no items for this motive that have been previously validated, four items were constructed based on the description of this motive by Dichter (1966) and by Engel et al. (1993). The items were worded as closely as possible to the items for the other constructs used that were derived from Hennig-Thurau et al. (2004). The items for this construct were: In general, I post online
Laura Dirksz, 5754461 28 about products/companies because “…I think an ad is entertaining”, “I am intrigued by an ad (it is original/different)”, “…I think the ad is clever” and “…I like to quote lines or slogans from the ad”. The reliability analysis revealed a Cronbach’s alpha of .98, therefore all four items were retained. The scores across the four items were averaged to give an overall score for message involvement. A higher score indicated more motivation for engaging in eWOM because of entertainment derived from posting about an advertisement rather than a product itself.
Product involvement. This construct assessed the consumer’s motivation to engage in
eWOM in order to relieve an overflow of positive emotions following a gratifying and
pleasurable purchase or usage experience with the product. It was measured by three items that were formulated based on the description of this motive by Dichter (1966), Engel et al. (1993) and Sundaram et al. (1994). The items presented to the respondents were: In general, I post online about products/companies because “…I am excited that I own the product/use the
service”, “…I have a great experience with the product/service”, “…I am really interested in the product/service”. Careful consideration was taken into the wording of these items in order to ensure similar sentence structure to those of Hennig-Thurau et al. (2004). The scale had a Cronbach’s alpha of .87. The overall score for product involvement was calculated by averaging the respondent’s ratings across the three items. Higher scores on this scale indicated a greater motivation to engage in eWOM because of excitement resulting from use of the product or company.
Social benefits. The social benefits construct assessed the consumer’s eWOM
communication motivated by social integration purposes and the social benefits the consumer expects to gain from their eWOM communication. The three items were adapted from Hennig-Thurau et al. (2004) and included; In general, I post online about products/companies because
Laura Dirksz, 5754461 29 “…I believe a chat with like-minded people is always a nice thing”, “…it is fun to communicate this way with other people” and “…I make/maintain friends this way”. In this sample the
Cronbach’s alpha was .83, thereby verifying the internal consistency of this scale.
Helping the company. Helping the company was measured by two items originally
developed by Hennig-Thurau et al. (2004). This construct assessed the consumer’s motivation to engage in eWOM because of a desire to help the company become or remain successful. The items were; In general, I post online about products/companies because “…I am so satisfied with a company and its product that I want to help the company to be successful” and “…in my own opinion, good companies should be supported”. The internal consistency of this scale was also validated for this sample (α = .84). Responses on the two items were averaged to an overall score of eWOM motivation caused by a desire to help the company.
Self-enhancement. This construct assessed the consumer’s motivation to engage in
eWOM in order to receive positive recognition from others and to project a desired image. The four items that were used to measure self-enhancement were derived from Hennig-Thurau et al. (2004). The items were; In general, I post online about products/companies because “…I feel good when I can tell others about my buying success”, “…I can tell others about a great experience”, “this way I can express my joy about a good buy” and “…my contributions show others that I am a clever customer”. Responses on the items were averaged to provide a measure for eWOM motivated by self-enhancement (α = .93).
Advice seeking. This construct regarded the consumer’s motive to engage in eWOM to
obtain advice to resolve a problem with the (usage of) the product. This construct was measured by two items developed by Hennig-Thurau et al. (2004). The items were; In general, I post
Laura Dirksz, 5754461 30 online about products/companies because “…I expect to receive tips or support from other users” and “…I hope to receive advice from others that help me solve my problems”. Cronbach’s alpha for this scale was .89. The average score from these two items were used as a measure for advice seeking as a motivating factor for eWOM.
Economic incentives (2 items). This construct was measured by two items adapted from
Hennig-Thurau et al. (2004). It reflected the degree in which an economic incentive served as a motivation for the consumer to engage in eWOM communications. The items were; In general, I post online about products/companies because “…of the incentives I receive” and “…I receive a reward for the writing”. However, the internal reliability of this scale was very poor (α = .29). This scale only consisted of two items, therefore removal of an item were not possible solutions to increase the internal consistency of this scale. Therefore, it was not able to compute a reliable scale for measuring eWOM motivation driven by economic incentives. This factor was not included in subsequent analyses.
Concern for other consumers. Originally this scale consisted of four items that were
developed by Hennig-Thurau et al. (2004). The reliability analysis revealed a Cronbach’s alpha of .70. However, a review of the content of the items indicated the possibility of two factors consisting of two items each. The two factors had higher reliabilities than the original scale consisting of four items. The first two items assessed the consumer’s motivation to engage in eWOM because of a desire to help other consumers make the right purchase decision. These items were; In general, I post online about products/companies because “…I want to help others with my own positive experience”. These two items were renamed into the scale “Helping other
Laura Dirksz, 5754461 31 The other two items originally belonging to the concern for other consumer scale were: “…I want to warn others of bad products/companies” and “…I want to save others from having the same negative experience as me”. These items reflected a motivation for eWOM because of a desire to spare other consumers from a negative experience. Therefore, these two items were renamed into the scale “Warning other consumers” (α = .82). A similar separation of the items and renaming of the scales was also done by Yap et al. (2012).
Vengeance (2 items). This constructs focused on the consumers desire to retaliate against
the company after a negative experience as the motive for eWOM. It was measured by two items adapted from Hennig-Thurau et al. (2004). The items were; In general, I post online about products/companies because “…the company harmed me, and now I will harm the company!” and “…I want to take vengeance upon the company” (α = .94).
Anxiety reduction. Anxiety reduction reflected a consumer’s motive for eWOM
communication which is rooted in a desire for catharsis to ease anger, anxiety or frustration following a negative experience. This construct was measured by two items adapted from
Hennig-Thurau et al. (2004). The items were; In general, I post online about products/companies because “…my contributions help me to shake off frustrations about bad buys” and “…I like to get anger off my chest”. The Cronbach’s alpha for this scale was marginal at .70.
4.4 Procedure
The survey was programmed using the Qualtrics online survey software. The survey was distributed through the researcher’s Facebook page and through e-mails during a two month period, from May to June 2014. The link to the survey was accompanied by a cover letter inviting respondents to participate in a survey about online postings about products or companies, see Appendix B.
Laura Dirksz, 5754461 32 The link directed the respondents to a brief introduction specifying the conceptualization of eWOM on each channel. Then, respondents were asked to indicate on which platform they had recently engaged in eWOM on, either Facebook, an opinion platform or both. Respondents who indicated not having engaged in eWOM on any channel were redirected to the end of the survey. The second question inquired the respondents’ frequency of eWOM engagement on the channel(s) they indicated having used for eWOM. Regarding Facebook, respondents were asked to indicate how frequently they posted status updates on their profile that mentioned a product or company. For opinion platforms, respondents were asked to indicate their frequency of postings that mention a product or company on opinion platforms, with examples such as Yelp.com, Tripadvisor.com, Beslist.nl, Tweakers.net. In the next block, respondents were presented with 28 statements intended to measure the eWOM motives. The survey was programmed to randomize the order in which the statements were presented. The last section of the questionnaire inquired the gender, age and occupation of the respondent. See Appendix C for the complete survey.
5. Results
Out of the 233 respondents, a total of 96 (41.2 per cent) reported having engaged in eWOM only on Facebook and 106 respondents (45.5 per cent) reported having engaged in eWOM only on Opinion platforms. Thirty-one respondents (13.3 per cent) had engaged in eWOM on both platforms. The respondents who had engaged in eWOM on both platforms were excluded from subsequent analyses since their responses pertained to both channels and therefore could not be distinguished per channel. Thus, the final sample consisted of two groups, the first group was Facebook eWOM engagers (N = 96) consisting of 42 men and 54 women aged 18 to 28 years (men: M = 23, SD = 2.5, women: M = 24, SD = 2.4) with yearly posting frequency ranging from 1 to 156 (M = 23, SD = 36). The second group was opinion platform eWOM
Laura Dirksz, 5754461 33 engagers (N = 106) consisting of 49 men and 57 women aged 18 to 28 years (men: M = 22, SD = 1.7, women: M = 23, SD = 2.6) with a yearly posting frequency ranging from 1 to 52 (M = 11,
SD = 12). The distribution of the number of eWOM posts was highly skewed for both channels.
A logarithmic transformation was applied to the outcome variables yearly Facebook posts and yearly opinion platform posts to address their skewed distribution. After this transformation the normality assumption was met for the use the transformed outcome variables yearly Facebook posts and yearly opinion platform posts for the parametric tests.
5.1 Relationship between eWOM motives and eWOM channel
The relationship between the ten eWOM motivation scales and eWOM posts on Facebook was investigated using Pearson product-moment correlation coefficient. Preliminary analyses were performed on each variable to ensure that the assumptions of linearity, normality and
homoscedasticity were not violated. The result of the correlation analyses are presented in Table 1. The correlations of the ten eWOM motivations with eWOM engagement on Facebook were low and not significant. This indicated that there was no linear relationship between the eWOM motives and the frequency of eWOM posts on Facebook in this sample.
Correlation coefficients were also computed for the ten eWOM motivation scales and eWOM posts on opinion platforms. Preliminary analyses were performed on the variables to ensure that the assumptions of linearity, normality and homoscedasticity were not violated. The result of the correlation analyses are presented in Table 2. The correlations of the ten eWOM motivations with eWOM engagement on opinion platforms were not significant. This indicated that there was no linear relationship between the eWOM motives and the frequency of eWOM posts on opinion platform in this sample.
Laura Dirksz, 5754461 34 Multiple regression analysis with the ten eWOM motives as the predictor variables and the number of eWOM posts on opinion platform or on Facebook could not result in significant models because of the lack of correlation between the ten eWOM motives and the number of eWOM posts on Facebook and the number of eWOM posts on opinion platforms, respectively.
Table 1. Means, Standard Deviations and Correlations for Facebook eWOM engagers (N = 96)
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Yearly Facebook Posts 23.26 35.93 -
2. Message Involvement 4.27 1.89 -.02 -
3. Product Involvement 4.43 1.41 -.01 .95** -
4. Helping Other Consumers 4.43 0.84 -.09 .74** .69** -
5. Warning Other Consumers 3.22 0.84 -.04 -.59** -.67** -.26* -
6. Self-Enhancement 4.36 1.48 .01 .45** .43** .34** -.53** -
7. Social Benefits 4.72 1.29 .02 .61** .64** .21* -.88** .77** -
8. Helping the Company 4.20 0.75 -.05 .46** .43** .44** -.35** .71** .49** -
9. Advice Seeking 3.60 1.90 0.01 0.20 .22* .54** 0.08 0.14 -.24* .28* -
10. Vengeance 2.49 0.82 0.01 -.91** -.87** -.56** .72** -.48** -.76** -.47** 0.13 - 11. Anxiety reduction 3.02 1.32 -0.03 -.57** -.54** -0.07 .55** -0.17 -.59** -0.08 .54** .81** - 12. Percieved Interaction 5.26 1.15 0.17 0.02 -0.01 -0.05 0.12 -0.11 -0.08 -0.10 -0.05 0.03 0.00 - 13. Percieved Reach 3.43 0.95 0.04 -0.02 -0.01 0.10 -0.03 -0.09 -0.08 -0.11 0.13 0.06 0.11 0.07 - *Correlation significant at the 0.05 level (2-tailed)
Laura Dirksz, 5754461 36
Table 2. Means, Standard Deviations and Correlations for opinion platform eWOM engagers (N = 106)
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Yearly opinion platform posts 10.40 11.97 -
2. Message Involvement 2.10 0.59 .046 -
3. Product Involvement 3.01 1.07 -.03 .55** -
4. Helping Other Consumers 4.89 1.45 .02 .30** .13 - 5. Warning Other Consumers 5.30 1.40 .20 -.06 -.65** .113 -
6. Self-Enhancement 3.86 1.30 -.13 .18 .51** .09 -.76** -
7. Social Benefits 3.09 1.09 -.14 .15 .16 .14 -.52** .84** -
8. Helping the Company 4.77 1.66 -.08 .21* .60** .25* -.65** .93** .73** -
9. Advice Seeking 3.26 1.23 -.17 .10 .67 ** -.25* -.96** .80** .50** .72** -
10. Vengeance 4.11 1.58 .14 -.52** -.67** .07 .80** -.69** -.40** -.55** -.84** - 11. Anxiety reduction 3.25 1.00 .18 .22* .04 -.37** .55** -.48** -.45** -.29** -.41** .38** - 12. Percieved Interaction 3.36 1.20 -.16 .02 -.16 .18 .16 -.22 -.04 -.20* -.23* .17 -.02 - 13. Percieved Reach 5.46 1.08 -.04 -.08 -.11 .00 .02 -.12 -.08 -.16 -.06 .04 -.10 .17 - *Correlation significant at the 0.05 level (2-tailed)
The number of posts on Facebook or on an opinion platform could not be predicted by any of the ten motives for eWOM, consequently the hypotheses could not be tested in the way that they were formulated. The relationships between the motive and the number of eWOM posts was not present and thus the effects of the motives on the channel choice could not be investigated. Therefore, it was chosen to test whether there were any differences in the means on each motive between the Facebook posters and opinion platform posters.
5.2 Comparison of motivations between eWOM on Facebook and on opinion platforms
A series of independent sample t-tests was conducted to compare the motivations of those who engaged in eWOM on Facebook (N = 96) with those who engaged in eWOM on opinion platforms (N = 106). The effect sizes of these differences were also calculated in order to assess the strength of the difference between the groups. The guidelines proposed by Cohen (1988) were used to interpret the effect sizes.
Significant differences were found for nine out of the ten eWOM motives. The motive where no significant differences between the groups was found was advice seeking t(160.29) = 1.49, p = .14. In total, Facebook posters scored higher on a three eWOM motives compared to opinion platform posters. The first motive was message involvement. The results indicated a significant difference between the two groups, t(111.68) = 10.79, p < .001. Those who engaged in eWOM on Facebook (M = 4.27, SD = 1.89) were more motivated by message involvement than those who engaged in eWOM on opinion platforms (M = 2.10, SD = .59). The magnitude of the difference in the means (mean difference = 2.17, 95 % CI: 1.77 to 2.57) was large (d = 1.55). The scores for product involvement also significantly differed for the groups, t(176.64) = 8.00, p < .001), with higher product involvement scores for those who engaged in eWOM on Facebook
Laura Dirksz, 5754461 38 (M = 4.43, SD = 1.41) than those on opinion platforms (M = 3.01, SD = 1.07). The mean
difference was 1.42 (95 % CI: 1.07 to 1.77) and had a large effect size (d = 1.13). The third motive with a significant difference was self enhancement. Facebook posters (M = 4.36, SD = 1.48) were more motivated by self enhancement than opinion platforms posters (M = 3.86, SD = 1.30). This difference was significant with t(190) = 2.52, p < .05. The Cohen‘s d of 0.33
indicated a small effect size (mean difference = 0.50, 95 % CI: 0.11 to 0.88). Another motive where Facebook posters scored significantly higher on than opinion platform posters was social
benefits, t(200) = 9.75, p < .001, with Facebook posters (M = 4.72, SD = 1.29) being more
motivated by social benefits than opinion platforms posters (M = 3.09, SD = 1.08). The magnitude of the mean difference of 1.63 (95 % CI: 1.30 to 1.97) was large (d = 1.37).
Those who engaged in eWOM on opinion platforms scored significantly higher on five motives compared to those who engaged in eWOM on Facebook. The first motive was helping
the company, t(148.81) = - 3.23, p < .05, with opinion platforms posters (M = 4.77, SD = 1.66)
scoring higher on this motive than Facebook posters (M = 4.20, SD = .74). The effect of this mean difference was moderate (d = - 0.50, mean difference = 0.54, 95 % CI: - 0.93 to – 0.22). The second motive was vengeance t(160.81) = - 9.29, p < .001, opinion platform posters (M = 4.11, SD = 1.58) were more motivated by vengeance than Facebook posters (M = 2.49, SD = .81). The magnitude of the difference in means (mean difference = 1.62, 95 % CI: - 1.96 to – 1.27) was large (d = - 1.28). The means on helping other consumers t(170.76) = - 2.71, p < .05 were also significantly different between the groups, with respondents who engaged in eWOM on opinion platforms (M = 4.88, SD = 1.45) indicating a higher motivation to help other
consumers than those on Facebook (M = 4.43, SD = .84). Cohen’s d indicated a small effect (d = - 0.38) of the mean difference of 0.45 (95 % CI: - 0.78 to – 0.13). Mean scores on anxiety
Laura Dirksz, 5754461 39
reduction also differed between the groups. Opinion platforms posters (M = 3.24, SD = .99)
scored significantly higher on anxiety reduction t(175.60) = 8.00, p < .001 than Facebook posters (M = 3.02, SD = 1.32). The Cohen’s d was – 0.20, indicating a small effect (mean difference = 0.22, 95 % CI: - 0.55 to 0.10). A significant difference was also found for the motive warning
other consumers, t(173.90) = - 12.89, p < .001, with a large effect size (d = - 1.79) of the mean
differences (mean difference = 2.08, 95 % CI: - 2.40 to – 1.75).Those who posted on opinion platforms (M = 5.30, SD = 1.40) were more motivated by warning other consumers than those who posted on Facebook (M = 3.22, SD = .84).
Laura Dirksz, 5754461 40
6. Discussion and conclusions
This study sought to test the proposition that consumer’s eWOM motives would predict the likelihood of posting eWOM on either Facebook or on a consumer opinion platforms. It was hypothesized that consumers motivated by message involvement, self-involvement, social benefits and anxiety reduction were more likely to engage in eWOM on Facebook than on an opinion platform. Second, it was hypothesized that consumers motivated by helping the
company, advice seeking, vengeance, concern for others and self-enhancement were more likely to engage in eWOM on an opinion platform than on Facebook. Multiple regression analyses were supposed to be conducted to delineate the relationship between the eWOM motives and the eWOM channels. However, the initial correlation analysis showed very weak relationships which were not significant and thus indicated that these relationships could occur simply by chance. Since a relationship between the variables was not found, the multiple regression analysis would not be able to indicate the strength of the cause and effect relationship between the eWOM motives and the eWOM channels.
Subsequently, it was chosen to examine whether there were any differences between the motives that drove eWOM on Facebook and the motives that drove eWOM on opinion
platforms. The results indicated that eWOM on Facebook was more driven by message
involvement, product involvement, self enhancement and social benefits than eWOM on opinion platforms. Whereas eWOM on opinion platforms was more driven by wanting to warn and help other consumers, take vengeance on the company, help the company and reduce anxiety.
The large effect for the differences in the underlying eWOM motive of vengeance and warning other consumers imply that eWOM on opinion platforms is more characterized by a desire to spare other consumers from an unsatisfactory experience or to deter others from
Laura Dirksz, 5754461 41 patronizing the product or company. This is in accordance with the proposition that the consumer might best achieve the goal of retaliation and the goal of warning other consumers by exposing the negative experience to the most amount of people. This is more attainable on opinion platforms which are public, than on a personal Facebook profile which is private. eWOM on Facebook was more characterized by a desire to share excitement resulting from either use of the product or from the advertisement and from a desire to interact with other users. These reflect the social aspects of Facebook of self-disclosure and social interaction to build and maintain
relationships (Ellison et al., 2007; Lampe et al., 2006). The differences between motives underlying eWOM posts on Facebook and posts on opinion platform possibly suggest that the consumer considers which channel is better suited to achieve the goal for posting. Further research could investigate what the characteristics are that the consumer considers when
choosing a channel for eWOM expression. Research could also uncover whether this process of channel choice takes form in a conscious or subconscious consideration of the channel
characteristics and eWOM goals.
The lack of a relationship between the eWOM motives and the number of eWOM posts in this research could be attributed to differences in the sample. For this study the sample was collected through Facebook and e-mail thereby excluding those who post on opinion platforms that do not have a Facebook account. Whereas in Hennig-Thurau et al. (2004) collected the sample directly through opinion platforms. Thus the possibility of a sampling bias should be considered, and the lack of results could be attributed to this and not necessarily the variables that were researched. Second, the main argument was that SNS, and specifically Facebook possessed a unique set of characteristics that differentiated it from opinion platforms. Facebook had some characteristics related to offline WOM, such as the already established social
Laura Dirksz, 5754461 42 relationships between the user and its network. This lead to the inclusion of other motives found in research on offline WOM. The basis for this study was the framework developed by Hennig-Thurau et al. (2004) and the purpose was to explore the motives in a different setting. Research on eWOM motives by Hennig-Thurau et al. (2014) did manage to find a relationship between the motives and the frequency of posting. eWOM communication was represented by a single
parameter, eWOM frequency. The instructions on the survey prior to the assessment of eWOM motives did not ask the respondents to recall one specific incident. The questions were
formulated to inquire the respondents’ general motives for engaging in eWOM. It should be considered that multiple occasions of posting came to the respondents’ minds which lead to the indication of multiple motives.
The adoption of both technologies should also be considered as a factor that may influence eWOM expression and channel choice. If the consumer who has a Facebook account has to set up a new account to post on an opinion platform or is unfamiliar with the opinion platform functionalities, it entails an extra effort. Another possible limitation is the definition used for eWOM on Facebook. The conceptualization of eWOM on Facebook being limited to postings on a consumer’s personal profile was to allow an accurate comparison with opinion platforms. On Facebook, the concept of eWOM could be broadened to include content about products or companies that are not originally produced by the consumer such as re-sharing posts and liking posts, but also check-ins and liking pages that signal brand preference.
Additionally, consumers on Facebook might also post on company or product pages and not on their own profile. Thereby they can reach a specific audience, being other consumers who also share an interest in the product or company. Posting on the company page could also allow a respond from the company itself, which is beneficial for consumers who are seeking advice.
Laura Dirksz, 5754461 43 Thus the other functionalities of Facebook should also be considered as a means for eWOM. These could include sharing through instant messaging, sharing on profiles of friends, sharing in groups and sharing on brand pages. Within Facebook where eWOM is most likely to be shared and which motive lead to their expression.
This study was not able to assess whether consumers driven to engage in eWOM by a specific motive engaged more often in eWOM on one channel compared to another. However, it did reveal that in general, posts on each respective channel were more driven by certain motives. The differences in eWOM motives on each channel suggests that particular motives are more present in eWOM on one channel over the other. Thus further research is needed in order to establish whether eWOM motives actually influence the channel choice for engaging in eWOM.
