24-6-2016
‘EWOM on the
Facebook wall, which
is the fairest brand of
them all?’
The role of platform type and source
credibility in the relationship between
EWOM and consumer engagement
Sirkka van Loon 10345329
MASTER THESIS
UNIVERSITY OF AMSTERDAM
GRADUATE SCHOOL OF COMMUNICATION MASTER PERSUASIVE COMMUNICATION SUPERVISOR: S.F. BERNRITTER
1
Abstract
The aim of the current study was to investigate the influence of exposure to positive and negative Electronic Word of Mouth (EWOM) on consumer engagement. Based on Source
Credibility Theory it was expected that this relationship would be mediated by source
credibility. Moreover, based on Social Identity Theory and previous studies it was expected that EWOM on different platform types would vary in degree of source credibility. Results collected through an online experiment indicated that EWOM had a negative effect on cognitive consumer engagement, while source credibility had a positive effect on affective and active consumer engagement. No evidence was found, however, for the expected mediation and moderation effects. These findings extend the existing body of literature on EWOM by investigating the credibility of EWOM on different platform types. In addition managerial implications for dealing with EWOM on different platform types are provided.
2
Introduction
Social media have given a new dimension to building strong brands and the creation of communities. Social platforms like Facebook, Twitter and Instagram now give consumers the chance to follow, like and share their favorite brands. In addition it is possible to directly contact and interact with brands through social media by posting messages on their company page or by reacting on posts made by a brand (Kietzmann, Hermkens, McCarthy & Silvestre, 2011). On the other hand consumers can also interact with each other about brands by sharing experiences, writing reviews, recommending products and creating brand communities. These online interactions are called Electronic Word of Mouth (EWOM) which is defined as online communication that offers consumers the chance to express their personal, unbiased opinions and recommendations about products and brands (Hennig-Thurau, Gwinner, Walsh &
Gremler, 2004).
The Dutch brand Hema has for example her own official Hema Facebook pages for her different stores, but there also exist several Hema communities and fan pages for the brand. One of the largest fan pages is the ‘We love Hema’ community page that was specially created for and by Hema fans all over the world. The page has 7.600 likes and the Facebook users on this page share information and opinions with each other about Hema products in a positive way. Whereas on the official Hema Facebook page consumers more often tend to express their critique and complaints. Thus there are clear differences in the sentiment and manner of communication between the official Hema Facebook pages and the Hema fan pages. These differences cause a problem, however, since the different nature of the interactions can as a result have differing effects on the consumer that reads this messages. The question is therefore what these different effects are, how they can be explained and what they mean for managing brands in a social media environment and Facebook in particular.
To answer this question a closer look at the two different platform types is needed. On the one hand there is the page created by the brand, the company generated platform and on
3
the other hand there are brand communities created by consumers, the consumer generated platforms. This distinction between company and consumer generated platforms has not always been taken into account in previous research in the field of EWOM. The effects of EWOM on these platforms does differ, however, since the interactions on both platforms are by nature different. The interactions on consumer generated platforms are more
conversational and aimed at discussing the brand without company representatives observing, while consumers on company generated platforms more often make statements directed at the brand rather than to start a conversation with other consumers (Van Noort & Willemsen, 2011). Therefore it is important to study the influence of positive compared to negative EWOM through different types of platforms on social media.
In addition, The Social Consumer Decision Journey states that social media can affect consumers at every stage of the consumer decision journey, creating a winding instead of the traditional linear path from consideration to purchase (Court, Elzinga, Mulder & Vetvik, 2009). During this journey consumers’ experiences influence which brand they prefer and the potential advocacy of these consumers influences other consumers. Since consumers often remain engaged after making a purchase and publicly promote or disparage the purchased products and brands on social media (Court et al., 2009). It is therefore important not just to look at concepts from the traditional consumer decision journey, like purchase intentions and product evaluations, but to examine consumer experiences, bonding and engagement as well. The current study will therefore examine the influence of EWOM on consumer engagement, which can be described as the collection of experiences consumers have with a brand and the behavioral manifestations they perform in relation to this brand (Calder, Malthouse &
Schaedel, 2009; Van Doorn et al., 2010).
Previous literature suggests the credibility of the source plays an important role in the relationship between EWOM and consumer engagement. Since the platform where EWOM is
4
uttered could play a part in the perceived credibility of the source of that EWOM message. Previous research has found that corporate websites and consumer review websites can differ in the extent to which they generate interest for a product (Bickart & Schindler, 2001).
Moreover, the credibility of EWOM on company generated pages is supposedly less than that of consumer generated pages (Tsao & Hsieh, 2015). All together the current study will therefore focus on answering the following research question:
RQ: What is the influence of positive, in contrast to negative, EWOM on consumer engagement through the credibility of the source and what moderating role does platform type play in this relationship?
By answering this research question, the present research will provide three key contributions to both research and practice. First of all the existing knowledge about EWOM is extended and the effects of positive and negative EWOM on different platform types are shown. With these findings we will be able to answer the question what the different effects of positive and negative EWOM are on consumer compared to company generated platforms. Allowing us to assess the feasibility of different word of mouth marketing strategies within these platforms.
Secondly, by researching the differences in the amount of source credibility of EWOM on different platform types the current research answers the call from King, Racherla and Bush (2014) for more research into the ways trust and the anonymity of the online
environment change the power of EWOM. In doing so, it will be made clear if differences in the perceived credibility of an EWOM source will increase or decrease the influence of EWOM. Moreover, two possible drivers of consumers’ evaluation of source credibility will be examined: EWOM valence and platform type. These two drivers and source credibility have not yet been empirically examined in the combination that is conceptualized in the current study. This study therefore extends and connects existing research on EWOM, source credibility and different platform types.
5
Third, this study draws from The Social Consumer Decision Journey by stating that online experiences with brands and other consumers, like EWOM, can increase consumer engagement. By doing this the current research moves away from the level of the traditional consumer decision journey towards a model that takes the innovative features of social media into account. This also answers the earlier call made by Hollebeek, Glynn and Brodie (2014) for more research into the under-explored concept of consumer engagement and the extension of this principle to consumer interactions with and about brands on social media.
In the following section a short literature review of what is already known about EWOM from research up until now will be presented. After that the central concepts consumer engagement, source credibility and platform type will be defined and a review of several theories and empirical findings in relation to these concepts will be presented. Based on these findings several hypotheses will be formulated which then will be tested through an experiment. Finally the results from this experiment and a conclusion and discussion will be presented.
Theoretical background
Electronic word of mouth
EWOM has several distinctive characteristics that set it apart from traditional forms of word of mouth. Hennig-Thurau et al. (2004) have described EWOM as the online conversations about brands or products by consumers. These conversations take place in real-time on platforms like Twitter, blogs, review pages or Facebook. Because of the public character of these platforms a message about a brand can immediately be seen by a large amount of users and, in contrast to traditional word of mouth, the words remain in the public sphere long after they have been uttered (Hennig-Thurau et al., 2004). While traditional and electronic word of mouth both concern the recommendation or disparagement of a product or brand, the internet and social media have made the spread of word of mouth faster and easier (Sen & Lerman, 2007). In addition, the anonymity of the internet has decreased the barriers to voice unpopular
6
opinions (Hennig-Thurau et al., 2004). As a result consumers are not afraid to make their dissatisfaction with a product or brand known.
Additionally, EWOM has proven to be an extremely influential source of information for consumers. Consumers that were exposed to information from online discussions about a product showed greater interest in that product than did consumers who were exposed to marketer-generated sources (Bickart & Schindler, 2001). This is because the exposure to personal interactions about a brand has more influence on consumer choices than advertising has (López & Sicilia, 2013). As a result, consumers are able to make or break products and brands with what they write online (Fournier & Avery, 2010). Some companies have learnt to take advantage of this, however, and use word of mouth for their own marketing purposes. Research has found these companies that implement word of mouth marketing strategies to be more successful (Pfeffer, Zorbach & Carly 2014). EWOM can thus be an important tool in online brand management and the influence of the consumers that write these messages should not be underestimated.
The influence of EWOM has already been confirmed in several studies. So has EWOM been found to influence the brand image (Sandes & Urdan, 2013), product attitude (Sen & Lerman, 2007), consumer engagement (King et al., 2014) and purchase intention (Chevalier & Mayzlin, 2006) in relation to a product or brand. In all of these studies positive EWOM was found to have a positive effect, while negative EWOM was found to have a negative effect. Hennig-Thurau, Wiertz and Feldhaus (2015) for example found that negative EWOM about a recently released movie tweeted from consumer generated platforms caused a decrease in the amount of people visiting that movie in theaters. However, positive EWOM about this movie did not increase the amount of visitors. On the other hand, research has indicated that positive EWOM can increase sales, while negative EWOM can cause a decrease in sales (Chevalier & Mayzlin, 2006). Moreover, both the valence and the quantity
7
of EWOM has appeared to be able to increase purchase intentions (Park, Lee & Han, 2007; Xia & Bechwati, 2008). Since the quantity of EWOM signals the popularity of the product to other consumers (Park et al., 2007).
These contradicting findings in the influence of positive and negative EWOM on purchase intentions and sales can be explained by the negativity bias that exists online. This principle states that negative messages are more rare and weigh therefore more heavily in the eventual purchase decision (Hennig-Thurau et al., 2015). A negativity bias has, however, only been confirmed for Twitter and online review sites. With online reviews this negativity bias only exists for utilitarian products. For hedonistic products a positivity bias has become apparent, where receivers pay more attention to and are more influenced by positive reviews (Sen & Lerman, 2007). Because there are no conclusive findings for positivity nor negativity
biases on Facebook and Twitter and online review sites strongly differ from Facebook on key
features like the amount of characters used, technical functions and the amount of social interaction (Schweidel & Moe, 2014) an equal influence of positive compared to negative EWOM is expected on Facebook. In other words, the current study argues that on average positive EWOM will positively influence consumers to the same extend as negative EWOM will negatively influence consumers when posted on Facebook.
Consumer engagement
As social media and the EWOM that is posted within these media are able to influence every stage of the consumer decision journey and more (Court et al., 2009), new concepts for measuring brand performance within the social media environment have become apparent (Hollebeek et al., 2014). A key new concept is consumer engagement. Described by Van Doorn et al. (2010) as consumers’ behavioral manifestations with a focus on a brand or organization, which go beyond making a purchase and result from the motivation to act. When a consumer is strongly engaged with a brand he or she might for example perform
8
behavioral manifestations like increased online interaction with and about that brand, liking the brand page, sharing brand-related content, improving brand meaning and taking part in value co-creation (De Vries & Carlson, 2014; Men & Tsai, 2013). More specifically these manifestations that consumer engagement consists of can be divided into brand-related cognitive, affective and behavioral activities that occur during or in relation to consumer interactions with and about brands (Hollebeek et al., 2014). Cognitive brand-related activities can be explained as the level of elaboration and thought processing that occurs in consumer interactions with and about brands (Hollebeek et al., 2014), e.g. the consumer thinks a lot about a specific brand and what it means to him or her. Affective brand-related activities on the other hand capture the degree of affect that a consumer experiences in consumer
interactions with and about brands (Hollebeek et al., 2014), e.g. the emotional connection the consumer feels with a specific brand. Finally the behavioral brand-related activities concern the amount of time and effort spent on a brand in consumer interactions with and about brands (Hollebeek et al., 2014), e.g. the amount of time spent interacting about brands online and liking brand pages. Together these brand-related activities construct a consumer’s brand experience. The engagement with a brand therefore includes the collection of experiences a consumer has had with this brand (Calder et al., 2009).
The influence of online interactions on consumer engagement follows from Uses and
Gratifications Theory (Blumler & Katz, 1974). According to this theory consumers seek a
social interaction gratification, which is shown to be resolved by social interactions about brands on social media and Facebook in particular (Jahn & Kunz, 2012). As EWOM is a form of social interaction, it is expected that EWOM can resolve this need for social interaction. Moreover, the social interactions EWOM consists of can cause a feeling of empathy among the readers of these messages. In a way the writers of EWOM messages are experiencing the product for the other consumers. The personal stories that result from this are both
9
entertaining and educating, drawing the reader in and causing this reader to empathize with the feelings of the writer and become involved with the subject that is been written about. In effect, a vicarious experience is created where the consumer vicariously learns about the product or brand (Bickart & Schindler, 2001). Vicarious learning stems from Social Cognitive
Theory (Bandura, 1986) and means people learn new behaviour through others. So, by
observing other people’s experiences with a product or brand, someone can learn about the punishments or rewards of this behaviour (Bandura, 2002). When someone writes a negative EWOM message about a specific product for example, other consumers can observe the negative outcomes of buying this product which might add to the collection of negative experiences with the brand and decreases consumer engagement. On the other hand, positive EWOM will add to the collection of positive experiences with the brand and will therefore increase consumer engagement.
A field experiment by Algesheimer, Borle, Dholakia and Singh (2010) confirmed that positive feedback within an online brand community increased participation in this brand community. Since participation in an online brand community is considered a brand-related activity, consumer engagement was thus increased by positive messages. Negative feedback on the other hand reduced community participation and consequently consumer engagement (Algesheimer et al., 2010). This happens because the positive or negative experiences consumers have within online communities influence the amount of engagement with that community and the brand that particular community is built around (Nambisan & Baron, 2007). The focus of these studies was, however, specifically on consumers participating in online communities and not the influence of positive and negative EWOM messages on consumer engagement in general. A different study by De Vries and Carlson (2014) that did take into account the effects of positive and negative EWOM on consumer engagement in general, found that online social interactions about a brand were able to influence consumer
10
engagement. Specifically, positive online social interactions about a brand increased consumer engagement, while negative online social interactions about a brand decreased consumer engagement (De Vries & Carlson, 2014). Positive EWOM is hence in the current study expected to increase consumer engagement with the brand, while negative EWOM will decrease consumer engagement with the brand. The following hypothesis was therefore developed:
H1: Positive EWOM will increase consumer engagement, while negative EWOM will decrease consumer engagement.
Source credibility
The influence of EWOM on consumer engagement can be explained by the credibility of the source. Opinions and reviews about a brand or product are often considered more trustworthy when they have been written by fellow consumers who have no vested interest in the brand or product and who are perceived to have no intentions to manipulate other consumers (Bickart
& Schindler, 2001). This can be explained further bythe Source Credibility theory, which
states that source expertise and source bias determine the extent to which this source is perceived as a credible source. Source bias refers to the trustworthiness and possible bias of the information provided by a source or the incentives this source receives for providing certain information (Brown, Broderick & Lee, 2007). A source could for example have commercial motives for posting certain product or brand related information.
Next to this, the expertise of a source determines the extent to which this source is considered competent in relation to the information that is provided (Brown et al., 2007; Moran & Muzellec, 2014). An example of this would be if the source described a real-life experience with the product or brand. Based on Source Credibility Theory, a source that has a lot of expertise and is not biased will be considered a credible source (Brown et al., 2007). In addition, a source that is considered credible is found to be more persuasive than a source that
11
is not considered as credible. As the credibility of the source may spillover to the message and as a result lead to more persuasion (Van Noort, Antheunis & Van Reijmersdal, 2012). When a credible source writes a positive review about a brand on social media for example, the readers of this message could be persuaded to like or buy this brand or at least consider it.
However, it is not that easy to determine if a source of EWOM has expertise or is trustworthy. In contrast to traditional word of mouth where interactions take place face-to-face and the receiver of the information can draw upon social context cues to decide whether the source of this information is credible or not, EWOM is posted online by an often
anonymous source (Filieri, 2016). In addition, the source and receiver of EWOM do not have direct contact or prior relationships with each other, which might hinder the development of trust and assessment of source credibility (Filieri, 2016). But, because the receiver of EWOM knows little about the source itself within the anonymity of the social media environment, the content of the message becomes the most important source of information to assess the credibility of the source (Filieri, 2016; Park et al., 2007). The clarity, valence and balance of the information the source provides are in this situation the key features for evaluating its credibility (Moran & Muzellec, 2014).
Research investigating the relationship between EWOM valence and source credibility discovered that when cognitive personalization is high, meaning the reader is able to immerse himself into the story or situation that is described in the message, positive EWOM messages are considered more credible. As a result of this higher credibility the EWOM message had a stronger impact on purchase intentions (Xia & Bechwati, 2008). Negative EWOM, on the other hand, did not have the same increased credibility when cognitive personalization was high (Xia & Bechwati, 2008). These findings were corroborated by Tsao and Hsieh (2015), who found positive EWOM messages to increase consumers’ purchase intentions via source credibility. In addition, empirical findings revealed that source credibility increased the
12
influence that EWOM has on consumers and that positive EWOM was in this case more influential than negative EWOM (López and Sicilia, 2014). This effect of positive EWOM presumably occurs because positive messages are in general more common and therefore less mistrusted by consumers (López and Sicilia, 2014).
The increased influence of EWOM through source credibility has been confirmed by Laczniak, DeCarlo and Ramaswami (2001) as well. They discovered that the credibility of the source mediated the relationship between EWOM and brand evaluations, where a higher credibility of the information caused a stronger impact on brand evaluations afterwards (Laczniak et al., 2001). A more recent study by O’Reilly, MacMillan, Mumuni and
Lancendorfer (2016) revealed that source credibility indeed determines the impact of EWOM messages on consumers. When a source was considered to have sufficient levels of expertise and trustworthiness consumers were more inclined to use and act on the information in the EWOM message (O’Reilly et al., 2016).
Therefore, following the findings from these previous studies that positive EWOM is considered more credible than negative EWOM, it is expected that positive EWOM will have a higher source credibility than negative EWOM. In turn, the amount of source credibility is expected to influence consumer engagement. Since previous findings have revealed that a higher source credibility is able to increase the impact of EWOM on various consumer behaviors we specifically expect a high source credibility to increase consumer engagement. Whereas a low source credibility is expected to decrease consumer engagement. These expectations are reflected in the following hypotheses:
H2: Positive EWOM messages will have a higher source credibility than negative EWOM messages.
H3: A high source credibility will increase consumer engagement, while a low source credibility will decrease consumer engagement.
13 Company vs. consumer generated platforms
The relationship between source credibility and EWOM is in the current study expected to be strongly influenced by the type of platform that is used to communicate information about a product or brand. In online platforms a distinction can be made between consumer generated platforms and company generated platforms. A company generated platform is an official page created by a company or brand used to get in touch with consumers and provide information about the brand and its products (Tsao & Hsieh, 2015). In contrast to this, a consumer generated platform can be defined as a fan page or online community created by consumers out of appreciation for the brand or to share information and experiences with the brand and its products with other consumers (Van Noort & Willemsen, 2012).
A large part of the online engagement consumers have with brands takes place outside of company generated platforms, for example in Facebook groups created by individual consumers or brand communities like the Volkswagen club for car owners (Schamari & Schaefers, 2015). Bickart and Schindler (2001) showed that people who had to read online reviews on consumer generated platforms for 3 months had more interest in the product they read reviews about than people who read messages on corporate websites for 3 months (Bickart & Schindler, 2001). However, in this study participants were asked to look for messages on review sites and corporate websites themselves causing characteristics of the online platform to possibly obscure the results.
A study that did make use of reviews that were consistent across different conditions, discovered that both the quality of the EWOM messages and the type of platform influenced credibility and purchase intentions. EWOM messages that were high in quality and were posted on independent platforms had a higher credibility than qualitatively high EWOM messages on corporate platforms (Tsao & Hsieh, 2015). As the motives of the source for posting EWOM on a corporate platform can be called into question. The source might have
14
posted the message to be able to benefit from the company for example (Tsao & Hsieh, 2015). These hidden commercial motives are expected to be mostly prevalent for positive EWOM about a brand on the company page of that brand. In contrast, Bronner and De Hoog (2010) discovered that there was equal trust in EWOM messages on both consumer and company generated platforms. The majority of the messages on consumer generated platforms was positive, however (Bronner & De Hoog, 2010). This is in line with the literature on brand communities in which it is argued that members of brand communities often interact with each other about brands in a positive manner and feel a personal connection to the group (Zaglia, 2013).
This characteristic of brand communities can be explained by Social Identity Theory (SIT) (Tajfel & Turner, 1979). According to SIT individuals categorize themselves as being a part of various social groups and each membership of a social group represents a social identity to an individual. SIT further argues that individuals strive to see themselves in a positive light and create a positive self-concept through enhancing their social identity within a social group (Ranaweera & Jayawardhena, 2014). Consequently, by spreading positive or negative information about a product or brand that is in line with the beliefs of an individual’s social group, the social identity of this individual within the group will be reinforced.
Members of a brand community share a social identity and feel strongly connected to the brand as it represents their common identity. Because of this, whenever negative information about the brand is encountered, members feel their social identity is threatened and will counter or distort the information to respond in favor of the brand (Chang, Hsieh & Tseng, 2013). The spread of positive information about the brand on the other hand will reinforce the source’s social identity within the brand community. Even when an individual does not strongly identify with a brand community, they will still act congruently with the rest of the group to gain social acceptance (Chang et al., 2013).
15
Based on the finding that community members will counter any negative information that threatens the social identity of their community it is expected that negative EWOM on consumer generated platforms will be considered less credible. Because consumer generated platforms often consist of fans and admirers of a brand who form a brand community with a common social identity, they will counter any negative brand information to protect this identity. Positive EWOM on the other hand will be more appreciated and will be considered more credible on consumer generated platforms as this reinforces the in-group social identity. In contrast, positive messages on company generated platforms will be considered as less credible due to possible hidden commercial motives of the source. From this line of reasoning the following hypothesis was formulated:
H4: The influence of EWOM on source credibility depends on the type of platform the EWOM message is posted on: Positive EWOM posted on a consumer generated platform and negative EWOM posted on a company generated platform will increase source credibility, while negative EWOM posted on a consumer generated platform and positive EWOM posted on a company generated platform will decrease source credibility.
16
Based on the hypotheses developed in the previous sections the conceptual model shown in Figure 1 was developed.
Figure 1 Conceptual model H2 H4 H3 H1 Method
Sample & Design
A convenience sample consisting of 142 individuals from the general Dutch population above 18 voluntarily took part in the current research. The sample had an average age of 28 (M = 28.29, SD = 13.03), 64.3% was female and the majority of the participants, 90.5%, were Facebook users. However, because the non-Facebook users did not significantly differ from the Facebook users the anticipated removal of the non-Facebook users from the analyses was not necessary. All participants were randomly assigned to one of four conditions in an online experiment with a 2 (positive vs. negative EWOM) x 2 (consumer vs. company generated platform) factorial between subjects design.
Procedure
Participation in this experiment took place online through Qualtrics. Participants received a link to the experiment after which they could voluntarily decide to participate in the
Positive vs. Negative EWOM Platform type Consumer vs. Company created Consumer engagement Source credibility
17
experiment. First, participants were exposed to a screenshot of a positive or negative EWOM message on an either consumer or company generated platform of the European brand True fruits. After the exposure to the EWOM message participants were asked to fill out a short questionnaire, measuring the credibility of the source and the amount of consumer
engagement with the brand. At the end of the questionnaire a manipulation check was performed by asking several questions about the EWOM message participants had been exposed to. Lastly, participants had to answer a few more personal questions about their age, gender and Facebook usage.
Manipulations
The manipulations in this experiment consisted of screenshots of real Facebook pages which were edited to show an either positive or negative EWOM message. The Facebook pages that had been selected were about the brand True fruits. This is an European brand that sells a variety of different smoothies 100% made of fruit. The brand is not very well-known in the Netherlands, but it is very popular in Germany, Switzerland and Austria. Because of this existing attitudes and intentions in relation to the product or brand will not influence the results of this study. Next to this, True fruits can be considered a rather neutral product that is relevant to both men and women. Because the product is unfamiliar to participants they will rely more heavily on evaluations of others, as EWOM often plays an important part in the adoption of new products (López & Sicilia, 2013). The EWOM messages were designed for this specific experiment to make sure the messages were completely similar to each other except for the words expressing a positive or negative sentiment. This was done to guarantee the internal validity of this study as much as possible. The positive message was formulated in the following way: ‘True fruits is the best smoothie I’ve ever tasted, you should definitely try
this!’. The negative message was formulated as follows: ‘True fruits is the worst smoothie I’ve ever tasted, you should definitely not try this!’
18
The second manipulation concerned the type of platform on which the messages were shown. Platform type was described as the online environment in which the EWOM messages are placed. This factor can vary between platforms that are consumer generated and platforms that are company generated (Van Noort & Willemsen, 2012). In the current experiment Facebook was chosen as the social media channel to present the EWOM messages on. Since Facebook has both a variety of company pages as well as fan pages and communities. In addition, Facebook offers great possibilities for manipulation and it is widely used within the Dutch population. The True fruits company Facebook page was used as a template for the company generated platform. Next to this, the same template was used for the consumer generated platform with a few adaptations to the page description and headline. The EWOM messages that were formulated were edited onto the screenshots. The complete screenshots can be found in appendix 1 (Figure 2, 3, 4 and 5). By making use of existing Facebook pages the external validity of this experiment is guaranteed since the stimuli are very realistic.
Dependent measures
The first dependent variable that was used in this study is the credibility of the source. Source credibility can be described as the extent to which participants consider the source of the EWOM message to be trustworthy and to possess a considerable amount of expertise
(Willemsen, Neijens & Bronner, 2012). To measure this concept an existing scale developed by Ohanian (1990) was used to measure the credibility of the source. This scale consists of four items measuring the trustworthiness of the source, including the items ‘honest/dishonest’ and ‘sincere/insincere’. Another four items measured the expertise of the source, including items as ‘experienced/inexperienced’ and ‘informed/uninformed’. Participants could answer these items on a 7-point Likert scale in which the positive half of the item was placed at point 7 and the negative half of the item was placed at point 1. The total amount of items and the scale reliability can be found in Table 2 and 3 in appendix 2.
19
The dependent variable consumer engagement was conceptualized as the behavioral manifestations of consumers with a focus on a product or brand, which go beyond the mere intention to make a purchase and construct the overall brand experience (Van Doorn et al., 2010). These behavioral manifestations can be divided into brand-related cognitive, affective and behavioral activities. Therefore the consumer engagement scale developed by Hollebeek et al. (2014) that has integrated a cognitive, affective and activation component of consumer engagement was adopted. As was to be expected following Hollebeek et al. (2014), the three factors all loaded on separate components and were hence divided into three separate scales. Cognitive consumer engagement included three items, like for example ‘When I would use
True fruits, I would be stimulated to learn more about this brand’. Affective consumer
engagement consisted of four different items, including ‘Using True fruits would make me
happy’. Active consumer engagement included another three items, like for example ‘When I’m looking for a smoothie, I would choose True fruits’. Participants could answer these items
on a 7-point Likert scale, ranging from ‘strongly disagree’ to ‘strongly agree’. The complete items and reliability of the scales can be found in Table 4, 5 and 6 in appendix 2.
Manipulation checks
To check if the manipulations had been successful and to guarantee the internal validity of the measurements, four different manipulation checks were performed. First of all, the
participants were asked which type of platform they had recognized when being exposed to the screenshot (consumer or company generated). This was measured by making use of a simple multiple choice question, in which participants could answer if they thought the Facebook page was either created by A) a company/brand or B) a consumer/fan. After this participants were asked if they had experienced the message they had seen as rather positive or negative. To measure this participants were asked to rate the message on a scale of 1 to 5, where 1 stood for ‘very negative’ and 5 stood for ‘very positive’. In addition, participants were
20
asked if they had been familiar with the products of True fruits before participating in the current experiment to make sure they were really not familiar with the product or brand beforehand. This was measured with the response categories ‘Yes’, ‘I don’t remember’ and ‘No’.
Analyses
To analyze the data that was collected with the questionnaire, first several manipulation and randomization checks were carried out. For these manipulation and randomization checks multiple ANOVA, Chi-square and independent samples t-tests were chosen. For the main analyses a regression analysis was selected in combination with a PROCESS analysis based on model 7 from Hayes (2013). With the use of a PROCESS analysis it is possible to test the entire mediation-moderation that was conceptualized in this study within one analysis.
Results
Randomization checks
To check if the randomizations of age, general preference to drink True fruits, brand familiarity and product attitude were successful multiple one-way ANOVA analyses were performed. The analyses revealed that there was no main effect of EWOM on age, F(1, 121) = 0.02, p = .892, brand familiarity, F(1,122) < 0.01, p = .955 nor on product attitude, F(1,122) = 3.61, p = .060. There was, however, a main effect of EWOM on the general preference to drink True fruits F(1,122) = 5.70, p = .019. Participants in the positive EWOM conditions generally preferred to drink the smoothie more often (M = 1.20, SD = 0.57) than participants in the negative EWOM conditions (M = 0.94, SD = 0.66). Because of this main effect the general preference to drink True fruits was controlled for in the main analyses. Next to this the analyses revealed that platform type did not have a main effect on age, F(1,121) = 0.42, p = .517, general preference to drink True fruits, F(1,122) = 0.19, p = .666, brand familiarity,
21
analyses reveal an interaction effect of EWOM and platform type for age, F(1,121) = 0.68, p = .411, general preference to drink True fruits, F(1,122) = 2.36, p = .127, brand familiarity,
F(1,122) = 0.28, p = .596 nor for product attitude, F(1,122) = 0.15, p = .698. This means all
but general preference to drink True fruits were successfully randomized.
Another two randomization checks were performed for gender and Facebook usage using Chi-square tests. The results from these tests indicated no differences between EWOM conditions on gender χ²(1, N = 126) = 0.09, p = .853 and between platform type conditions on gender χ²(1, N = 126) = 2.37, p = .140. Neither were there any differences between EWOM conditions on Facebook usage χ²(1, N = 126) = 1.21, p = .367 and between platform type conditions on Facebook usage χ²(1, N = 126) = 1.62, p = .237. Therefore it can be concluded that the randomization of gender and Facebook usage has been successful as well.
Manipulation checks
To see if the manipulations of the different factors were successful several manipulation checks were performed. The first manipulation check evaluated the manipulation of EWOM by making use of an independent samples t-test. Results indicated that the difference in the average score on EWOM between the manipulation check variable and the negative EWOM conditions (M = 1.80, SD = 0.91) was significantly different from the average score on EWOM between the manipulation check variable and the positive EWOM conditions (M = 3.67, SD = 0.98), t(124) = -11.16, p < .001. Meaning that the manipulation of EWOM was successful, as participants experienced the negative EWOM message as negative and the positive EWOM message as positive.
A second independent samples t-test was performed to check the manipulation of platform type. The results from this t-test showed that the difference between the average score on platform type between the manipulation check variable and the company generated platform conditions (M = 0.82, SD = 0.39) significantly differed from the average score on
22
platform type between the manipulation check variable and the consumer generated platform conditions (M = 0.53, SD = 0.50), t(124,117.8) = -3.66, p < .001. Based on these findings the manipulation of platform type can be considered successful, since participants recognized the consumer generated platform as created by a consumer/fan and the company generated platform as created by a company/brand.
Analyses
To test the conceptual model developed in this study a PROCESS analysis by Hayes (2013) was performed using 5000 bootstrap resamples. First we looked at the main effect of EWOM on consumer engagement. EWOM had a significant negative main effect on the cognitive component of consumer engagement (B = -0.53, 95% BCBCI [-0.95, -0.10]). Which means that in contrast to what is stated in hypotheses 1, positive EWOM decreases cognitive
consumer engagement while negative EWOM on the other hand increases cognitive consumer engagement. In contrast, EWOM did not have a significant main effect on the affective
component of consumer engagement (B = -0.11, 95% BCBCI [-0.49, 0.27]), neither was there a significant main effect of EWOM on the activation component of consumer engagement (B = -0.12, 95% BCBCI [-0.54, 0.30]). Therefore hypotheses 1 was rejected.
After that the mediation of the relationship between EWOM and consumer
engagement by source credibility was examined. Results of the PROCESS analysis revealed no significant effect of EWOM on source trustworthiness (B = 0.01, 95% BCBCI [-0.62, 0.63]) nor source expertise (B = 0.28, 95% BCBCI [-0.34, 0.89]). Based on these findings it can be concluded that hypothesis 2 is rejected. Since the findings reveal that EWOM does not appear to have an effect on source credibility.
In addition, the effect of source credibility on consumer engagement was analyzed. There was no significant effect of source trustworthiness on the cognitive component of consumer engagement (B = 0.13, 95% BCBCI [-0.04,0.30]). Source expertise did not have a
23
significant effect on the cognitive component of consumer engagement either (B = 0.12, 95% BCBCI [-0.05, 0.29]). There was a significant positive effect, however, of source
trustworthiness on the affective component of consumer engagement (B = 0.18, 95% BCBCI [0.03, 0.33]), but no similar significant effect was found for source expertise (B = 0.12, 95% BCBCI [-0.04, 0.27]). This means that the more a source is considered trustworthy, the more affectively engaged consumers will be with the brand. Source trustworthiness, however, did not have a significant effect on the activation component of consumer engagement (B = 0.11, 95% BCBCI [-0.06, 0.28]). In contrast, source expertise did have a significant positive effect on the activation component of consumer engagement (B = 0.27, 95% BCBCI [0.11, 0.44]). Meaning that a source that is perceived to have a lot of expertise will cause consumers to be more actively engaged with the brand. Based on these findings hypothesis 3, which expected source credibility to positively influence consumer engagement, was partially supported.
Finally the moderation effect of platform type on the relation between EWOM and source credibility was tested. Platform type did not moderate the relationship between EWOM and source trustworthiness (B = 0.05, 95% BCBCI [-0.84, 0.95]) nor the relationship between EWOM and source expertise (B = -0.40, 95% BCBCI [-1.28, 0.48]) in this model. This means that the expected differences in the credibility of positive and negative EWOM on consumer and company generated platforms, that were formulated in hypothesis 4, are rejected. An overview of the results can be found in Table 1.
24
Table 1
Effect sizes (B) for each relationship between variables (n = 126)
EWOM B Source trustworthiness B Source expertise B EWOM - 0.01 0.28 Platform type x EWOM - 0.05 -0.40 Cognitive consumer engagement -0.53* 0.13 0.12 Affective consumer engagement -0.11 0.18* 0.12 Active consumer engagement -0.12 0.11 0.27** Note. *p < .05; **p < .01; ***p < .001.
Conclusion & Discussion
Conclusion
The present study set out to investigate the effects of positive compared to negative EWOM on consumer engagement. These effects were expected to be mediated by the amount of source credibility consumers experienced and moderated by the type of platform on which the
EWOM message was presented. The results have demonstrated that EWOM did not have the
expected result on consumer engagement. Only the cognitive component of consumer
engagement proved to be significantly influenced by EWOM and all three of the components demonstrated a negative influence of EWOM. This finding is in contrast with the expectation that positive EWOM would cause consumers to be more engaged with the brand, while
25
negative EWOM would cause consumers to be less engaged with the brand. Based on the studies of Hennig-Thurau et al. (2015) and Sen and Lerman (2007) it could be argued,
however, that negative EWOM does indeed attract more attention and has a higher diagnostic value, as it is more rare and unexpected, causing consumers to be more engaged in a negative way by negative EWOM. The findings in this study would in that case confirm the existence of a negativity bias for Facebook as well.
Moreover, source credibility did not have the expected mediation effect on the relationship between EWOM and consumer engagement. EWOM did not prove to
significantly influence source expertise nor source trustworthiness. This might be explained by the fact that the EWOM message was very short and did not offer many opportunities for the receiver to immerse him- or herself into the experience of the source. Since it has been confirmed in previous research that whenever cognitive personalization is high, EWOM can increase source credibility (Xia & Bechwati, 2008). Source credibility did have a partial positive effect on the affective and activation components of consumer engagement, however. When the trustworthiness of the source was high consumers were more affectively engaged with the brand as was predicted in H3. When source expertise was high, consumers tended to be more actively engaged with the brand, as also predicted in H3. Presumably, when the source is perceived to have more expertise consumers feel more capable to use the product themselves and take action, as health communication literature on self-efficacy has indicated before (Conner & Sparks, 2005). When the source is trustworthy on the other hand, this might offer consumers a sense of safety and generate positive feelings towards the brand (Aguirre, Mahr, Grewal, Ruyter & Wetzels, 2015).
The relationship between EWOM and source credibility revealed not to be
significantly moderated by platform type. Meaning that platform type did not play any role in this model. Looking at the effects of platform type, however, it became clear that the source
26
of a positive EWOM message on both the company and consumer generated platform was considered to be more trustworthy than the negative EWOM message. This is in line with the study by Bronner and De Hoog (2010), who discovered that there existed equal trust in EWOM messages on company and consumer generated platforms. The positive EWOM message on the company generated platform had the highest level of trustworthiness though. The source of a positive EWOM message on the consumer generated platform was on the other hand perceived to have more expertise, while the source of a positive EWOM message on the company generated platform was perceived to have less expertise. This partially confirms the expectations in the current study that positive EWOM messages on consumer generated platforms are perceived as more credible, while positive EWOM messages on company generated platforms are considered less credible.
Limitations & Directions for future research
As any study, this study has some limitations. First of all the limited number of participants that took part in the study has decreased the power of the statistical analyses. 25% of the participants dropped out of the experiment before fully completing the questionnaire. In addition to this, a convenience sample of university students was used. Since student samples can significantly differ on various demographic variables from the rest of the population, the external validity of this study is negatively affected. Future research with larger sample sizes drawn from the general population could therefore provide more generalizable results.
Furthermore, theory stated that the effects of positive and negative EWOM on
consumer generated platforms and brand communities in specific would occur if an individual felt a strong connection to the community and identified with it. In addition, individuals that did not have a strong connection to the community would still act congruently with the
perceptions of the group to gain social acceptance (Chang et al., 2013). The community in this study, however, was not able to resemble the true community feeling and interaction of a
27
brand community. Because of this, the lack of connection of participants with the community and brand and the absence of social pressure from community members, can possibly have led participants not to act congruently with group norms. Since the group norms were not as evident as they would have been in a real-life setting. As a result, the external validity of the study has been compromised. Future research could therefore look at ways to capture the true community feeling and study responses towards EWOM of community members within this setting.
The current research made use of a for the participants unfamiliar brand. This was done to avoid the influence of existing attitudes and engagement on the results and find a pure effect of EWOM on consumer engagement. A limitation of this was, however, that
participants had difficulty answering in-depth questions about the unfamiliar brand based on only one EWOM message on a Facebook page. In future, research could replicate this study making use of a more familiar brand to see if this reveals different effects.
Moreover, the stimulus material in this study focused primarily on Facebook as a social media platform. The use of other social media channels might reveal other effects, because the distinction between consumer and company generated platforms is rather different on Twitter, online review sites or blogs. A direction for future research would therefore be to look into the effects of EWOM on consumer engagement on both company and consumer generated platforms and make comparisons between different social media channels.
Managerial implications
The present study has revealed that EWOM does play a part in the level of engagement consumers have with a brand. Therefore, it might be useful for brand managers to take EWOM into account in the building and maintenance of a strong brand. Creating an open online environment where consumers can comment and interact with the brand can increase
28
consumer engagement. Negative EWOM causes a stronger engagement and should therefore not be banned or deleted from social media pages. Brand managers do need to deal with negative EWOM, however, by developing effective webcare strategies. As previous research has indicated, it is better to respond to negative EWOM in a calm and reasonable way making use of a conversational human voice (Van Noort & Willemsen, 2011). Then, the negative information in the EWOM message can be countered with a positive response.
Moreover, based on the results it might be relevant for brands to implement EWOM marketing strategies. Making use of online platforms to stimulate and spread EWOM about the brand. By stimulating consumers to post and comment about the brand online engagement can be increased. The establishment of company generated customer service pages where consumers can share their problems or negative experiences with a brand and get a constructive response from the brand, could be a helpful tool in increasing consumer
engagement through negative EWOM. As the results have demonstrated EWOM on company generated platforms is considered more trustworthy. Therefore, the focus should be on these platforms. However, further research into the differences between consumer and company generated platforms is needed. What we do know at this point is that webcare on company generated platforms is considered much less intrusive than on consumer generated platforms (Van Noort & Willemsen, 2011). This means that stimulating negative EWOM on company generated platforms will be more effective, as it is easier to counter with webcare and it is considered more trustworthy.
Additionally, this study looked at the influence of source credibility on consumer engagement and found that a credible source can increase the consumer engagement with a brand. As a result brand managers should not try to write EWOM messages themselves to increase engagement. Since there is a commercial motive involved, the source might not be considered trustworthy, decreasing the credibility of the source. Of course the temptation to
29
write a negative EWOM message and counter it with a smart webcare strategy exists. The negative effects of being unmasked are more severe, however, than the amount of consumer engagement this strategy would gain. Since consumers prefer overt marketing above covert marketing strategies and finding out about the implementation of covert strategies afterwards causes them to see the brand more negatively than before (Aguirre et al., 2015). Using word of mouth marketing to stimulate consumers and specifically opinion leaders to share their product experiences would therefore be a more rewarding strategy, as this is the time where brands get consumers to play their game.
References
Aguirre, E., Mahr, D., Grewal, D., De Ruyter, K., & Wetzels, M. (2015). Unraveling the personalization paradox: The effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing, 91(1), 34-49. doi:10.1016/j.jretai.2014.09.005
Algesheimer, R., Borle, S., Dholakia, U.M., Singh, S.S. (2010). The impact of customer community participation on customer behaviors: An empirical investigation.
Marketing Science, 29(4), 756-769. doi:http:// dx.doi.org/10.1287/mksc.1090.0555
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (2002). Social cognitive theory of mass communication. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 121-154). Mahwah, NJ: Lawrence Erlbaum.
Bickart, B. & Schindler, R.M. (2001) Internet forums as influential sources of consumer information. Journal of Interactive Marketing, 15(3), 31-40. doi:10.1002/dir.1014
Blumler, J.G., & Katz, E. (1974). The uses of mass communications: Current perspectives on
30
Bronner, F., & De Hoog, R. (2010). Consumer-generated versus marketer-generated websites in consumer decision making. International Journal of Market Research, 52(2), 231-248. doi:http://dx.doi.org/10.2501/S1470785309201193
Brown, J., Broderick, A.J., & Lee, N. (2007). Word of mouth communication within online communities: Conceptualizing the online social network. Journal of Interactive
Marketing, 21(3), 2-20. doi:10.1002/dir.20082
Calder, B.J., Malthouse, E.C., & Schaedel, U. (2009). An experimental study of the relationship between online engagement and advertising effectiveness. Journal of
Interactive Marketing, 23(4), 321-331. doi:10.1016/j.intmar.2009.07.002
Chang, A., Hsieh, S.H., & Tseng, T.H. (2013). Online brand community response to negative brand events: The role of group eWOM. Internet Research, 23(4), 486-506.
doi:http://dx.doi.org/10.1108/IntR-06-2012-0107
Chevalier, J., & Mayzlin, D. (2006). The effect of word of mouth on sales: Online book reviews. Journal of Marketing Research, 43(3), 345-354. Retrieved from
http://www.jstor.org/stable/30162409
Conner, M., & Sparks, P. (2005). Theory of planned behaviour and health behaviour. In M. Conner & P. Norman (Eds.), Predicting health behaviour: Research and practice with
social cognition models (2nd ed., pp. 170-222). Maidenhead, UK: Open University
Press.
Court, D., Elzinga, D., Mulder, S., & Vetvik, O.J. (2009). The consumer decision journey.
McKinsey Quarterly, (3), 96-107. Retrieved from
mckinsey.com/insights/marketing_sales/the_consumer_decision_journey De Vries, N. J., & Carlson, J. (2014). Examining the drivers and brand performance
implications of customer engagement with brands in the social media environment.
31
Filieri, R. (2016). What makes an online consumer review trustworthy?. Annals of Tourism
Research, 58, 46-64. doi:10.1016/j.annals.2015.12.019
Fournier, S., & Avery, J. (2011). The uninvited brand. Business Horizons, 54(3), 193-207. doi:10.1016/j.bushor.2011.01.001
Hennig‐Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word‐of‐mouth via consumer‐opinion platforms: What motivates consumers to articulate themselves on the Internet? Journal of interactive marketing, 18(1), 38-52. doi:10.1002/dir.10073
Hennig-Thurau, T., Wiertz, C., & Feldhaus, F. (2015). Does Twitter matter? The impact of microblogging word of mouth on consumers’ adoption of new movies. Journal of the
Academy of Marketing Science, 43(3), 375-394. doi:10.1007/s11747-014-0388-3
Hollebeek, L.D., Glynn, M.S., & Brodie, R.J. (2014). Consumer brand engagement in social media: Conceptualization, scale development and validation. Journal of Interactive
Marketing, 28(2), 149-165. doi:10.1016/j.intmar.2013.12.002
Jahn, B., & Kunz, W. (2012). How to transform consumers into fans of your brand. Journal of
Service Management, 23(3), 344-361.
doi:http://dx.doi.org/10.1108/09564231211248444
Kietzmann, J.H., Hermkens, K., McCarthy, I.P., & Silvestre, B.S. (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business
Horizons, 54(3), 241-251. doi:10.1016/j.bushor.2011.01.005
King, R.A., Racherla, P., & Bush, V.D. (2014). What we know and don’t know about online word-of-mouth: A review and synthesis of the literature. Journal of Interactive
Marketing, 28(3), 167-183. doi:10.1016/j.intmar.2014.02.001
32
word-of-mouth communications: An attribution theory perspective. Journal of
Consumer Psychology, 11(1), 57-73. doi:10.1207/S15327663JCP1101_5
López, M., & Sicilia, M. (2013). How WOM marketing contributes to new product adoption.
European Journal of Marketing, 47(7), 1089-1114.
doi:http://dx.doi.org/10.1108/03090561311324228
López, M., & Sicilia, M. (2014). Determinants of E-WOM influence: The role of consumers' internet experience. Journal of Theoretical and Applied Electronic Commerce
Research, 9(1), 28-43. doi:http://dx.doi.org/10.4067/S0718-18762014000100004
Men, L.R., & Tsai, W.S. (2013). Beyond liking or following: Understanding public
engagement on social networking sites in China. Public Relations Review, 39(1), 13-22. doi:10.1016/j.pubrev.2012.09.013
Moran, G., & Muzellec, L. (2014). eWOM credibility on social networking sites: A framework. Journal of Marketing Communications, 1-13.
doi:10.1080/13527266.2014.969756
Nambisan, S., & Baron, R.A. (2007). Interactions in virtual customer environments: Implications for product support and customer relationship management. Journal of
Interactive Marketing, 21(2), 42-62. doi:10.1002/dir.20077
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness and attractiveness. Journal of Advertising, 19(3), 39-52. Retrieved from http://www.jstor.org/stable/4188769
O’Reilly, K., MacMillan, A., Mumuni, A.G., & Lancendorfer, K.M. (2016) Extending our understanding of eWOM impact: The role of source credibility and message relevance.
Journal of Internet Commerce, 15(2), 77-96. doi:10.1080/15332861.2016.1143215
33
purchasing intention: The moderating role of involvement. International Journal of
Electronic Commerce, 11(4), 125-148. doi:10.2753/JEC1086-4415110405
Pfeffer, J., Zorbach, T., & Carley, K. M. (2014). Understanding online firestorms: Negative word-of-mouth dynamics in social media networks. Journal of Marketing
Communications, 20(1-2), 117-128. doi:10.1080/13527266.2013.797778
Ranaweera, C., & Jayawardhena, C. (2014). Talk up or criticize? Customer responses to WOM about competitors during social interactions. Journal of Business Research,
67(12), 2645-2656. doi:10.1016/j.jbusres.2014.04.002
Sandes, F.S., & Urdan, A.T. (2013). Electronic word-of-mouth impacts
on consumer behavior: Exploratory and experimental studies. Journal of International
Consumer Marketing, 25(3), 181-197. doi:10.1080/08961530.2013.780850
Schamari, J., & Schaefers, T. (2015). Leaving the home turf: How brands can use webcare on consumer-generated platforms to increase positive consumer engagement. Journal of
Interactive Marketing, 30, 20-33. doi:10.1016/j.intmar.2014.12.001
Schweidel, D.A., & Moe, W.W. (2014). Listening in on social media: A joint model of sentiment and venue format choice. Journal of Marketing Research, 51(4), 387-402. doi:http://dx.doi.org/10.1509/jmr.12.0424
Sen, S., & Lerman, D. (2007). Why are you telling me this? An examination into negative consumer reviews on the web. Journal of Interactive Marketing, 21(4), 76-94. doi:10.1002/dir.20090
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin, & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks-Cole.
Tsao, W., & Hsieh, M. (2015). eWOM persuasiveness: do eWOM platforms and product type matter? Electronic Commerce Research, 15(4), 509-541.
34
doi:10.1007/s10660-015-9198-z
Van Doorn, J., Lemon, K.N., Mittal, V., Nass, S., Pick, D., Pirner, P., & Verhoef, P.C. (2010). Customer engagement behavior: Theoretical foundations and research directions.
Journal of Service Research, 13(3), 253-266. doi:10.1177/1094670510375599
Van Noort, G., Antheunis, M. L., & van Reijmersdal, E. A. (2012). Social connections and the persuasiveness of viral campaigns in social network sites: Persuasive intent as the underlying mechanism. Journal of Marketing Communications, 18(1), 39-53.
doi:10.1080/13527266.2011.620764
Van Noort, G., & Willemsen, L. M. (2012). Online damage control: The effects of proactive vs. reactive webcare interventions in consumer-generated and brand generated platforms. Journal of Interactive Marketing, 26(3), 131-140.
doi:10.1016/j.intmar.2011.07.001
Xia, L., & Bechwati, N.N. (2008). Word of mouse: The role of cognitive personalization in online consumer reviews. Journal of Interactive Advertising, 9(1), 3-13.
doi:10.1080/15252019.2008.10722143
Zaglia, M. E. (2013). Brand communities embedded in social networks. Journal of Business
35
Appendix
1. Stimulus material
Figure 2
36
Figure 3
37
Figure 4
38
Figure 5
39 2. Scale variables
Table 2
Source trustworthiness scale
Items Factor loadings
I think the source of the Facebook post is... Undependable-Dependable
.66
I think the source of the Facebook post is... Dishonest-Honest
.85
I think the source of the Facebook post is... Unreliable-Reliable
.89
I think the source of the Facebook post is... Insincere-Sincere
.87
I think the source of the Facebook post is... Untrustworthy-Trustworthy
.92
Based on a principal component factor analysis with varimax rotation, all items of the source trustworthiness scale loaded on one component with an Eigenvalue of 3.53. This scale explains 70.64% of the total variance and proved to be highly reliable (α = .89). Table 3
Source expertise scale
Items Factor loadings
In my opinion the source of the Facebook post is... Not an expert-Expert
.82
In my opinion the source of the Facebook post is... Inexperienced-Experienced
.87
In my opinion the source of the Facebook post is... Unknowledgeable-Knowledgeable
.90
In my opinion the source of the Facebook post is... Unqualified-Qualified
.88
In my opinion the source of the Facebook post is... Unskilled-Skilled
40
Using a principal component factor analysis with varimax rotation, it became clear that all items of the source expertise scale loaded on one component with an Eigenvalue of 3.81. This scale explained 76.27% of the total variance and is highly reliable (α = .92).
Table 4
Cognitive consumer engagement scale
Items Factor loadings
Using True Fruits will get me to think about True Fruits
.83
I would think about True Fruits a lot when I'm using it
.85
Using True Fruits would stimulate my interest to learn more about True Fruits
.78
A principal component factor analysis with varimax rotation, showed that all the items of the cognitive consumer engagement scale loaded on one component with an Eigenvalue of 2.01. The scale explains 67.08% of the total variance and is considered reliable (α = .75).
Table 5
Affective consumer engagement scale
Items Factor loadings
I would feel very positive when I use True Fruits
.91
Using True Fruits would make me happy .91
I would feel good when using True Fruits .92
I would be proud to use True Fruits .80
Based on a principal component factor analysis with varimax rotation, the items of the affective consumer engagement scale all loaded on one component with an Eigenvalue of 3.13. This scale is highly reliable (α = .90) and explains 78.28% of the total variance.
41
Table 6
Active consumer engagement scale
Items Factor loadings
I would spend a lot of time using True Fruits compared to other brands
.81
Whenever I would be having a smoothie, I would use True Fruits
.90
True Fruits is one of the brands I would use when I have a smoothie
.88
With an Eigenvalue of 2.24, all the items of the active consumer engagement scale proved to load on one component. This finding was based on a principal component factor analysis with varimax rotation. The scale is highly reliable (α = .83) and explains 74.81% of the total variance.
Table 7
Attitude scale
Items Factor loadings
I think True Fruits is... Bad-Good .89
I think True Fruits is... Unfavorable-Favorable
.92
I think True Fruits is...Unappealing-Appealing
.87
With an Eigenvalue of 2.39, all three items of the attitude scale loaded on one component. These findings were discovered using a principal component factor analysis with varimax rotation. This scale proved to be highly reliable (α = .87) and explains 79.51% of the total variance.
42
Table 8
Purchase intention scale
Items Factor loadings
The next time I need a smoothie, I will choose True Fruits
.92
If I had needed a smoothie during the past year, I would have chosen True Fruits
.91
In the next year if I need a smoothie, I will select True Fruits
.94
A principal component factor analysis with varimax rotation proved that all the items of the purchase intention scale loaded on one component with an Eigenvalue of 2.56. The scale is highly reliable (α = .91) and is able to explain 85.39% of the total variance.
