The effect of using electronic-word-of-mouth
on customer-based brand equity:
Do attitudinal brand loyalty and homophily
matter?
by
Annabel Annen
2
The effect of using electronic-word-of-mouth
on customer-based brand equity:
Do attitudinal brand loyalty and homophily
matter?
Groningen, 17th of June 2013
Master thesis
University of Groningen, the Netherlands Faculty of Economics and Business Master Marketing Management
Annabel Annen
Nieuwe Ebbingestraat 49b
9712 NE Groningen, the Netherlands +31(0)627530461
annabelannen@live.nl student number: 2059738
Supervision
University of Groningen, Faculty of Economics and Business, Department of Marketing First supervisor: dr. P. C. (Peter) Verhoef, Professor of Marketing
3
MANAGEMENT SUMMARY
In the recent past, campaigns with online involvement of customers did not exist. At that time product and brand related information was mainly spread by companies and between customers only offline interactions existed (word-of-mouth). In contrast, nowadays customers not only interact with companies, but by using social media also with other customers.
The effects of using eWOM on the level of brand equity are investigated in this paper. Strong equity brands are valuable for a company because they lower risks of a company. The level of brand equity a customer perceives may depend on the positive or negative content in an eWOM. Therefore, the main research question is:
“Does using positive or negative electronic-word-of-mouth (eWOM) affect customer-based brand equity (CBBE)?
This study further investigates the moderating role of attitudinal brand loyalty and the role of homophily on our main effect.
The research question was answered by using four hypotheses. First, the main research question was answered by investigating whether we can assume that brand equity is higher for customers receiving positive eWOM than for customers receiving negative eWOM (H1). Subsequently, the first moderator was investigated by checking whether the effect of eWOM on brand equity will be reduced by attitudinal brand loyalty (H2). Thereafter, it was analysed whether attitudinal brand loyalty is
significantly higher for Apple than for Nokia, in order to associate Apple correctly as a high loyalty brand and Nokia as a low loyalty brand. In addition, the second moderator effect was investigated by analysing whether the effect of eWOM on brand equity will be increased by homophily (H3).
In total, 236 respondents participated in the questionnaire (of which mainly young adults). The
questionnaire included a 2 (exposure to a positive review vs. negative review) x 2 (high attitudinal brand loyalty vs. low attitudinal brand loyalty) x 2 (high homophily vs. low homophily) experimental design. Subjects were exposed to either a positive eWOM review about a smartphone brand, or a negative eWOM review about a smartphone brand. For the manipulation of the attitudinal brand loyalty
construct, Apple and Nokia were taken as brands. Homophily was manipulated by explicitly quoting the occupation of the writer of the positive or negative eWOM review (student vs. working person) and comparing this with the occupation of the respondent.
4 The main recommendation that result out of these findings is that especially brands that customers have low attitudinal brand loyal feelings towards should concern about the impact of negative eWOM, since these brands do not benefit from any protection resulting from the strong influence of attitudinal brand loyalty on brand equity.
The fact that the hypotheses in this study are not significant might be caused by several limitations that will be discussed here briefly. First we used snowball sampling what did not lead to a good reflection of the population.
Second, the fact that the hypotheses did not show significant outcomes might be caused by the choice of Apple and Nokia as brands for the manipulation of the attitudinal brand loyalty construct.
Respondents have a clear en deeply held opinion towards these brands, which is not easy to change by means of just one short eWOM review with positive or negative valence. In future studies it is preferable to use newer or less well-known brands, to make sure that respondents do not have strong and hardly changeable believes and attitudes.
A third limitation is that the effect of just one short eWOM review could be questioned. Credibility and persuasiveness would have been increased when the volume of the eWOM reviews that respondents were exposed to were higher (Yang, Kim, Amblee and Jeong, 2012).
A fourth limitation is that in the questionnaire no manipulation was included that checked whether the eWOM was read by the respondent, and therefore we can only assume the respondent did read the review.
A fifth limitation is the fact that in this study homophily was solely based on similarities in the
occupation (student vs. working) of the writer of the eWOM review and the respondent. Several other factors might also influence whether the respondent who reads the review perceives homophilic ties towards the writer of the review, such as ethnicity, education, lifestyle and social or professional status (e.g. Rogers, 1983; Brown, Broderick and Lee, 2007).
5
PREFACE
The past six months as a student are comparable to me as being a fulltime employee. For the first time in my student period, I had a lot of structure in my life. My friends knew that they could find me in the University Library on a daily base and that I was not easy to persuade for some drinks in a pub. Just because I had an important job to fulfill - that was, stick to the thesis deadlines. After a period of five months hard work and discipline, I accomplished my goal by handing in this piece of paperwork. It has been said that writing a thesis forces you into live in a social isolation. However, I did not perceive it that way. Although I was in the University Library 24/7, the lunch and coffee breaks in the canteen were very fun and pleasant.
This thesis as presented in front of you is of great value for me. Not only because it is an achievement of my master Marketing, but also because it symbolizes the end of my period as being a student.
Without the help of certain persons, writing this thesis did not go so smoothly.
First I would like to thank my supervisor from the University of Groningen, dr. Peter Verhoef, for his support. What I liked about his way of supervision was that he was really clear and direct in presenting his comments. When something was not sufficient, he did not keep up appearances. Therefore, you always knew what to expect. I further liked that when he had some critics he also came with suggestions for improvements, which gave me new and fresh energy to proceed.
Second I would like to thank my parents for their support and ongoing faith in me. They always have been able to put their interests aside to let me succeed in everything I wanted to pursue.
Third, I would like to thank my uncle for his interest in and his critical view on my thesis. I really
appreciated that he was always available to answer my pressing questions and to give me new insights. Furthermore, I would like to thank my brother, because when I started to panic, he listened to my story and spoke some encouraging words to cheer me up.
At last, I appreciated the help of my friends, since they all did their best to get me enough respondents within a very short time by spreading the questionnaire among their friends, colleagues and on
Facebook. I also valued the spare time with them. Meet with my friends was a good remedy for me to relieve some stress and to recharge.
I realized that by handing in my thesis, my years as a student have come to an end. I have had a great time here in Groningen, I met wonderful people, experienced and learned a lot and I enjoyed the student life fully. Whatever the future will bring, I will always remember and refer to this time of my life in a positive way.
Thanks to you all,
6
TABLE OF CONTEN TS
1. Introduction p. 8
2. Theoretical framework p. 11
2.1 Customer Engagement Behavior (CEB) p. 11 2.1.1 Word-of-mouth and electronic-word-of-mouth (eWOM) p. 12 2.2 Customer-based brand equity (CBBE) p. 12
2.3 Conceptual model p. 13
2.4 eWOM and its potential effects on customer-based brand equity p. 14 2.5 Attitudinal brand loyalty and its potential moderating effects p. 14 2.6 Homophily and its potential moderating effects p. 16
3. Research method p. 18
3.1 Experimental design p. 18
3.2 Data collection and sample p. 18
3.3 Measuring variables p. 19
4. Results p. 23
4.1 Demographical findings p. 23
7
5. Conclusion and recommendations p. 29
5.1 Discussion of the findings p. 29 5.2 Limitations and directions for further research p. 30
References p. 33
8
1. INTRODUCTION
Would it not be great to compose your own hamburger for McDonald’s restaurants with your own favorite ingredients? This idea became reality when McDonald’s restaurants introduced the “My Burger” campaign in Germany last year. In this crowdsourcing project, McDonald’s restaurants set up a platform to engage customers in the process of product development –that is, designing their own favorite hamburger. Customers could compile their custom-made hamburger out of more than 70 different ingredients and they could name their hamburger afterwards. As a result, customers were creating a tremendous buzz for McDonald’s restaurants by promoting their favorite hamburger on social networking sites with the aim of generating votes. Out of the ten finalists, the five best hamburgers were selected to be sold in 1,415 McDonald’s restaurants in Germany for a limited period of time. The success of this crowdsourcing campaign became visible in the fact that more than 333,000 Germans have designed a hamburger and that more than 5,000,000 Germans voted online for one of the
hamburgers. This example is a great success story of involving customers into the product development process.
In the recent past, these kinds of campaigns with online involvement of customers did not exist. At that time product and brand related information was mainly spread by companies and between customers only offline interactions existed (word-of-mouth). The recent growth of consumer-generated media (CGM) has changed the unidirectional interaction from companies to customers into a bidirectional interaction between customers and companies (Onishi and Manchanda, 2012). Customers not only interact with companies, but by using social media also with other customers and these interactions are bidirectional in nature. According to Libai et al. (2010), customer-to-customer (C2C) interaction can be defined as “the transfer of information from one customer (or a group of customers) to another customer (or a group of customers) in a way that has the potential to change their preferences, actual purchase behavior, or the way they further interact with others.”
9 The rapid growth of customer engagement behaviors shows that customers have become serious partners for companies and brands to look at, since customers are not only “consumers” anymore, but nowadays they have also acquired the role of “producer” (of content), “distributor” (e.g. of reviews) and the role of “retailer” (e.g. on eBay). As Leeflang (2011) has stated in his paper, “firms no longer control marketing, but rather customers define what a company is (and is not).”
The collection and usage of information resulting from customer engagement behaviors might affect customers’ opinions towards the brand or the company. Some researchers (e.g. Chen, Wang and Xie, 2011; Godes et al., 2005) have argued that customer choice is affected in a direct and meaningful way by the opinions or actual purchase decisions of other customers. However, the direction of this effect may depend on whether these CEBs are positive or negative in nature. Therefore, in this research we will focus on the effect of positive and negative CEBs on customers’ opinions.
CEB is a broad construct (Van Doorn et al., 2010). This study is focused on one form of CEB –that is, electronic-word-of-mouth (eWOM). This relatively new type of communication enables customers to share product or brand experiences and opinions (e.g. recommendations, complaints, advices) with multiple other customers at once (e.g. Gruen et al, 2006; Hennig-Thurau et al., 2004). Customers can not only generate eWOM, but they can also collect and use the positive or negative content of the eWOM for making product decisions. This paper focuses on the effects of the latter use of eWOM- that is, the effect of using information that is provided online by other customers. Given the distinct potential effects of internet communication (directed to multiple individuals and available to other customers for an indefinite period of time), the effects of eWOM deserve the serious attention of marketing
researchers and managers (Hennig-Thurau et al., 2004). Since internet communication is directed to multiple individuals and available for an indefinite period of time, it can be expected that internet communication significantly influences product adoption, purchase decisions as well as post-product perceptions.
This study is aimed at investigating the effects of using eWOM on the level of brand equity. Strong equity brands are valuable for a company because they lower risks of a company. In this paper, brand equity will be viewed from a customer mind-set. Therefore, we will focus on customer-based brand equity (CBBE). The level of brand equity may depend on the positive or negative content of information in the eWOM. Therefore, the main research question is:
“Does using positive or negative electronic-word-of-mouth (eWOM) affect customer-based brand equity (CBBE)?
10 these underlying feelings and emotions. The experiment is aimed at investigating the moderating role of attitudinal brand loyalty on our main effect. Therefore, the sub-question in this study is:
“To what extent is attitudinal loyalty to the brand or firm moderating the relationship between using electronic-word-of-mouth (eWOM) and customer-based brand equity (CBBE)?”
Furthermore, the strength of eWOM may depend on the degree of homophily between the reviewer and the reader of the eWOM. Prior research has indicated that customers attach more value to a review if the reviewer and reader of the review have similar characteristics. Therefore, the simulated review contains a textbox in which is noted whether the reviewer is a student or a working person. Ties are defined as homophilic if they are matching. Since homophily might impact the relationship of eWOM on brand equity, the second sub-question is:
“To what extent is homophily moderating the relationship between using electronic-word-of-mouth (eWOM) and customer-based brand equity (CBBE)?”
The research questions in this paper will be exposed to a literature study and a field research in the form of an experiment as well. The experiment will take place according to a 2x2x2 experimental design (see Table 1).
Table 1: The 2x2x2 experimental design used in this study
Positive eWOM Negative eWOM
Apple Nokia Apple Nokia
High homophily A C E G
Low homophily B D F H
The paper is organized as follows: in the theoretical framework –chapter 2- the key definitions used for this research are explained and prior literature related to this topic will be discussed. The hypotheses and conceptual model are clarified in this chapter as well. Thereafter, the research method is covered in chapter 3, including the experimental design, data collection method, sample and measuring variables. In chapter 4, results of the analyses are discussed and chapter 5 is left for conclusions,
11
2. THEORETICAL FRAMEWORK
This chapter will cover the theoretical foundation that serves as the basis for answering the research questions. In paragraph 2.1 the overall concept “customer engagement behaviour” is subject to a literature study. Since this research focuses solely on one particular part of CEB, namely “electronic-word-of-mouth”, in 2.1.1 a special paragraph is devoted to describe this construct. Then, the dependent variable “customer-based brand equity” will be covered in paragraph 2.2. Hereafter, the aim of this research is visualized in a conceptual model in paragraph 2.3. Subsequently, the potential effects of electronic-word-of-mouth on brand equity are discussed in paragraph 2.4. Next, the “attitudinal brand loyalty” construct is discussed in paragraph 2.5 and literature on “homophily” can be found in paragraph 2.6.
2.1 Customer Engagement Behavior (CEB)
This research is about customer engagement behavior, CEB in this paper. This relatively new term was initially introduced in the paper of Van Doorn et al. (2010), and is a compilation of several phenomena that also have been studied in the past, mainly as customer-to-customer interactions (C2C). One of the key elements of CEB is that it is about “interactions between customers”. C2C-interactions can occur through observational learning as well as through verbal communications (Libai et al., 2010). By verbal communications, word-of-mouth (WOM) is meant. Word-of-mouth has always been interesting for managers to understand, because it is often an important driver of customer behavior such as the adoption of a new technology, the decision to watch a television show, or the choice of which laptop to purchase. It might affect awareness in some cases, or preferences in others (Godes and Mayzlin, 2004). Although the fact that word-of-mouth in history took place solely in the offline environment, nowadays word-of-mouth has gained an important role in the online environment as well due to the more central position of internet in customers’ daily life. Since word-of-mouth is an integral part of C2C-interactions, the interactions between customers have broadened onto the online environment. This corresponds to the paper of Libai et al. (2010), in which it is stated that C2C- interactions have both offline and online dimensions. Electronic word-of-mouth (eWOM) is not the only way in which customers are able to communicate online. Other online communication possibilities are among others blogging, voting, co-creation and YouTube videos.
In the CEB definition, also the “engagement” dimension is incorporated. Engagement takes places in different settings, among others in media, brand, customer, employee, organization and product settings. A good definition of engagement to start with comes from Calder and Malthouse (2008), who distinguish media engagement from “mere linking”, implying that engagement is a stronger state of connectedness between the customer and the media than liking alone.
12 alignment with Van Doorn et al. (2010) its definition of customer engagement, similarities and overlap in definitions can be found in other literature. Practically all definitions of engagement constitute elements of “interactive experiences”, “beyond mere involvement”, “motivational drivers” and “behavioral outcomes” (e.g. Brodie et al., 2011; Higgins, 2006; Hollebeek, 2011; Mollen and Wilson, 2010). Another point worthy to note is that according to the paper of Hollebeek (2011), engagement is thought to reflect a process in which engagement intensity may develop over time. This implicates that the level of engagement is not stabile but can fluctuate when exposed to certain factors, such as for example eWOM.
2.1.1 Word-of-mouth and electronic- word-of-mouth (eWOM)
Although CEBs can occur in several forms (e.g. voting or co-creation), this research focuses solely on one particular engagement manifestation- that is, electronic word-of-mouth (eWOM). eWOM is a derivative of the initial concept of word-of-mouth that takes place in an offline environment through verbal communications. Many researchers have found that word-of-mouth has a significant impact on product attitudes and consumer choice (e.g. Engel et al., 1969; Brown, Broderick, & Lee, 2007). In addition, prior research also found that word-of-mouth (both verbal and electronic) is more effective and persuasive than traditional mass media, such as personal selling and various types of advertising (e.g. Bickart and Schindler, 2001; Richins, 1983).
Word-of-mouth can have both positive and negative consequences for a firm. In general, positive word-of-mouth can be expressed in customers’ willingness to recommend products to others (Gruen et al., 2006). Following this way of thinking, negative word-of-mouth is customers’ willingness to discommend products to others.
The internet enables customers to engage in eWOM by sharing their opinions on, and experiences with goods and services with a multitude of other customers. In this study, we will refer to eWOM as described the paper of Gruen et al. (2006), who define eWOM communication as “any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet.”
We suggest that eWOM with positive valence is likely to have more positive outcomes for the firm than eWOM with negative valence, which is likely to have more negative outcomes for the firm.
2.2 Customer-based brand equity (CBBE)
This paper investigates whether using positive or negative information in eWOM has an impact on the equity of the brand or the company the eWOM was about. To investigate this effect, we first have to clarify a proper definition of brand equity.
13 some strong, favorable and unique brand associations in memory. These associations include product-related or non-product-product-related attributes; functional, experiential, or symbolic benefits. Reiterated by Stahl et al., (2012), it is imperative that any customer mind-set measure of equity include both
awareness/ familiarity and brand associations. Hence, a brand has positive CBBE when customers react more favorably to a product and the way it is marketed when the brand is identified than when it is not. These reactions towards a brand having strong equity result in generating “proper cognitive appraisals and emotional reactions” towards that brand (Keller, 2008). Defining CBBE according the customer mind-set and based on customers’ memory-based brand associations have been taken up in many scientific papers (e.g. Aaker, 1993; Christodoulides et al., 2012; Cobb-Walgren et al., 1995; Dawar and Pillutla, 2000; Stahl et al., 2012; Pappu et al., 2005; Rego et al., 2009; Roehm; Brady, 2007).
From now on we will refer to this CCBE-construct simply by the term brand equity. For companies, building and maintaining strong brand equity is very important, since the benefits of brand equity for firms are amongst others that a strong brand will lead to higher perceived quality, higher customer preferences, higher purchase intentions, lower customer price sensitivity, less vulnerability to
competitive actions, greater trade support, larger margins and it facilitates repeat-purchasing behavior (e.g. Aaker and Jacobson, 1994; Cobb-Walgren et al., 1995; Keller, 2008; Rego et al., 2009).
2.3 Conceptual model
In this section both the main and moderator effects of this research are clarified. As can be seen in figure 1, customer-based brand equity is the dependent variable in this study. Positive electronic-word-of-mouth and negative electronic-word-electronic-word-of-mouth are the independent variables. The effect of these independent variables on customer-based brand equity will be studied in hypotheses 1.
14 Figure 1: Conceptual model used in this study
2.4 eWOM and its potential effects on customer-based brand equity
As mentioned, eWOM is a manifestation of customer engagement and it may have consequences for brand equity. Whether these consequences are positive or negative might depend on the (positive or negative) valence of the eWOM. When returning to the general customer engagement topic, positive customer engagement includes those actions that in the short and long run have positive consequences – financial and non-financial for the firm. CEBs with positive valence might evoke positive consumption emotions in the customer’s brain such as hope, happiness, joy and surprise (Verhoef and Lemon, 2013). These emotions might increase the equity customers attach to the brand. Since eWOM is an expression of CEB, we suggest that eWOM with positive valence will result in higher levels of brand equity.
In the same study, Verhoef and Lemon (2013) posit that negative CEBs raise emotions that include aspects of anger, depression and guilt. These emotions might decrease the equity customers attach to the brand. Therefore, we suggest that eWOM with negative valence will result in lower levels of brand equity. Hence, we suggest that:
H1: Brand equity is higher for customers receiving positive eWOM than for customers receiving negative eWOM.
2.5 Attitudinal brand loyalty and its potential moderating effects
15 well as attitudinal components should be taken into consideration when studying the concept.
According to Kumar and Reinartz (2006), behavioral loyalty refers to “the observed action customers have demonstrated towards a particular product or service. Attitudinal loyalty, by contrast, refers to the perceptions and attitudes a customer has toward a particular product or service”. Although behavioral loyalty may be a result of attitudinal loyalty, purchases may be driven by other factors as well.
Therefore, the attitudinal aspects of loyalty might be even more important. Many researchers have studied loyalty as a combination of behavioral and attitudinal loyalty, which can be referred to as “true loyalty” (e.g. Day, 1969; Dick and Basu, 1994; Iglesias et al., 2011; Kumar and Reinartz, 2006). In addition, many researchers also focused mainly on the attitudinal dimension of brand loyalty (e.g. Chaudhuri and Holbrook, 2001; Oliver, 1999). Oliver (1999) describes attitudinal loyalty as “a deeply held commitment to rebuy or repatronize a preferred product/ service consistently in the future, thereby causing repetitive same-brand or same brand-set purchasing, despite situational influences and marketing efforts having the potential to cause switching behavior”. In this research we will focus solely on attitudinal loyalty.
Companies are eager to create loyal customers, since brand loyalty may result in several advantages for a company. In the framework of Dick and Basu (1994), three main consequences resulting from
customer loyalty are described: (1) search motivation, (2) resistance to counter persuasion and (3) word-of-mouth. In addition, brand loyalty can have several advantages such as customers’ willing to pay a premium (Chaudhuri and Holbrook, 2001), higher purchase frequency acceleration, higher share of wallet, increased retention and increased lifetime duration (Keller, 2008).
Now the concept of attitudinal brand loyalty is clear, we explain how to apply this construct in this paper. Attitudinal brand loyalty will serve as a moderator in this research. In that perspective, we suggest that brand loyalty has an impact on the relationship between electronic-word-of-mouth and customer-based brand equity. Prior research claimed that loyalty is strongly related to commitment, satisfaction, brand affect and trust (e.g. Ahluwalia et al., 2000; Oliver, 1999; Chaudhuri and Holbrook, 2001). Since loyalty is related to commitment, implications that hold for this construct may also hold for the loyalty construct. Research of Ahluwalia et al. (2000) found that the response patterns of high- and low-committed customers to positive and negative information are very different. They found that the impact of positive information is enhanced when customers are high-committed, compared to when customers are low-committed. Taking this into account, we suppose that the effect of positive eWOM on brand equity is higher when a customer is brand loyal, compared to when a customer is not brand loyal. However, we assume that beyond a certain point, a ceiling-effect will occur for brand loyal customers who are exposed to positive eWOM, since you cannot become more delighted if you already have reached the maximum level. Beyond that point, the increase of the effect will be reduced when a customer is brand loyal, compared to when a customer is not brand loyal. Concluding from this, we suggest this moderating effect to be non-linear (concave).
16 customers will not strongly adopt negative eWOM into their evaluation. This results in the following hypothesis:
H2: The effect of eWOM on brand equity will be reduced by attitudinal brand loyalty. This paper focuses on the telephone branch in which several smartphone brands can be detected. To assess any difference in attitudinal brand loyalty, Apple and Nokia are taken as brands along which this construct is measured. We assume Apple to have a majority of highly attitudinal loyal customers. In contrast, we assume Nokia to have a minority of highly attitudinal loyal customers. Hence, we assume that attitudinal brand loyalty is significantly higher for Apple than for Nokia. This assumption is tested in our study by means of a manipulation check.
2.6 Homophily and its potential moderating effects
As can be seen in figure 1, “homophily” serves as the second moderator in this research.
Homophily is studied many times in marketing and psychology literature. One good definition to start with comes from Rogers (1983), who stated that homophily explains group composition in terms of the similarity of members’ characteristics: the extent to which pairs of individuals are similar in terms of certain attributes, such as age, gender, education, or lifestyle. A more abstract definition comes from a study of McPherson et al. (2001), in which is stated that homophily is the principle that a contact between similar people occurs at a higher rate than among dissimilar people. Or simply said by Braun and Bonfrer (2011), homophily infers that people who are similar to one another are more likely to interact with one another. Other authors who have formed a somewhat similar definition of homophily are e.g. Brown and Reingen, 1987; Galbreth et al., 2007; Nitzan and Libai, 2011 and Yang and Allenby, 2003.
In prior research (e.g. Brown and Reingen, 1987; Nitzan and Libai, 2011), homophily is measured in terms of the percentage of similar characteristics a pair of connected customers share. People who share a lot of similar demographic profiles, can be identified as “demographic neighbors”, as introduced by Yang and Allenby (2003).
Many different customer characteristics might drive homophily. Lazarsfeld & Merton (1954) described two types of homophily based on these characteristics. These authors distinguish (1) status homophily, in which similarity is based on informal, formal, or ascribed status, from (2) value homophily, which is based on values, attitudes, and beliefs. In this research, homophily will be measured based on one customer characteristic, –that is, a distinction between “employee” and “student”. This is linked to “status homophily”, as identified by Lazarsfeld & Merton (1954).
17 objectively assess similarity based on distinguishable characteristics such as looks and appearance. When not any cues are available, customers have to evaluate eWOM persuasiveness solely on content characteristics, shared group interests and group mind-set (Brown, Broderick and Lee, 2007; Park and Min Lee, 2009). As mentioned, in the experimental phase of this study, we simulated a review that includes whether the reviewer is an “employee” or a “student”. By providing this cue, it is easier for customers to identify some kind of homophily with the writer of the review.
There can be discerned several benefits and drawbacks of homophily in the literature. First benefit is that the complexity of an individual’s thought process during reasoning can be reduced by homophily, as opposed to situations where it is hard to determine audiences’ characteristics and opinions, in which case a more complex elaboration is needed to create a message that emphasizes argument quality (Nekmat, 2012). Another benefit is that customers are more likely to trust the endorsements of people whose preferences they share (Feick and Higie, 1992). Moreover, participating in homophilic groups leads to enhanced profits for the firm (Galbreth et al., 2012). A drawback is that homophily limits people’s social worlds in a way that has powerful implications for the information they receive, the attitudes they form, and the interactions they experience (McPherson & Smith-Lovin, 1987).
In this study will be investigated whether homophily affects the relationship between eWOM and brand equity. People interact with each other when forming opinions and preferences towards brands and products. Preferences and choice behavior are therefore influenced by customer’s own tastes and the tastes of others. People who identify with a particular group often adopt the preferences of that group, which results in interdependent choices (Yang and Allenby, 2003). Narayanan and Nair (2013) support this finding by suggesting that homophily cause people to behave similarly in product adoption behavior. This suggests that the likelihood of imitation is greater among people who show homophilic ties. Rogers (1983) indicated that homophilic ties are perceived as more credible than heterophilic ones, suggesting that homophilic sources can be perceived as more influential. Brown et al. (2007)
extrapolate this finding by concluding that homophily in online word-of-mouth influences customer’s decision making and attitude formation.
Concluding from the above, we assume homophily to affect the relationship between eWOM and brand equity. To test this moderation effect, we have compiled the following hypothesis:
18
3. RESEARCH METHOD
In this chapter, the data collection method will be discussed. First, practical issues concerning the type of questions, sample size and the distribution of the questionnaire are described in paragraph 3.1. Thereafter, the questionnaire design is discussed in more depth in paragraph 3.2 by discussing all manipulations that are devised to obtain the results. Subsequently, paragraph 3.3 discusses the way in which each construct will be measured by referring to other authors’ scales that are used in this study. 3.1 Experimental design
In this study we investigated the effects of positive eWOM and negative eWOM on brand equity and we investigated the moderating role of attitudinal brand loyalty and homophily. A 2 (exposure to a positive review vs. negative review) x 2 (high attitudinal brand loyalty vs. low attitudinal brand loyalty) x 2 (high homophily vs. low homophily) experimental design was used. Subjects were exposed to either a positive eWOM review about a smartphone brand, or a negative eWOM review about a smartphone brand. For the manipulation of the attitudinal brand loyalty construct, Apple and Nokia were taken as brands. Apple is supposed to have a majority of highly attitudinal loyal customers, –that are customers with strong empathetic feelings towards Apple. In contrast, Nokia is supposed to be a brand that has a minority of highly attitudinal brand loyal customers. Whether these assumptions about the brand loyalty of Apple and Nokia are true were also measured by the questions about ‘attitudinal brand loyalty’ in the questionnaire as a means of a manipulation check.
Homophily was manipulated by explicitly quoting the ‘occupation’ of the writer of the positive or negative eWOM review (student vs. working person). In the questionnaire also the occupational status of the respondent was assessed. In the case of equality (a student reads a review of a student or a working person reads a review of a working person), homophily is assumed to be high. In the case of inequality (a student reads a review of a working person or a working person reads a review of a
student), homophily is assumed to be low. The experimental design used in this study was realized using eight versions (A to H) of the questionnaire (Table 2).
Table 2: the 2 x 2 x 2 experimental design used in this study
Positive eWOM Negative eWOM
Apple Nokia Apple Nokia
High homophily A C E G
Low homophily B D F H
3.2 Data collection and sample
19 to participate in this study to other people they know. Moreover, a link to the questionnaire was put on Facebook, so that Facebook-friends did get the possibility to share the link with their friends on
Facebook or Twitter and so on. To keep the response rate high, the questionnaire did take no longer than five minutes to complete. In total, 236 respondents completed the questionnaire. Based on the questionnaire, respondents were divided in a high homophily group and a low homophily group. Number of respondents in each cell of the design is shown in Table 3.
Table 3: Number of respondents in the experimental design used in this study Positive eWOM Negative eWOM
Apple Nokia Apple Nokia
Student 25 (A) 23 (C) 28 (E) 61 (G) Working person 24 (B) 25 (D) 25 (F) 25 (H)
Table 3 shows that all the versions of the questionnaire were completed by a sufficient number of respondents. However, version G was filled in by more respondents. This was caused by the snowball effect of the Facebook recruitment.
3.3 Measuring variables
The questionnaires assessed the following variables. Attitudinal brand loyalty
Four items were used to assess attitudinal brand loyalty to Apple and the same four items were used to assess attitudinal brand loyalty to Nokia. Items were derived from Bearden and Netemeyer (2011). Initially, these items were designed by Thomson, MacInnis and Park (2005). In this study, we used the items “loved”, “delighted”, “connected” and “bonded”, that were translated in Dutch. Respondents were given the following instruction: ‘think for about ten seconds about the brand Apple (and to Nokia thereafter): Subsequently indicate to what extent the following words describe your feelings to Apple (or Nokia)’. Items could be answered on a 7-point bipolar scale (1=describes poorly, 7= describes very well). The reliability of the brand loyalty scale Apple was Alpha=.89. The reliability of the brand loyalty scale of Nokia was Alpha=.81. To hide the intention of the research, respondents also filled in the loyalty scale for HTC (Alpha=.92) and the brand loyalty scale for Samsung (Alpha=.93).
Brand equity
20 brand’ (1=disagree, 7=agree). The reliability of brand equity for Apple was Alpha=.73 and the reliability of brand equity for Nokia was Alpha=.77.
Subjective price balance
To assess the subjective price balance respondents were asked whether the price setting of Apple, Nokia, HTC and Samsung was 1=too low, 7=too high. In the analysis, scores of 1, 2 and 3 were used as indications of under priced products, and scores of 4 and 5 were used as an indication of fairly priced products and scores of 6 and 7 were used as indications of overpriced products.
Purchase decision
To assess the effects of the positive and negative eWOM on purchase decision, respondents were provided a decisional task. Respondents were given the following instruction: ‘Imagine you intend to buy a new smartphone subscription. You are interested in the following offer: 200 minutes, 500 MB internet, unlimited SMS and a two-year contract period. Study the following four offers and then indicate your preference’.
After the information was provided, respondents were asked to indicate their preference for one of the four offers. The offers are shown in figure 2.
Figure 2: Four smartphone offers respondents were exposed to in the questionnaire
21 Demographic variables
Respondents were asked about their principal daily activities (student, work, other). These data were used to assess the level of homophily with the reviewer. Further, respondents indicated their gender and age.
Table 4 visualizes how each construct is measured:
Table 4: construct, source and measurement of variables used in the questionnaire Construct Sources Measures and scales Attitudinal brand loyalty Consumers’ emotional
attachments to brands (Bearden and Netemeyer, 2011)
Bipolar scales (1= describes poorly and 7= describes very well)
Loved Delighted Connected Bonded
Question 1 - 4; 6 - 9; 11 – 14; 16 – 19 Brand equity (1) Quality and preference (Sloot,
Verhoef and Franses, 2005)
Bipolar scales (1= very low and 7= very high)
To my opinion, the quality of products of brand X are
To my opinion, my preference for products of brand X are
Question 12 - 13 Brand equity (2) (Verhoef, Langerak and
Donkers, 2007)
Bipolar scales (1= disagree and 7= agree)
Brand X is a strong brand Brand X is a well-known brand Brand X is an attractive brand Brand X is a unique brand Question 22 - 27
Identification Bipolar scales (1= did not identify at all and 7=did really identify)
22 Subjective price balance Bipolar scales (1= too low and 7= too high)
To my opinion, the price balance of brand X is
Question 5; 10; 15; 20
Purchase decision Multiple choice
A: Nokia Lumia 900 B: HTC One S C: Apple iPhone 5
D: Samsung Galaxy S3 i9300 blue Question 28
Homophily Dichotomous question
Occupation [student/working/ other]
Question 29
Demographic variables Dichotomous question and open question Gender [male/ female]
23
4. RESULTS
In this chapter the outcomes of the questionnaire will be discussed. First some demographics are shown and thereafter each research topic will be discussed in a separate paragraph.
4.1 Demographical findings
After collecting 236 respondents, data analyses were performed in SPSS. First some demographics were computed to get a proper overview of the data. These results are shown in Table 5.
Table 5: Demographic characteristics
n % year Gender Male 89 37.9 Female 146 62.1 Age (mean) 25.7 Occupation Student Working Other 154 77 5 65.3 32.6 2.1
As can be seen in Table 5, the sample in this study consists out of 89 male respondents and 146 female respondents. One respondent did not note the gender question. The range of the respondents lies between 14 and 59 year. The most noted age was 21 years old (36 respondents). 78.0% of the respondents were between 17 and 27 years old.
4.2 Results of eWOM on brand equity
In order to answer hypothesis 1, it is tested whether brand equity is influenced by type of eWOM (positive or negative). Therefore, a new brand equity variable is composed that includes the mean brand equity scores. Thereafter, a One-Way ANOVA is performed. Table 6 shows the outcomes of this test. Table 6: Effects of eWOM on brand equity
Positive eWOM (n=97)
Negative eWOM (n=139)
F P
Brand equity (mean) 5.02 4.89 .802 .371
24 4.3 Results on attitudinal brand loyalty
Although there cannot be detected an effect of eWOM on brand equity for the two smartphone brands Apple and Nokia together, there is a significant difference between an eWOM (positive or negative) of Apple and Nokia on brand equity (Table 7).
Table 7: The effects of eWOM and attitudinal brand loyalty on brand equity Positive eWOM Negative eWOM
Apple 5.72 5.63
Nokia 4.30 4.43
The main effect of type of eWOM (positive or negative) was not significant (F(1,236)=.018, n.s.). However, the main effect of Apple or Nokia (attitudinal brand loyalty) was very significant
(F(1,236)=111.485, p<.001). This means that brand equity strongly depends on the previous attitudinal brand loyalty with significant higher scores for the Apple brand.
The interaction between attitudinal brand loyalty and type of eWOM (positive or negative) was not significant (F(1.236)=.814, n.s.). This means that hypothesis 2, which states that the effect of eWOM on brand equity will be reduced by brand loyalty, is not supported.
Attitudinal brand loyalty did not appear to have a moderator effect between type of eWOM (positive or negative) and brand equity (Table 7). Therefore, we additionally investigated the moderator effect of brand loyalty on subjective price level, since subjective price level was used in the questionnaire as an additional measure of brand equity. We assessed the subjective price level by asking respondents to indicate their price perception of the smartphones on a scale between 1=too low and 7=too high. All respondents indicated the price level of the four smartphone brands separately.
Subsequently, a Two-Way ANOVA is performed for the smartphone brands Apple and Nokia separately to test the effects of eWOM and attitudinal brand loyalty on subjective price level. Results are shown in Table 8.
Table 8: The effects of eWOM and attitudinal brand loyalty on subjective price level Positive eWOM Negative eWOM F P
Apple 5.77 5.93 .503 .959
Nokia 4.11 4.12 .738 .790
Table 8 shows that for Apple, the subjective price level in the positive eWOM condition is not
25 In order to test whether attitudinal brand loyalty is indeed significantly higher for Apple than for Nokia, first the mean attitudinal brand loyalty scores are computed for Apple, Nokia, HTC and Samsung separately (Table 8). In this way, four new variables were created including the mean attitudinal brand loyalty scores for the four smartphone brands. Subsequently, a paired-samples t-test was performed to compute the differences in attitudinal brand loyalty between the different brands.
Table 9: Mean attitudinal brand loyalty scores for Apple, Nokia, HTC and Samsung Brand loyalty
Apple 5.40
Nokia 3.87
HTC 3.83
Samsung 4.71
The paired-samples t-test indicated that the mean attitudinal brand loyalty of Apple was significantly higher than the mean attitudinal brand loyalty of Nokia (t=12.75, df=235, p<.001). Therefore, this manipulation check indicates that Apple was correctly seen as the condition with the high attitudinal brand loyalty, while Nokia was correctly seen as the condition with the low attitudinal brand loyalty. From this we can conclude that attitudinal brand loyalty is indeed significantly higher for Apple than for Nokia.
Subsequent analysis indicated that the mean attitudinal brand loyalty of Apple was significantly higher than the mean attitudinal brand loyalty of HTC (t=13.79, df=235, p<.001). Moreover the mean attitudinal brand loyalty of Apple was significantly higher than the mean attitudinal brand loyalty of Samsung as well (t=5.71, df=235, p<001). The t-test further indicated that the mean attitudinal brand loyalty of Samsung is significantly higher than the mean attitudinal brand loyalty of Nokia (t=7.84, df=235, p<001). The assumption that the mean attitudinal brand loyalty of HTC is equal to the mean attitudinal brand loyalty of Nokia is not significant (t=.33, df=235, n.s.).
4.4 The effects of eWOM for Apple products
26 Table 10: Attitudinal brand loyalty of Apple as a moderator between eWOM and brand equity
Positive eWOM Apple (n=49)
Negative eWOM Apple (n=53)
Apple high loyalty 6.16 6.17 Apple low loyalty 5.30 5.10
The type of eWOM of Apple (positive or negative) did not influence the mean brand equity (F(1,102)=.397, n.s.).
The level of attitudinal brand loyalty with Apple did significantly influence mean brand equity
(F(1,102)=42,058,p<.001). The correlation between attitudinal brand loyalty of Apple and brand equity is .66 (p<.001).
The interaction between type of eWOM (positive or negative) and attitudinal brand loyalty with Apple was not significant (F(1,102)=.518, n.s.). This means that the effect of eWOM (positive or negative) does not depend on the level of attitudinal brand loyalty with Apple products. This contradicts hypothesis 2, which states that the effect of eWOM on brand equity will be reduced by attitudinal brand loyalty.
4.5 The effects of eWOM for Nokia products
Subsequently, we performed the same procedure for the Nokia brand. So, we split the attitudinal brand loyalty scores for Nokia into two groups based on its median (4.0). Scores with a maximum of 4.0 were split into ‘low loyalty towards Nokia’, and scores higher than 4.0 were split into ‘high loyalty towards Nokia’. Thereafter, we performed a Two-Way ANOVA for Nokia as well. The outcomes are shown in Table 11.
Table 11: Attitudinal brand loyalty of Nokia as a moderator between eWOM and brand equity Positive eWOM Nokia
(n=49)
Negative eWOM Nokia (n=53)
Nokia high loyalty 5.48 5.63 Nokia low loyalty 5.90 5.62
The type of eWOM of Nokia (positive or negative) did not influence the mean brand equity (F(1,102)=.100, n.s.)
Level of brand loyalty with Nokia did not significantly influence mean brand equity (F(1,102)=1.298, n.s.). This means that with the Nokia products attitudinal brand loyalty did not relate significantly to brand equity, in contrast to Apple.
27 4.6 Homophily as a moderator between eWOM and brand equity
In this study, homophily was measured according to the ‘occupation’ construct (student vs. working person). Homophily is assumed to be high when a student reads a review of a student or when a working person reads a review of a working person. In case of inequality in occupation between the reviewer and the respondent, homophily is assumed to be low. In this way, homophily is divided into two groups. In order to test hypothesis 3, a Two-Way ANOVA is performed. Table 12 shows the outcomes of this test.
Table 12: The effect of homophily as a moderator between eWOM and brand equity Positive eWOM (n=97) Negative eWOM (n=138) Low homophily 5.11 4.79 High homophily 4.94 4.97
The type of eWOM (positive or negative) did not influence brand equity (F(1,235)=.973, n.s.). Homophily did not significantly influence mean brand equity (F(1,235=.004, n.s). This means that homophily did not relate significantly to brand equity. The interaction between type of eWOM (positive or negative) and homophily was not significant (F(1,235)=1.396, n.s.). This means that the effect of eWOM does not depend on homophily with the reviewer. This contradicts with hypothesis 3, which states that the effect of eWOM on brand equity will be increased by homophily.
4.7 Results on purchase decision
Out of the 236 respondents who had to make a product decision for one of the four smartphone brands, 114 respondents choose for the Apple smartphone compared to only 15 respondents who preferred to buy the Nokia smartphone. Table 13 visualises how preferences for a smartphone (Apple or Nokia) are distributed after being exposed to an Apple review (positive or negative).
Table 13: Effects of type of eWOM on smartphone preference
Positive Apple review Negative Apple review Apple preference (n=114) 44.9% 46.2% Nokia preference (n=15) 71.4% 25.0%
Out of the respondents who were exposed to a positive Apple review, 44.9% preferred to buy the Apple smartphone subsequently. Even more respondents (46.2%) preferred to buy the Apple smartphone after being exposed to a negative Apple review.
28 Nokia preference is based on a very small sample. The choice for either Apple or Nokia appears to be random, so the type of eWOM seems not to affect smartphone preference.
4.8 Results on subjective price level
Respondents rated the price setting of Apple, Nokia, HTC and Samsung on a 7-point bipolar scale. In order to perform this test, price levels were recoded into three groups. Score 1 and 2= under priced; score 3, 4 and 5 = fairly priced and score 6 and 7= overpriced. Thereafter, the subjective price level per smartphone brand is computed in percentages. Results are shown in Table 14.
Table 14: Subjective price level per smartphone brand
Apple HTC Nokia Samsung
Under priced 5.1% 12.7% 17.4% 15.7%
Fairly priced 22.5% 72.4% 76.7% 70.8%
Overpriced 72.5% 14.9% 5.9% 13.6%
As can be seen in Table 13, Apple is seen as a smartphone brand with prices that are perceived to be too high by 72.5% of the respondents. In contrast, only 5.9% of the respondents claim that Nokia products are overpriced. Out of the four smartphone brands, Nokia is perceived most often (76.6%) as a brand with a correct price setting. In contrast, only 22.5% of the respondents claim that the Apple
29
5. CONCLUSION AND RECOMMENDATIONS
After interpreting the results in chapter 4, in this chapter we will elaborate on the findings. In addition, the limitations of this study will be identified and future research avenues will be suggested.
5.1 Discussion of the findings Hypothesis 1
The main subject of study was to investigate whether brand equity is higher for customers receiving positive eWOM than for customers receiving negative eWOM. After testing this assumption in hypothesis 1, this statement cannot be confirmed. This implicates that the type of eWOM (positive or negative) does not influences the level of brand equity.
Hypothesis 2
In hypothesis 2 is assumed that the effect of eWOM on brand equity will be reduced by attitudinal brand loyalty. We checked for this hypothesis in several ways.
First, we analysed this assumption by testing the interaction between attitudinal brand loyalty (of Apple and Nokia) and type of eWOM (positive or negative). This analysis showed no significant outcomes and therefore, we cannot support that the effect of eWOM on brand equity will be reduced by brand loyalty. However outside the initial study design, this analysis did find a significant direct effect of attitudinal brand loyalty (of Apple and Nokia) on brand equity. This means that customers that score higher on attitudinal brand loyalty score higher on brand equity as well. This indicates that brand loyalty is a strong driver of brand equity (direct effect), with significant higher scores for Apple than for Nokia.
Subsequently, we analysed whether the effect of eWOM (positive or negative) on brand equity of Apple will be reduced by attitudinal brand loyalty towards Apple by splitting the Apple loyalty construct based on its median into a high loyal group and a low loyal group. It is found that the effect of eWOM (positive or negative) does not depend on the level of attitudinal brand loyalty with Apple products. This
contradicts hypothesis 2, which states that the effect of eWOM on brand equity will be reduced by attitudinal brand loyalty.
However it was not tested in one of the hypotheses, in this analysis a significant direct effect of attitudinal brand loyalty towards Apple products on brand equity of Apple is found. This means that brand loyalty towards Apple is a strong driver of brand equity of Apple (direct effect).
Thereafter, the same median split analysis for Nokia was performed. The effect of eWOM (positive or negative) does not depend on the level of attitudinal brand loyalty with Nokia products. This contradicts hypothesis 2, which states that the effect of eWOM on brand equity will be reduced by attitudinal brand loyalty.
30 From these analyses, we can conclude that attitudinal brand loyalty can be seen as a more important variable that influences the level of brand equity than type of eWOM (positive or negative), as originally assumed in our hypotheses. Argumentations that support that type of eWOM has no effect are that customers are not influenced by a positive eWOM simply because they are already loyal. This confirms our theory about the ceiling-effect, as proposed in paragraph 2.5. Highly loyal customers are not influenced by a negative eWOM either, because highly loyal customers will not adopt this negative content. This finding supports our defensive-theory, as proposed in paragraph 2.5.
In sum, we can conclude from these analyses that instead of the moderating role of attitudinal brand loyalty, there is a direct link between attitudinal brand loyalty and brand equity. This finding was outside the scope of this study and was therefore unexpected. This finding seems especially to be true for the Apple brand. For Apple, attitudinal brand loyalty appears to have such a strong effect that after being exposed to a negative Apple eWOM even more respondents preferred the Apple brand than after being exposed to a positive eWOM of Apple.
Therefore, we can conclude that the assumption that it is easy to destroy the equity of a brand by means of a negative eWOM is true for brands that possess weak brand loyalty, since a direct and strong relationship is detected between attitudinal brand loyalty en brand equity. This implicates that brands with low attitudinal brand loyal customers are more vulnerable to negative eWOM, in that sense that its brand equity will easily be impaired.
To distinct high loyalty from low loyalty, we tested the loyalty construct for Apple and Nokia (and HTC and Samsung to distract the respondent from the goal of this study). Analysis shows that Apple is correctly associated as a high loyalty brand and Nokia is correctly associated as a low loyalty brand. Hypothesis 3
This study further assessed the moderating role of homophily. Hypothesis 3, which states that the effect of eWOM on brand equity will be increased by homophily, could not be confirmed. This implicates that equality in occupation (student or working) between the reviewer and reader of the eWOM, does not influence the level of brand equity. So the credibility of the eWOM is not dependent on similarities in occupation.
5.2 Limitations and directions for further research
In this paragraph we will discuss the limitations of this study. In addition, avenues for further research will be discussed.
Generalization limitation
31 Limitation of Apple and Nokia as chosen brands for the manipulation
The fact that the hypotheses show no significant outcomes might be caused by the choice of the brands for the manipulation. Apple and Nokia are chosen as brands to make sure every respondent knows the brand and therefore can form an opinion towards the brand. However, these brands probably are too strong. Since respondents have a clear en deeply held opinion towards Apple and Nokia, these brand attitudes are not easy to change by means of just one eWOM review with positive or negative valence. For the manipulation to be significant, it might have been better to choose for fictive, no existing brands, since respondents have to form an opinion or attitude about that brand while reading the eWOM review. Although using fictive brands works well in some cases, it would not work in this study. The reason here fore is we cannot assess brand loyalty by letting respondents answer loyalty items about brands they have never heard before. Another option that might have led to better outcomes of the hypotheses is to choose for newer or less well-known brands, to make sure respondents do not have formed strong and unchangeable believes and attitudes in mind. In that case, it is more likely that respondents change their opinion or attitude about that brand while reading the eWOM review. Low volume of eWOM results in limited quality of eWOM manipulation
Third limitation is that for the eWOM condition, a short review is simulated. Based on the valence of this little review, the level of brand equity was supposed to be more positive or more negative than the loyalty scores reported for that brand. The effect of such a short review could be questioned.
In literature can be found that customers build higher credibility on products with larger WOM volume. This is because customers, trough knowledge of other customer’s experiences concerning a particular product, can reduce uncertainty associated with the product (Yang, Kim, Amblee and Jeong, 2012). Therefore, the likelihood of differences in brand equity would probably have been higher when respondents were exposed to several forms of positive (or negative) information during a fifteen minutes presentation held by smartphone experts, or when respondents were exposed to a database with multiple reviews that all confirmed the positive (or negative) valence.
No manipulation was included that checked whether the eWOM was read by the respondent Fourth limitation is that in the questionnaire, no question was included that controlled for the fact whether the respondent indeed read the content of the review carefully. Therefore, we can only assume the respondent did read the review, but we cannot prove whether this is the case.
Limited quality of the homophily manipulation
32 The threat of hypothesis guessing
To cover hypothesis guessing, the brand loyalty construct was not only assessed for Apple and Nokia, but also for HTC and Samsung to distract respondents from the goal of the questionnaire. The brand equity construct did not include items to distract the respondent. However, 65.3% of the respondents in this study consist out of students. Students are exposed to many questionnaires and research studies and therefore, it might be the case that these respondents are guessing what the real purpose of the study is. When these respondents subsequently base their behavior on what they guess instead of the manipulation, this is a validity threat. A limitation of this study is that the questionnaire did not check for hypothesis guessing.
Future research should focus on the direct link between attitudinal brand loyalty and brand equity. Especially brands that customers have low attitudinal brand loyal feelings towards should concern about the impact of negative eWOM, since these brands do not benefit from any protection resulting from the strong influence of attitudinal brand loyalty on brand equity.
33
REFERENC ES
Aaker, D. A. (1996). Measuring brand equity across products and markets. California Management Review, 38, 102 – 120.
Aaker, D. A. and Jacobson, R. (1994). The financial information content of perceived quality. Journal of Marketing Research, 31, 191 –201.
Ahluwalia, R., Burnkrant, R. E. and Unnava, H. R. (2000). Consumer response to negative publicity: The moderating role of commitment, Journal of Marketing Research, 37, 203 – 214.
Bearden, W.O., and Netemeyer, R. (2011). Handbook of Marketing Scales (3nd ed.). Thousand Oaks, Sage Publications.
Bickart, B. and Schindler R. M. (2001). Internet forums as influential sources of consumer information. Journal of Interactive Marketing, 15, 31 – 40.
Braun, M. and Bonfrer, A. (2011). Scalable inference of customer similarities from interactions data using Dirichlet processes. Marketing Science, 30, 513 – 531.
Brown, J. J. and Reingen, P. H. (1987). Social ties and word-of- mouth referral behavior. Journal of Consumer Research, 14, 350 – 62.
Brodie, R. J., Ilic, A., Biljana, J. and Hollebeek, L. (2011b), Consumer engagement in a virtual brand community: an exploratory analysis. Journal of Business Research, 66, 105 – 114.
Brown, J., Broderick, A. J., and Lee, N. (2007). Word of mouth communication within online communities: Conceptualizing the online social network. Journal of Interactive Marketing,
21, 2 – 20.
Calder, B. J. and Malthouse, E. C. (2008). Media engagement and advertising effectiveness, in Kellogg on advertising and media, Bobby J. Calder, ed. Hoboken, NJ: Wiley, 1-36.
Chaudhuri, A. and Holbrook, M. B. (2001). The chain of effects from brand trust and brand affect to brand performance: The role of brand loyalty. Journal of Marketing, 65, 81 – 93.
Chen, Y., Wang, Q. and Xie, J. (2011). Online social interactions: A natural experiment on word-of-mouth versus observational learning. Journal of Marketing Research, 48, 238 – 254.
34 Cobb-Walgren, C. J., Beal, C. and Donthu, N. (1995). Brand equity, brand preferences, and purchase intent. Journal of Advertising, 24, 25 - 40.
Dawar, N. and Pillutla, M. M. (2000). Impact of product-harm crises on brand equity: The moderating role of consumer expectations. Journal of Marketing Research, 37, 215 – 226.
Day, G. S. (1969). A two-dimensional concept to brand loyalty. Journal of Advertising Research, 9, 29 – 35.
Dick, A. S. and Basu, K. (1994). Customer loyalty: Toward an integrated conceptual framework. Journal of the Academy of Marketing Science, 22, 99 – 113.
Doorn, van, J., Lemon, K. N., Mittal, V., Nass, S., Pick, D. and Verhoef, P. C. (2010). Customer Engagement Behavior: Theoretical foundations and research directions. Journal of Service Research, 13, 253 – 266. Engel, J. E., Blackwell, R. D., Kegerreis, R. J. (1969). How information is used to adopt an innovation. Journal of Advertising Research, 9, 3 – 8.
Feick, L. and Higie, R. A. (1992). The effects of preference heterogeneity and source characteristics on ad processing and judgments about endorsers. Journal of Advertising, 21, 9 – 23.
Galbreth, M. R., Ghosh, B. and Shor, M. (2012). Social sharing of information goods: Implications for pricing and profits. Marketing Science, 31, 603 – 620.
Godes, D. and Mayzlin, D. (2004), Using online conversations to study word-of-mouth communication. Marketing Science, 23, 545 - 560.
Godes, D., Mayzlin, D., Chen, Y., Das, S., Dellarocas, C., Pfeiffer, B. et al. (2005). The firm’s management of social interactions. Marketing Letters, 16, 415 – 428.
Gruen, T. W., Osmonbekov, T. and Czaplewski, A. J. (2006). eWOM: The impact of customer-to customer online know-how exchange on customer value and loyalty. Journal of Business Research, 59, 449 – 456. Hennig-Thurau, T., Gwinner, K. P., Walsh, G., and 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, 38 – 52.
Higgins, E. T. (2006). Value from hedonic experience and engagement. Psychological Review, 113, 439 - 460.
35 Iglesias, O., Singh, J. J. and Batista-Foguet (2011). The role of brand experience and affective commitment in determining brand loyalty. Brand Management, 18, 570 – 582.
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57, 1 – 22.
Keller, K. L. (2008). Strategic Brand Management (3rd edn). Upper Saddle River, NJ: Prentice-Hall. Kumar, V. and Reinartz, W. J. (2006). Customer Relationship Management: A Databased Approach. John Wiley and Sons Ltd.
Leeflang, P. (2011). Paving the way for “distinguished marketing”. International Journal of Research in Marketing, 28, 76 – 88.
Libai, B., Bolton, R., Bügel, M. S., de Ruyter, K., Götz, O., Risselada, H. and Stephan, A. T. (2010). Customer-to-customer interactions: Broadening the scope of word of mouth research. Journal of Service Research, 13, 267 - 282.
McPherson, M. and Smith-Lovin, L. (1987). Homophily in voluntary organizations: Status distance and the composition of face-to-face groups. American Sociological Review, 53, 370 – 379.
McPherson, M., Smith-Lovin, L. and Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415 – 444.
Mollen, A. and Wilson, H. (2010). Engagement, telepresence and interactivity in online consumer experience: Reconciling scholastic and managerial perspectives. Journal of Business Research, 63, 919 - 25.
Narayanan, S. and Nair, H. S. (2013). Estimating causal installed-base effects: A bias-correction approach. Journal of Marketing Research, 50, 70 – 94.
Nekmat, E. (2012). Message expression effects in online social communication. Journal of Broadcasting & Electronic Media, 56, 203 – 224.
Nitzan, I., and Libai, B. (2011). Social effects on customer retention. Journal of Marketing, 75, 24 – 38. Onishi, H., and Manchanda, P. (2012). Marketing activity, blogging and sales. International Journal of Research in Marketing, 29, 221 – 234.
36 Park, C. and Lee, T. M. (2009). Information direction, website reputation and eWOM effect: A moderating role of product type. Journal of Business Research, 62, 61 – 67.
Rego, L. L., Billett, M. T. and Morgan, N. A. (2009). Consumer-based brand equity and firm risk. Journal of Marketing, 73, 47 - 60.
Richins M. L. (1983). Negative word-of-mouth by dissatisfied consumers: a pilot study. Journal of Marketing, 47, 68 – 78.
Roehm, M. L. and Brady, M. K. (2007). Consumer responses to performance failures by high-equity brands. Journal of Consumer Research, 34, 537 - 545.
Rogers E. M. (1983). Diffusion of innovations. New York, NY: Free Press.
Rust, R. T., Zeithaml, V. A. and Lemon, K. (2004). Customer-centered brand management. Harvard Business Review, 82, 110 – 118.
Sloot, L. M., Verhoef, P. C. and Franses, P. (2005). The impact of brand equity and the hedonic level of products on consumer stock-out reactions. Journal of Retailing, 81, 15 – 34.
Stahl, F., Heitmann, M., Lehmann D. R. and Neslin, S. A. (2012). The impact of brand equity on customer acquisition, retention, and profit margin. Journal of Marketing, 76, 44 – 63.
Thomson, M., MacInnis, D. J. and Park, C. W. (2005). The ties that bind: Measuring the strength of consumers’ emotional attachments to brands. Journal of Consumer Psychology, 15, 77 – 91.
Yang, S. and Allenby, G. M. (2003). Modeling interdependent consumer preferences. Journal of Marketing Research, 40, 282 – 294.
Yang, Y., Kim, W., Amblee, N. and Jeong, J. (2012). The heterogeneous effect of WOM on product sales: Why the effect of WOM valence is mixed. European Journal of Marketing, 46, 1523 – 1538.
Verhoef, P. C., Langerak, F. and Donkers, B. (2007). Understanding brand and dealer retention in the new car market: The moderating role of brand tier. Journal of Retailing, 83, 97 – 113.
37
APP ENDIX
Questionnaire
Mijn naam is Annabel Annen, master student Marketing aan de Rijksuniversiteit Groningen. Voor mijn master thesis zou ik u willen vragen deze enquête in te vullen. De enquête gaat over online
consumentengedrag in de telecomindustrie en zal ongeveer 5 minuten van uw tijd in beslag nemen. U zou mij ontzettend helpen als u deze enquête invult! Uw antwoorden zijn anoniem en zullen
vertrouwelijk worden behandeld.
Bij voorbaat heel erg bedankt voor uw deelname!
Met vriendelijke groet, Annabel Annen
Denk voor ongeveer 10 seconden aan het merk Nokia. Geef vervolgens aan in welke mate de volgende woorden uw gevoelens beschrijven ten aanzien van Nokia.
1. Geliefd
omschrijft slecht 1 –2 – 3 - 4 – 5 – 6 - 7 omschrijft goed 2. Opgetogen
omschrijft slecht 1 –2 – 3 - 4 – 5 – 6 - 7 omschrijft goed 3. Verbonden
omschrijft slecht 1 –2 – 3 - 4 – 5 – 6 - 7 omschrijft goed 4. Gehecht
omschrijft slecht 1 –2 – 3 - 4 – 5 – 6 - 7 omschrijft goed 5. De prijsstelling van Nokia vind ik
te laag 1 – 2 – 3 – 4 – 5 – 6- 7 te hoog
Denk voor ongeveer 10 seconden aan het merk HTC. Geef vervolgens aan in welke mate de volgende woorden uw gevoelens beschrijven ten aanzien van HTC.
6. Geliefd
omschrijft slecht 1 –2 – 3 - 4 – 5 – 6 - 7 omschrijft goed 7. Opgetogen
