Brand authenticity in the age of globalization:
The effects of authentic cues on
emotional brand attachment and word of mouth
Master thesisMSc in Business Administration - Marketing
Name: Suzanne Bos Student nr: 11122749
Date of submission: 27-01-2017
MSc. in Business Administration – Marketing Track UvA
2 Statement of originality
This document is written by Suzanne Bos who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating them. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Abstract
Although most marketing related studies nowadays focus on outside-in customer orientation, another stream proposes that a different approach may help brands as well: an inside-out product orientation. This inside-out product orientation in the form of brand authenticity may help customers to find back their identity in this era of globalization. Brands can influence the degree to which customers perceive a brand as authentic via authentic cues. Some of these cues are objective (such as brand scandals surrounding the brand) and some are subjective (based on the marketing communications of the brand).
This paper considers the effects of authentic cues on perceived brand authenticity and the effects of perceived brand authenticity on emotional brand attachment and word of mouth. An extensive analysis showed that the results in most of the cases led to (partially) significant relationships.
Key words: authentic cues, perceived brand authenticity, emotional brand attachment, marketing
4 Index 1. Introduction……… 6 2. Literature review………... 9 2.1Authenticity……… 9 2.1.1 Authenticity in marketing ………. 9 2.1.2 Authentic cues ………... 9
2.1.3 Perceived brand authenticity……….. 12
2.2 Marketing skepticism ………... 15
2.3 Consequences of PBA ………... 16
2.3.1 Emotional brand attachment……….. 17
2.3.2 Word of mouth ……….. 20 2.4 Conceptual model ………. 22 3. Method ………... 23 3.1 Procedure ……….. 23 3.2 Measures……… 23 3.3 Sample ……….. 25 4. Results ……… 26 4.1 Respondents ……….. 26 4.2 Data preparation ……… 26 4.3 Hypotheses testing ……… 29 4.3.1 Hypotheses 1a+b ………... 29 4.3.2 Hypotheses 2a+b+c ………... 29 4.3.3 Hypothesis 3 ……….. 30 4.3.4 Hypothesis 4 ……….. 30 4.3.5 Hypotheses 5a+b ……… 30 4.3.6 Hypotheses 6 a+b……… 30 4.3.7 Hypotheses 7a+b………. 31 5. Discussion ……….. 35 5.1 General discussion ………... 35 5.2 Theoretical implications ………... 37 5.2 Managerial implications ………... 37
5.3 Limitations and future research ……… 39
References ………. 41
Appendix ………... 47
Appendix 1: Demographics ………... 47
Appendix 2: Independent sample t-test ………. 47
5
Appendix 4: Model fit summary ……… 48 Appendix 5: Hypotheses ……… 49 Appendix 6: Survey ………... 52
6 1. Introduction
In this era of globalization it is a challenge for countries and individuals ‘to find a healthy balance between preserving a sense of identity, home and community and doing what it takes to survive within the globalization system’ (Friedman, 2012, p. 35). Although a considerable part of the literature nowadays focuses on customer orientation, Moulard, Raggio & Folse (2016) suggest that this may not be the only appropriate strategic decision. Their study demonstrates the positive impact of a product orientation: an orientation reflected by the brand’s interests and expertise that dictates marketing decisions. Moulard et al. (2016) suggest that production orientation is a synonym for brand authenticity, because this means that customers perceive a brand to be directed by the interest in their own products and not driven by the customers’ wants and needs. Thus, in an attempt to cope with this search for identity by customers, companies try to position themselves as authentic: ‘emphasizing the timeless value desired by consumers while downplaying commercial motives’ (Beverland, 2005, p.460). This is in line with the article of Napoli, Dickinson-Delaporte & Beverland (2016) which states that the benefits of a perceived authentic brand are ‘a brand’s ability to fulfil an individual’s self-identity project and the emotional connection established between a brand and its consumers’ (p. 4). Beverland & Farrelly (2010) state that customers have certain personal goals in mind when consuming a perceived authentic product, because these products may have different identity benefits, such as control over their life or mastering certain skills.
According to the Authentic Brand Index (http://www.authenticbrandindex.com/), customers nowadays are drawn toward brands with original stories, engaging identities and sincere commitment to deliver what they promise. Based on several studies of the Authentic Brand Index, customers are 60% more willing to advocate for a brand they perceive as highly authentic than for brands they perceive as inauthentic and the share of high value customers is 20% higher for highly authentic brands than for inauthentic brands.
Companies can position themselves as authentic by making use of authentic cues in their marketing communications. Considerable research has been done to the effects of authentic cues in marketing communications for hedonic products like wine, beer or MG cars (Beverland & Luxton, 2005; Beverland, 2006; Beverland, Lindgreen & Vink 2008; Leigh, Peters & Shelton, 2006). To date, only two studies have researched the effects of the use of authentic cues in marketing communications for mass market brands (Morhart, Malär, Guevremont, Girardin & Grohmann, 2015; Bruhn, Schoenmüller, Schäfer & Heinrich, 2012). However, the emphasis in these studies lies on the ‘heritage’ aspect of brand authenticity.
7
This aspect of brand authenticity is important for brands with a rich history and an important product heritage, such as wine brands or car brands like MG. Yet, mass market brands with a lesser rich history or lesser important product heritage can potentially also be perceived as authentic or product oriented, by positioning themselves as ‘real’ and ‘true towards their customers’. As there are already several brands (e.g. Rabobank, Pickwick, Albert Heijn) that approach to position themselves in such a way via their marketing activities, it is important to understand the effects of it. The emphasis in the current study is therefore on the effect of brand authenticity for mass market brands that position themselves as ‘real’ and ‘true towards their customers’. In this way, I will extend the model of Morhart et al. (2015) by adding an additional authentic cue.
An example of a brand that tries to position itself as real or true towards their customers is the Dutch supermarket leader Albert Heijn. Albert Heijn has just launched a new marketing campaign in which they show their true selves: ‘Everybody Appie’. In contrast with the fictional ‘mister Van Dalen’ - the supermarket manager whose stories have been told for ten years in a row - Albert Heijn now wants to tell its customers ‘real stories from real people’. In the new series of TV commercials, employees, customers and vendors of Albert Heijn tell their stories in relation to Albert Heijn, that is: why they are ‘Appie’. For example, an older lady, Magda, explains that in the beginning online shopping at Albert Heijn was quite a hassle, but now she has learned how to do it on her ‘senior’ computer, it is actually very easy. Besides that, she has a little crush on the delivery boy, who makes her even more enthusiastic about the service.
Morhart et al. (2015) state that perceived brand authenticity (hereafter: PBA) increases emotional brand attachment and word of mouth (hereafter: WOM). When customers feel emotionally attached to a brand, they are more willing to forsake personal resources - such as investments in time, money and energy- for this brand (Park, MacInnis, Priester, Eisingerich & Iacobucci, 2010). Therefore emotional brand attachment is seen as the ultimate goal for brands in establishing a commercial relationship with customers (Park et al., 2010). WOM has been proven to be twenty times as effective as traditional marketing (Trusov, Bucklin & Pauwels, 2009). It is therefore important that brands understand how they can make customers feel attached to the brand and how a brand can increase the WOM with regard to the brand.
8
For this reason, the following research question is formulated:
What is the effect of authentic cues in marketing communications on emotional brand attachment and word of mouth?
This study has several theoretical and managerial contributions. First, it sheds light on the effects of marketing communications for mass market brands that focus on being perceived as real or true towards their customers. In addition, this study shows the effects of marketing skepticism on brands that try to position themselves as real or true towards their customers, because one can ask themselves if people really believe the attempts of brands of trying to be perceived as authentic, or only see the commercial motives of the brand accompanied by the marketing communications. It is therefore expected, that the relationship between authentic cues and the way people perceive this as authentic will be moderated by marketing skepticism. Another contribution of this study is the insight in the effects of emotional brand attachment on WOM. In the article of Morhart et al. (2015) emotional brand attachment and WOM are approached as being two distinct constructs. However, according to several studies (Westbrook, 1987; Harrison-Walker, 2001; Fullerton, 2003; and Brown, Barry, Dacin & Gunst, 2005) WOM is a direct consequence of emotional brand attachment. It is therefore expected that the relationship between PBA and WOM is mediated by emotional brand attachment. Lastly, this study contributes as it shows the mediating effects of PBA on the relationship between the authentic cues and emotional brand attachment and WOM. According to the study of Morhart et al. (2015), brand attachment and WOM are direct effects of PBA. However, in this study I put forward the notion that PBA also functions as a mediator between authentic cues and emotional brand attachment and WOM. To test the effects of authentic cues in marketing communications for mass market brands on emotional brand attachment and WOM, a quantitative study in the form of a survey will be conducted for this study.
The remainder of this research is structured as follows: First, the relevant literature relating to authentic cues and their effects on emotional brand attachment and WOM will be reviewed. Second, the research design and method will be discussed, followed by the analysis of the results. Lastly, the results will be discussed and the implications and future research
suggestions will be given.
9 2. Literature review
To shed light on the empirical work that has addressed the relationship between authentic cues, emotional brand attachment and WOM, this chapter provides an overview of the relevant literature on these subjects. First, different definitions of authenticity will be compared, followed by the description of the different cues that can be used in marketing communications. Subsequently, the influence of marketing skepticism will be reviewed and we will take a look at the dimensions that influence if people perceive a brand as authentic. Finally, brand attachment and WOM will be explained.
2.1. Authenticity
2.1.1 Authenticity in marketing
The word authenticity finds his grounds in the Greek (‘authentikos) and Latin (‘authenticus’) language and means ‘original, genuine, authoritative; one who does something
himself or herself’ (http://wordinfo.info/unit/3771/ip:16) or ‘worthy of acceptance, not
imaginary, false or imitation, conforming to an original’ (Cappannelli & Cappannelli, 2004). According to Merriam-Webster’s Collegiate Dictionary (2016), something is authentic when it is ‘real or genuine, true and accurate and made to be or look just like an original’. However, different views exist on what authenticity really means for brands: ‘a subjective evaluation of genuineness ascribed to a brand by customers’ (Napoli, Dickinson, Beverland & Farrelly, 2014, p. 1091); ‘a rationally created characteristic informing an individual’s subjective perceptions’ of a brand rather than a characteristic immanent to an objective reality (Bruhn et al., 2012, p. 569) and; customers can find authenticity in different events and products, including functional and universal products, mainstream events and even in brands (Beverland & Farrely, 2010). The emphasis in these definitions lies on the subjective perspective of perceiving something as authentic. However, there exist three different perspectives on how customers perceive authenticity, namely: an objectivist perspective, a subjective or constructivist perspective and an existentialist perspective (Leigh et al., 2006). These will be further specified under authentic cues.
2.1.2 Authentic cues
Trying to position themselves as authentic via mass marketing communication may be difficult for brands as these mass marketing communications are often perceived as inauthentic. Beverland & Luxton (2005) state that authentic mass marketing communication needs to be tightly integrated with the internal operations of the company in order to be
10
perceived as authentic. Therefore, it is important to understand the different ways customers may perceive something as authentic. According to the article of Leigh et al. (2006), authenticity for brands is a threefold concept, based on different perspectives. These three perspectives are comparable with the three perspectives on authenticity meant in the article of Beverland et al. (2008). By using a different kind of authentic cues dependent on the perspective of authenticity, brands can influence the way customers perceive their brand as authentic. The three different perspectives and accompanying cues are as follows:
1. Objective authenticity (Leigh et al., 2006) or pure (literal) authenticity (Beverland et al., 2008): authenticity may be defined by judging objects or experiences according to objective criteria (Leigh et al., 2006). According to this perspective, authenticity can be seen as a ‘quality inherent in an object and evaluated by experts’ (Morhart et al., 2015, p. 201) or an artistic aura of originality (Leigh et al., 2006). The purpose of cues is to give evidence for what a brand claims to be (Morhart et al. 2015). Indexical cues with verifiable information about the brand can stimulate the way customers perceive a brand as (objective or (pure) literal) authentic (Grayson & Martinec, 2004; Beverland et al., 2008; Morhart et al., 2015). In the study of Morhart et al. (2015), two different forms of iconic cues are used. Actual behavior of brands in the form of brand scandals may function as an indexical cue because this is not in line with the moral principles of the brand and this places self-interest before other stakeholder’s interests (Morhart et al., 2015). This will negatively influence the way customers perceive a brand as authentic. Brand-congruent employee behavior may influence the PBA in a positive way when the employee behavior is in line with the promises made in a brand’s marketing communicating activities (Morhart et al., 2015).
2. Constructive authenticity (Leigh et al., 2006) or approximate authenticity (Beverland et al., 2008): ‘allows for different interpretations of reality on the basis of customers’ projections onto objects and is essentially symbolic’ (Leigh et al., 2006, p 483). According to this perspective, customers form their own real world based on different interpretations (Morhart et al., 2015) and their own personal experiences (Leigh et al., 2006). The purpose of cues for this perspective on authenticity is to provide abstract impressions about the essence of the brand (Morhart et al., 2015). Iconic cues with abstract impressions of the brand’s essence may stimulate the way customers perceive a brand as (constructive or approximate) authentic (Grayson & Martinec, 2004; Beverland et al., 2008; Morhart et al.,
11
2015). Morhart et al. (2015) use two different forms of iconic cues: a brand’s communication style emphasizing its roots and a brand’s communication style emphasizing a brand’s virtue.
3. Existential authenticity (Leigh et al., 2006) or moral authenticity (Beverland et al., 2008): a product of postmodern customers’ chase for experiences and amusement. This perspective on authenticity involves ‘personal or subjective feelings activated by the liminal process of activities’ (Leigh et al., 2006, p. 483). These perspectives on authenticity are related to one’s identity and it sees authenticity as a way of being true to one’s self (Morhart et al., 2015). The purpose of cues for this perspective on authenticity is to provide the customer ‘with a feeling that this brand will help achieve self-authentication through connecting with personal moral values’ (Beverland et al., 2008, p.8) or constructing their own identity (Morhart et al., 2015). Existential cues that can serve as an identity-related source may stimulate the way customers perceive a brand as (existentially or morally) authentic (Beverland et al., 2008; Morhart et al., 2015). Morhart et al. (2015) use ‘brand anthropomorphism’ as an existential cue. Brand anthropomorphism can be seen as the assignment of human characteristics to brands (Aaker & Fournier, 1995) which makes it easier to recognize the values of a brand (Morhart et al., 2015). When the characteristics of a brand are clear and support the development of the self-identity of customers, this may increase the PBA (Morhart et al., 2015).
Although these perspectives seem very different from each other, the article of Leigh et al. (2006) suggests that these three perspectives are interweaved with each other and that they all influence the way customers perceive a brand as authentic (or not). Morhart et al. (2015) state that people develop PBA out of a combination of objective facts (objective or (pure) literal authenticity), subjective mental associations (constructive or approximate authenticity) and existential motives connected to brands (existential authenticity or moral authenticity). This is in line with the conclusion of Napoli et al. (2014) and Grayson & Martinec (2004) that customers use a combination of cues in evaluating about the authenticity of brands. Therefore, this study will focus on both indexical, iconic and existential cues since people make use of a combination of cues to decide if they perceive something as authentic. For this we will use the authentic cues of Morhart et al. (2015), which are mentioned above. In addition, a new authentic cue is added to this model: ‘a real story communication style’. As mentioned before, authenticity can be found in the genuineness of products (Napoli et al., 2014). However, there is also an aspect in authenticity that indicates that something is real or
12
original and that someone does something himself or herself
(http://wordinfo.info/unit/3771/ip:16; Cappannelli & Cappannelli, 2004; Merriam-Webster’s Collegiate Dictionary, 2016). As customers use a combination of cues to decide if something is authentic, we can conclude that the extent to which customers believe a brand is real and true towards its customers may influence customers’ PBA (Morhart et al., 2015). This is in contrast with the conceptualization of Grayson & Martinec (2004) which says that authentic is not per definition a synonym for ‘true’ or ‘real’, because just being truthful does not mean that customers perceive a brand as authentic. As there are several brands that use real stories from real people in their marketing communications, it is interesting to examine the effects of this. By doing this, it is tested whether people really believe that the people in the marketing communications are real, and if they tell their own real story. Before the accompanying hypotheses will be stated, the dimensions that influence how people perceive a brand as authentic (PBA) will be reviewed. These dimensions will probably be influenced by aforementioned authentic cues.
2.1.3 Perceived brand authenticity
Different dimensions have been used to measure the PBA of customers. The article of Bruhn et al. (2012) identifies four dimensions: continuity, originality, reliability and naturalness; Napoli et al. (2014) identify three dimensions: quality commitment, heritage and sincerity; and Morhart et al. (2015) measure the PBA on the basis of these four dimensions: continuity, credibility, integrity and symbolism. Although these dimensions share some commonalities, they are each conceptualized in a different cultural context. Bruhn et al. (2012) state that one should be cautious with using these dimensions in a context different from the context in which they are developed. For this study - which will be conducted in the Netherlands - the dimensions of Bruhn et al. (2012) and Morhart et al. (2015) are both applicable. As the authentic cues of Morhart et al. (2015) are used in this study, their PBA dimensions will be used as well. In this way we will overcome the potential issues that are associated with cross-cultural validity (Napoli et al., 2016). The PBA dimensions of Morhart et al. (2015) are described as follows: 1.) Continuity refers to the brand’s history and stability over time, and the likelihood that it will persevere in the future, 2.) Credibility indicates the brand’s willingness and capability to live up to their promises, 3.) Integrity points out the virtue reflected in the brand’s intentions and in the value it conveys and 4.) Symbolism refers to the reflection of the brand’s values that help customers construct who they are. Based on these dimensions, Morhart et al. (2015) define PBA as ‘the extent to which customers
13
perceive a brand to be faithful toward itself (continuity), true to its customers (credibility), motivated by caring and responsibility (integrity) and able to support customers in being true to themselves (symbolism) ‘ (p.203). This definition of PBA will be used in this study as well.
The first indexical objective cue is customer’s knowledge of potential brand scandals. As long as customers have no other objective knowledge of a brand, they may rely on the actual behavior of brands. According to the study of Morhart et al. (2015), brand scandals may give an indication that the actions of a brand are not in line with the moral principles of the brand and that the brand places self-interest before other stakeholders interests. It is therefore expected that the more customers are aware of brand scandals, the lesser they perceive a brand as integer. Therefore, the following hypothesis will be tested:
H1a. Brand scandals decrease perceived integrity.
According to the article of Morhart et al. (2015) when the behavior of an employee is congruent with the promises made in a brand’s marketing communication activities, this may indicate that the brand is willing and capable to live up to its promises. Therefore, it is expected that the more customers perceive that the behavior is in line with the promises of the brand, the more customers perceive a brand as credible and thus able to live up to their promises (Burmann & Zeplin, 2005). However, as some brands have no employees in direct contact with their customers, ‘boundary spanning agent’ is used for this hypothesis instead of ‘employee’. The people in the marketing communications of the brand are thus seen as boundary spanning agents, as they are the face of the brands. The accompanying hypothesis is in line with the article of Morhart et al. (2015):
H1b. Brand congruent boundary spanning agent’s behavior increases perceived credibility.
A communication style that communicates the brand’s moral values, dedication in execution and the human factor may claim a sense of moral authenticity in the mind of the customers (Morhart et al., 2015) and it shows the brand’s commitment to strive for more than profitability and economic interest only (Beverland et al., 2008). This may help customers to see the virtue reflected in the brand’s intentions and in the value the brand conveys. As this is
14
an iconic (subjective) authentic cue, it differs per customer how this kind of communication is evaluated. It is expected that the more customers experience the virtue communicated by a brand, the more they may perceive a brand as integer (Morhart et al., 2015). For this reason, the following hypothesis is proposed:
H2a. A brand's communication style emphasizing its virtue increases perceived integrity.
Communicating with real people and their own real stories in relation to the brand can make customers believe that a brand is true towards its customers. In this way, the brand acts in line with the original meaning of ‘authentic’ as mentioned above: ‘one who does something himself or herself’ (http://wordinfo.info/unit/3771/ip:16). Claims that are made or stories that are told may be perceived as being more credible, as these stories are real stories and not fictional stories set up for marketing communication. This makes it more plausible that a brand is able to live up to its promises. Since this is an iconic (subjective) cue, it may differ per customer if they perceive a brand as credible. Therefore, the following hypothesis is stated:
H2b. A brand’s real story communication style increases perceived credibility.
A communication style that focuses on a brand’s roots may denote that customers perceive a brand as a stable brand with a history, and customers may trust the likelihood that the brand will persevere in the future (Morhart et al., 2015). It is therefore expected that a communication style that focuses on a brand’s roots may increase the continuity: the brand’s history and stability over time, and the likelihood that it will persevere in the future. This is in accordance with the article of Morhart et al. (2015). As this is an iconic (subjective) cue, customers may evaluate marketing communications focusing on the roots of a brand differently. The accompanying hypothesis is:
H2c. A brand’s communication style emphasizing its roots increases perceived continuity.
As mentioned before, in this age of globalization customers are more and more in search for their identity. They may find this through consumption of the products of a brand
15
that helps them to construct their identity (Arnould & Price, 2000; Leigh et al. 2006). When the characteristics of a brand are clear and support the development of the self-identity of customers, this may help customers to construct who they are and to support them in being their true selves (Aaker & Fournier, 1995; Morhart et al., 2015). It is therefore expected that a brand with clear human characteristics and features may increase the perceived symbolism of customers (Morhart et al., 2015). The accompanying hypothesis is:
H3. Brand anthropomorphism increases perceived symbolism. 2.2 Marketing skepticism
Obermiller & Spangenberg (1998) define skepticism toward advertising as ‘the tendency toward disbelief of advertising claims’ (p.160). Skepticism may develop when there are clear discrepancies between the claims made in marketing communications or when the claims made in marketing communications are difficult to verify (Ford, Smith & Swasy, 1990). The study of Obermiller, Spangenberg & MacLachlan (2005) suggests that the more skeptic customers are, the lesser they like advertisements, the lesser they rely on it and the lesser they attend on it. However, their study also suggests that customers are not totally immune for advertisements because they respond more positively to emotional claims in advertisements than to informational claims in advertisements. According to the article of Beverland & Farrelly (2010), a customer should be seen as ‘an active customer of authenticity instead of a passive receiver of information’. Customers thus do not simply accept attempts of brands trying to be perceived as authentic, but they evaluate the effects of the cues themselves. According to the Persuasion Knowledge Model, people react different to persuasion attempts of brands (Friestadt & Wright, 1994). Their response to these persuasion attempts depends on the topic knowledge, persuasion knowledge and agent (the one people see as responsible for the persuasion attempts) knowledge they have (Friestadt & Wright, 1994). The more persuasion coping knowledge people have, the more they are able to select and execute effective and suitable coping tactics (Friestadt & Wright, 1994). Campbell & Kirmani (2000) state that customers make use of their persuasion knowledge when there are clear ulterior motives of the brand accessible and when customers have the cognitive capacity. Consequently, the degree of marketing skepticism differs per person and per situation.
The potential risk of trying to be perceived as a highly authentic brand is the formation of a ‘doppelgänger brand image’: ‘a cultural backlash in the form of ‘a collection
16
of disparaging images and stories about a brand that are circulating in popular culture by a loosely organized network of customers, anti-brand activists, bloggers and opinion leaders in the news and entertainment media’ (Thompson, Rindfleisch & Arsel, 2006, p.50). Thomson et al. (2006) use Starbucks as an example: Founded as a single coffee shop 30 years ago, it has grown out to a worldwide known brand. The fact that they still position themselves as authentic has led to an astounding number of anti-Starbucks attacks. As there is probably more information available about the topic and the agent of mass market brands such as Starbucks, customers may become more skeptic about the marketing activities of these mass market brands. Therefore, it may be harder for mass market brands to be perceived as authentic.
It is expected that attempts of brands to express their heritage or virtue or to come across as being real, may be perceived as fake or as ulterior motives by customers (Morhart et al., 2015). The more topic knowledge, persuasion knowledge and agent knowledge customers have about brands, the more marketing skeptic they probably will be. Consequently, this may influence the relationship between the use of authentic cues and PBA. Therefore, the following hypothesis is proposed:
H4. Skepticism toward marketing weakens the relationship between iconic cues and the PBA dimensions continuity, integrity and credibility.
2.3 Consequences of PBA
The study of Napoli et al. (2016) demonstrates that brand authenticity is not fixed over time, but will position itself along a continuum. This is due to the decisions the brands make and the change of context of the brand. Although their findings suggest that authentic brands deliver the highest customer- and brand derived value, even when a brand has a moderate level of brand authenticity, they can still derive value (but to a lesser amount) (Napoli et al., 2016). Different strategies exist that can be used by brands to position themselves as authentic (Napoli et al., 2016). Dependent on where a brand stands according to the authenticity continuum, brands need to emphasize one of the different dimensions to deliver value. For example: brands that are new in a market need to gain trust and thus have to focus on ‘reliability’. Several studies show the advantages for brands with a high PBA: higher expected quality and trust (Moulard et al., 2016) and an increase in emotional brand attachment and WOM (Morhart et al., 2015).
17 2.3.1 Emotional brand attachment
According to the study of Price & Arnould (1999) customers and brands can develop commercial friendships on the basis of different elements: instrumentality, sociability and reciprocity. Although most friendships based on instrumental purposes are not likely to succeed, this is different for relationships between brands and customers, as those relationships are based on material benefits (Price & Arnould, 1999). Interactions between a brand and a customer provide opportunities for sociability. Reciprocity indicates that a friendship can vary along a continuum from agentic to communal. The longer a customer interacts with a brand, the more the commercial relationship transforms into a communal relationship (Price & Arnould, 1999). In this way customers become emotionally attached to a brand and perceive the brand as a friend. Morgan & Hunt (1994, p. 23) define commitment as ‘an exchange partner believing that an ongoing relationship with another is so important as to warrant maximum efforts at maintaining it; that is, the committed party believes the relationship is worth working on to ensure that it endures indefinitely’. Customers who are highly committed to a brand identify themselves more with the brand and they hold feelings of attachment to maintaining the valued relationship (Harrison-Walker, 2001). Commitment is a multidimensional concept that exists out of two components: affective commitment and continuance commitment (Allen & Meyer, 1990). Cognitive continuance commitment occurs when there is a ‘profit associated with continued participation and a cost associated with leaving’ (Kanter, 1968, p. 504). Affective commitment is the ‘affective or emotional attachment to the organization such that the strongly committed individual identifies with, is involved in, and enjoys membership in the organization’ (Allen & Meyer, 1990, p. 2) and the essence of it is that customers come to acquire an emotional attachment to a brand in the relationship (Fullerton, 2003). Because both components are very different from each other, they are not mutually exclusive and they vary along a continuum (Allen & Meyer, 1990). Both forms of commitment may be beneficial for a brand, but affective based commitment relationships are more effective on aspects such as brand advocacy and willingness to pay (Fullerton, 2003) and WOM (Westbrook, 1987).
According to Belk (1988, p 160) ‘we are what we have’, which means that we form our identity around our possessions. The social identity theory of Tajfel & Turner (1979) suggests that people derive their self-concept from their perceived membership of a relevant social group, and as mentioned above people also derive their self-concept from their possessions (Belk, 1988). Being more specific, we form our sense of self out of different social identities: the more important a specific identity is for ourselves, the more attractive
18
the accompanying products are (Kleine, Kleine & Kernan, 1993). Escalas & Bettman (2003, 2005, 2009) state that when brands make use of brand images that are in line with the ingroup (desirable reference groups and celebrities) of the customer, the customer is more likely to include this brand into their self-concept because it helps them to communicate their identity to others and to achieve one’s desired self. The study of Reimann, Castaño, Zaichkowsky & Bechara (2011) shows the neurological process of decreasing emotional arousal with regard to a brand over time, while inclusion of the brand into the self increases over time. Although identification with a brand may generate positive consequences for brands, a brief comment has to be made. Chernev, Hamilton & Gal (2011) suggest that the customer’s search for their identity can not only be satiated by a brand’s direct competitors, but also by unrelated brands (other product-categories) that help customers in their search for their identity. Thus, by positioning themselves as a ‘lifestyle brand’, brands may distinct themselves from their direct competitors. However, they still have to consider other ‘lifestyle brands’ in different product categories.
Emotional brand attachment happens when customers have feelings towards a brand (Thomson, MacInnis & Park, 2005) and it can be seen as the strength of the bond connecting the brand with the self (Park et al., 2010). The properties of brand attachment are reflected by two factors: brand-self connection and brand prominence (Park et al., 2010). Brand prominence is the degree of positive feelings and memories about a brand in relation to other brands. The more prominent a brand, the more as top of mind these positive feelings and memories are and thus the stronger the emotional brand attachment is (Park et al., 2010). Brand-self connection is the cognitive and emotional connection between the brand and the self (Park et al., 2010). Mittal (2006) states that the customer’s ‘self’ is comprised out of five
components, which are: 1. values and character, 2. competence and success, 3. social roles, 4.
body image, and 5. self-perceived personality traits. It differs per customer on which of these components their emphasis lies and what the proportion of each component is in their ‘self’. For example, some people value ‘competence and success’ more than their ‘body image’. This will influence the way someone behaves and articulates oneself. This five-component model of Mittal (2006) helps companies to understand what to communicate in order to engage customers whose self is in line with the components ‘that the brand can authentically engage’ (p.560). When a brand’s marketing communications are in line with important components of customers, there may be a stronger brand-self connection. Subsequently, the emotional brand attachment will be stronger. However, as the importance of these five
19
components differs per customer, the effects of these marketing communications on emotional attachment differ as well.
As mentioned before, according to the study of Park et al. (2010) brand attachment is the ultimate goal for brands in establishing a commercial relationship. Customers who are strongly attached to a brand, are more willing to forsake personal resources (e.g. investments
in time, money and energy) for this brand (Park et al., 2010). According to Malär, Krohmer,
Hoyer & Nyffenegger (2011) brands can increase emotional brand-attachment of customers by focusing their brand personality on the customer’s true selves instead of customer’s aspired selves. Brands should thus focus their communication on the components of the ‘true selves’ of customers who are related to the concept of the brand.
Arnould & Price (2000) state that the globalization influences the way we identify ourselves. Boundaries that functioned before as a reference point have become fluid because of this globalization and the rise of the Internet. To overcome this identity lessness, customers use objects or brands which they perceive as authentic to form their own identity (Arnould & Price, 2000), which is called an ‘authenticating act’. According to the article of Beverland & Farrelly (2010) customers assess the authenticity of products with personal goals in mind. The decision making process of assessing something as authentic stems from the desire for the real, true and genuine and is driven by a desire to draw different identity benefits (control, connection and virtue). In this way, customers adapt themselves to this era of globalization, by finding authenticity in different objects and linking this authenticity to themselves and their personal goals (Thomson et al., 2005). According to the study of Morhart et al. (2015) PBA increases emotional brand attachment. As mentioned above, to authentically engage customers, brands can engage in certain marketing activities. It is likely that customers respond positively to this, as authentic brands may help people in their search for identity in this age of globalization. Besides that, brands that are perceived as authentic may be evaluated as more credible, integer and continuous. When customers perceive a brand as authentic, they may show reciprocal dedicated behavior in the form of emotional brand attachment. It is therefore expected that brands that score high on PBA will also score higher on emotional brand attachment.
H5. Perceived brand authenticity has a positive effect on emotional brand attachment.
20 2.3.2 Word of Mouth
The traditional conceptualization of WOM is according to Godes et al. (2005) ‘the
one-to-one and face-to-face exchange of information about a product or service’ (2005, p. 416). Brown et al. (2005, p. 125) state that WOM communication is ‘any information about a target object (e.g. company, brand) transferred from one individual to another either in person or via some communication medium’. Because of lower trust of the traditional media, increasing product complexity and greater availability of information, WOM has become more and more important (Godes et al., 2005) According to the study of Trusov, Bucklin & Pauwels (2009) WOM has twenty times more positive effect on new customer acquisition than traditional marketing channels have. The study of Chevalier and Mayzlin (2006) suggests that online book reviews positively affect the sales of books and the study of Bone (1995) shows that WOM has an influence on short-term and long-term judgements of products.
According to the article of Morhart et al. (2015), PBA has a direct positive effect on customer’s WOM. However, in this study it is proposed that this relationship is mediated by emotional brand attachment. Commercial friendships as mentioned above are associated with positive WOM as an outcome (Price & Arnould, 1999) and commitment is most positively correlated to WOM when compared with brand satisfaction (De Matos & Rossi, 2008). The studies of Westbrook (1987), Harrison-Walker (2001) and Brown et al. (2005) suggest that affective commitment is positively related to WOM. This is in line with the study of Fullerton (2003) who suggests that affective commitment has a positive effect on customer advocacy. The studies of Brown et al. (2005) and De Matos & Rossi (2008) suggest that positive WOM is less dependent on satisfaction of the customer when this customer is highly committed to the brand.
The accompanying hypotheses therefore are:
H6a. Perceived brand authenticity has a positive effect on word of mouth.
H6b. Emotional brand attachment mediates the relationship between perceived brand attachment and word of mouth.
21
It is expected that there also exists a direct relationship between the authentic cues and emotional brand attachment and WOM and that this relationship is mediated by PBA. For example, customers may feel themselves lesser emotional attached to a brand or are lesser willing to provide WOM if they have heard about brand scandals. This because these customers probably will evaluate this brand as less integer (Morhart et al., 2015). The same applies for the other authentic cues, but in a positive way. Brands expressing their virtue in their marketing communications can appeal more reciprocal behavior of customers, because customers evaluate the brand as integer (Morhart et al., 2015). On the other hand, when customers are very skeptic about marketing, this relationship will probably be less strong (Obermiller & Spangenberg, 1995; Obermiller, Spangenberg & MacLachlan, 2005; Morhart et al., 2015). Therefore, it is expected that the mediation between the iconic cues and emotional brand attachment and WOM is moderated by marketing skepticism. The following hypotheses are stated:
H7a. The effect of authentic cues on (a) emotional brand attachment and (b) word of mouth is mediated by perceived brand authenticity.
H7b. Marketing skepticism moderates the mediation between (a) iconic cues and emotional brand attachment and (b) iconic cues and word of mouth.
22 2.4 Conceptual model
Based on the hypotheses stated above, the following conceptual model is developed: Figure 1. Conceptual model
23 3. Method
In this part the empirical part of the study will be provided. First, the characteristics of the collected sample and the research design will be explained. Secondly, the measures and measurement scales of the variables are described.
3.1 Procedure
To answer the research question and to test the developed conceptual model, a quantitative study in the form of a survey was conducted. The survey had a cross-sectional between-subjects design. The survey was administered online via Qualtrics and it consisted of closed fixed-response questions which required respondents to select an answer from a predetermined set of responses. By providing a lively introduction text, the survey tried to motivate and uplift the respondents to become more involved and to complete the survey.
At first, the respondents were randomly assigned to a commercial of seven different mass market brands. It was expected that based on four of the commercials (Dove 1, Rabobank, Albert Heijn and Pickwick) respondents would perceive the brand as highly authentic. According to several marketing related websites, the ratio behind the commercials of these four brands is to display the real stories from real customers or employees (e.g. Eck, 2016 and Brinks, 2016). According to an article about the Rabobank commercial, a Dutch study has shown that customers evaluate the Rabobank as more authentic because of this (http://www.marketingtribune.nl/bureaus/nieuws/2015/05/[commercial]-rabobank-maakt-hypotheek-persoonlijk/index.xml). It was expected that based on the other four commercials (Vanish, Ariel, Telfort and Dove 2) respondents would perceive these as less authentic, since they are selected on their scripted content.
Afterwards, respondents were provided with Likert scale questions in the same sequence. These Likert scale questions tapped into authentic cues, marketing skepticism, PBA, emotional brand attachment and WOM. At the end of the survey, the respondents were asked some demographic questions (e.g. age, educational background and gender). These questions were stated at the end of the survey since some of these questions might be considered as sensitive. A bogus question was asked to identify careless responses (e.g.: fill in ‘somewhat disagree’ to answer this question).
3.2 Measures
All constructs were measured using a seven point Likert-type scale. To ensure construct validity, scales from previous studies were adapted wherever possible.
24
Authentic cues were measured with a validated fourteen item scale, completed with one additional scale for iconic cues. For indexical cues, the validated three item seven Likert scale (1=strongly disagree, 7=strongly agree) with a Cronbach’s alpha of 0.89 from Morhart et al. (2015) was used to measure brand spanning boundary spanning agent behavior (e.g. The personal appearance of the employee of this brand is in line with appearance of the brand) and for brand scandals the validated one item seven Likert scale (1=never, 7=very often) from Wagner, Lutz & Weitz (2009) was used (e.g. How often have you heard/read about scandals pertaining this brand?)
For iconic cues, the validated three item seven Likert scale (1=strongly disagree, 7=strongly agree) with a Cronbach’s alpha of 0.82 from Morhart et al. (2015) was used to measure a brand’s communication style emphasizing a brand’s virtue (e.g.: The communication activities of this brand focus on delivering its promise to customers). Besides that, the validated four item seven Likert scale (1=never, 7=very often) with a Cronbach’s alpha of 0.82 from Morhart et al. (2015) was used to measure a brand’s communication style
emphasizing a brand’s roots (e.g.: The communication activities of this brand focus on
heritage). Additionally, to measure the effects of a real story communication style, a three item seven Likert scale (1=strongly disagree, 7=strongly agree) was added to the survey (e.g. the persons in the commercial are real customers of the brand).
For existential cues, the validated two item seven Likert scale (1=strongly disagree, 7=strongly agree) with a Cronbach’s alpha of 0.89 from Morhart et al. (2015) was used to measure brand anthropomorphism (e.g. I can easily imagine this brand as a person).
PBA was measured with the validated 15 item scale of Morhart et al. (2015) with each
a Cronbach’s alpha of >0.7. Continuity was measured with the four item seven Likert scale (e.g.: this is a timeless brand). Integrity was measured with the four item seven Likert scale (e.g.: this is a brand with moral principles). Credibility was measured with the three item seven Likert scale (e.g.: this is a brand that will not betray you). Symbolism was measured with the four item seven Likert scale (e.g. this is a brand that adds meaning to people’s live).
The seven item seven Likert (1=strongly disagree, 7=strongly agree) with a Cronbach’s alpha of 0.83 of Gaski & Etzel (1986) was used to measure marketing skepticism (e.g. Most advertising provides customers with essential information).
25
To measure emotional brand attachment, the ten item seven Likert scale (1=strongly disagree, 7=strongly agree) with a Cronbach’s alpha of 0.89 of Thomson et al. (2005) was used (e.g. Please indicate to what extent you agree that affectionate describes the brand).
WOM was measured with the three item seven Likert scale (1=strongly disagree,
7=strongly agree) with a Cronbach’s alpha of 0.96 of Price & Arnould (1999) (e.g. I would recommend this brand to someone who seeks my advice).
To test for the common method bias, centrality of work was measured with the three item seven Likert scale (1=strongly disagree, 7=strongly agree) of Miller, Woehr & Hudspeth (2002) with a Cronbach’s alpha of 0.84 (e.g. Even if I inherited a great deal of money, I would continue to work somewhere).
3.3 Sample
The population of interest of this study were Dutch customers who were familiar with the companies under investigation. The research was conducted using a non-probability convenience sample and was based on anonymous respondents. Respondents were approached via Facebook, e-mail, Whatsapp and face to face. This method was appropriate for this study as the population of interest for this study was basically the entire Dutch population and in this way it was possible to gain access to a large group of potential respondents. Demographic information (gender, age and educational background) was assessed to measure the similarity of the sample and to reassure that the survey was filled in by Dutch respondents only. The minimum number of responses needed for the results to be analyzable was calculated using the estimation of Miles & Shevlin (2001). This study measures 16 different constructs (demographic variables included) thus to find medium effects, a sample of 150 respondents was needed.
26 4. Results
In this chapter, the results of the data analysis will be presented. First, a summary of the respondents participated in this study will be provided. Second, a description of the data preparation will be given, followed by the actual hypotheses testing.
4.1 Respondents
Out of the 244 people that started the survey, 152 finished the survey completely, which equals a completion rate 62.3%. Of these 152 people, 31 people filled in the bogus question wrongly. Therefore, only 121 of the respondents were taken into account in the analysis. 34.2% of the respondents were male, 65.8% of the respondents were female. Over 80% (N=96) of the respondents were between 19 and 26 years old. Looking at education 96.7% (N=117) of the participants were highly educated (HBO or university). All demographics can be found in appendix 1.
4.2 Data preparation
Of the 244 respondents who filled in the survey, 123 were listwise deleted because of incomplete surveys or wrong answers to the bogus question. Afterwards, a few counter indicative items were recoded in order to measure the reliability of the measures with the Cronbach’s alpha.
Almost all of the constructs that are used in the survey for this study were adopted from previous studies and have proven to be reliable by having a Cronbach’s alpha higher than 0.7 in their original form. However, a few constructs have been adjusted for this study. To test the reliability of the (adjusted) constructs, the internal consistency method is used. One item that was used to measure the brand congruent boundary spanning agent behavior variable (‘what this person tells is not at odds with what the brand promises’), did not correlate with the other two items and was therefore deleted. A potential explanation for this is that this question contained a double denial, what made it potentially harder interpretable for the respondents. After the deletion of this item, the Cronbach’s alpha was 0.765, and has therefore an acceptable reliability level. The same applies for a brand’s communication style emphasizing its virtue. However, the Cronbach’s alpha was still weak (0.642) after deletion of the item (‘connection with the customers’), but it was decided to use this measure since it was a validated scale. As for the rest of the variables, no items needed to be deleted. Real story communication style -the only variable that was developed for this study- Cronbach’s alpha was 0.83.
27
Finally, the scale means of all items that were used to measure the variables were calculated. In order to give an overview of the variables used in this study a correlation matrix was developed with the variables’ means, standard deviations and reliability values (table 1). These outcomes indicate that a strong correlation exists between the variables, except for brand scandals and brand congruent boundary spanning agent behavior.
An independent sample t-test was conducted beforehand to compare the PBA dimensions in the real and fake scenario conditions. On average, respondents in the real
scenario scored higher on the PBA dimensions, except for credibility (continuity: Mreal
scenario=5.353, SDscenario=0.876, Mfake scenario=4.558, SDfake scenario=1.094, t(119)=4.364, p=0.000;
integrity: Mreal scenario=4.536, SDreal scenario=1.031, Mfake scenario=4.062, SDfake scenario=1.183,
t(119)=2.332, p=0.021; symbolism: Mreal scenario=4.174, SDreal scenario=1.335, Mfake scenario=3.177,
SDfake scenario=1.303, t(119)=4.151, p=0.000 and credibility: Mreal scenario=4.744, SDreal scenario=1.119, Mfake scenario=4.400, SDfake scenario=1.066, t(119)=1.730, p=0.086).
29 4.3 Hypotheses testing
In order to test the aforementioned hypotheses, structural equation modeling (hereafter: SEM) was executed in AMOS. As covariance is the basic statistic of SEM, the
goal of SEM is to test correlations between variables and to explain as much as possible of
their variance (Kline, 2005). By doing this, it was tested whether there exists a causal relationship between the independent variables and the dependent variables, that is: if the independent variables explain the variance in the dependent variables (Byrne, 2010). Therefore we expect linearity, additivity, independency and normality of the variables. The advantage of SEM is that instead of analyzing the variables of the model separately, SEM allows simultaneous analysis of all the variables in the model. Subsequently, there is lesser chance for measurement errors (Nachtigall, Kroehne, Funke & Steyer, 2003). Tables can be found in appendix 3,4 and 5.
4.3.1 Hypotheses 1 a+b
H1a. Based on the results, brand scandals have no significant effect on
perceived integrity as the path coefficient was -0.130 with a p-value of 0.275. Therefore, this hypothesis is rejected.
H1b. Brand congruent boundary spanning agent behavior has a direct significant effect on credibility as the path coefficient is 0.224 with a p-value of 0.005. For this reason, this hypothesis is accepted.
4.3.2 Hypotheses 2 a+b+c
H2a. This hypothesis is accepted as a brand’s communication style
emphasizing its virtue increases perceived integrity (path coefficient=0.308 with a p-value < 0.01).
H2b. The effect of a brand’s real story communication style on perceived
credibility is significantas the path coefficient is 0.268 with a p-value < 0.01. Subsequently,
this hypothesis is accepted.
H2c. A brand’s communication style emphasizing its roots increases perceived continuity as the path coefficient is 0.273 and the p-value is < 0.01. This hypothesis is accepted.
30 4.3.3 Hypothesis 3
H3. Based on the results, brand anthropomorphism increases perceived
symbolism since the path coefficient is 0.338 with a p-value < 0.01. This hypothesis is accepted.
4.3.4 Hypothesis 4
H4. Based on the results, there is no negative main effect of marketing
skepticism on integrity as the path coefficient is 0.049 and the p-value is 0.929. There is also no moderation effect as the interaction between marketing skepticism and a communication style emphasizing a brand’s virtue is not significant (path coefficient= -0.099 and p-value=0.359). The results indicate that there is a negative main effect of marketing skepticism on credibility as the path coefficient is -0.539 and p-value= <0.01. However, there is no moderation effect as the interaction between marketing skepticism and a real story communication style is not significant (path coefficient=0.023 and p-value is 0.632). There is no direct effect of marketing skepticism on continuity (path coefficient=0.311 with p-value=0.314) and there is also no significant interaction effect of marketing skepticism on continuity (path coefficient=-0.104 with p-value=0.134). As a consequence, this hypothesis is rejected.
4.3.5 Hypothesis 5
H5. a. Based on the results, both continuity and symbolism have a
significant effect on emotional brand attachment as the path coefficient of continuity is 0.278 with a p-value of <0.01 and the path coefficient of symbolism is 0.270 with a p-value of <0.01. Credibility and integrity do not have a significant effect on emotional brand attachment as the path coefficient of credibility is 0.100 with a p-value of 0.114 and the path coefficient of integrity is 0.158 with a p-value of 0.115. This hypothesis is accepted.
4.3.6 Hypotheses 6a+b
H6a. This hypothesis is accepted because both credibility (path coefficient=0.402 with a p-value <0.01) and integrity (path coefficient=0.250 and p-value <0.01) have a significant effect on WOM. For the other two PBA dimensions, the effect on
WOM is not significant (path coefficientsymbolism=0.007 with p-valuesymbolism=0.933, path
31
H6b. There is a direct effect of emotional brand attachment on WOM (path
coefficientemotional brand attachment=0.320 with p-valueemotional brand attachment=0.011). However, as
only credibility and continuity have a significant direct effect on WOM, only these PBA dimensions may be explained by emotional brand attachment. Since the path coefficient of credibility on emotional brand attachment is insignificant (see hypothesis 5a), this relationship will not explain the relationship between credibility and WOM. Only continuity has a significant effect on both emotional brand attachment and WOM. However, since there is no relationship between the other two PBA dimensions and WOM, this hypothesis is rejected.
4.3.7 Hypothesis 7a+b
H7a. a. For this hypothesis the user defined estimand of Gaskin was used.
This hypothesis is partially accepted as the path coefficients of two out of six of the relationships between the authentic cues and emotional brand attachment were mediated by the PBA dimensions. These authentic cues are as follows: a brand’s communication style emphasizing its roots has a path coefficient of 0.064 with a p-value of 0.027 and brand anthropomorphism has a path coefficient of 0.081 with a p-value of 0.001. As for the other relationships between the authentic cues and emotional brand attachment, no mediating effects were found. A real story communication style has a path coefficient of 0.034 with a p-value of 0.363, brand congruent boundary spanning agent has a path coefficient of 0.028 with a p-value of 0.299, a brand’s communication style emphasizing its virtue has a path coefficient of 0.047 with a pvalue of 0.161 and brand scandals has a path coefficient of -0.020 with a p-value of 0.221.
b. For this hypothesis the user defined estimand of Gaskin was used as well. This hypothesis is partially accepted, as three out of the six relationships between the authentic cues and WOM are mediated by the PBA dimensions. The relationship between real story communication style and WOM has a path coefficient of 0.108 and a p-value of 0.033 and the relationship between brand congruent boundary spanning agent behavior and WOM has a path coefficient of 0.090 with a p-value of 0.030. The relationship between a brand’s communication style emphasizing its virtue and WOM has a path coefficient of 0.119 and a p-value of 0.009. For the other three authentic cues, no significant effects were found. The relationship between brand anthropomorphism and WOM has a path coefficient of 0.028 and a p-value of 0.282. The relationship between brand scandals and WOM has no significant effects: path coefficient=-0.050 with p-value=0.221. The relationship between a brand’s
32
communication style emphasizing its roots and WOM has a path coefficient of -0.007 and a p-value of 0.788.
H7b. a. There exists a direct negative effect of marketing skepticism on emotional brand attachment (path coefficient=-0.948 with p=0.049). However, as no significant mediating effect between a brand’s communication style emphasizing its virtue and emotional brand attachment and a real story communication style and emotional brand attachment can be found (see H7a-a), there is also no moderating effect on this relationship. For the mediating effect of PBA on the relationship between a brand’s communication style emphasizing its roots and emotional brand attachment, no significant interaction effect was found (path coefficient=0.106 with p-value=0.08). As there are no significant moderating effects, this hypothesis is rejected.
b. Although there is a significant mediating effect of PBA for a brand’s
communication style emphasizing its virtue and WOM (see H7a-b), no moderating effect of marketing skepticism can be found (path coefficient=-0.035 with p-value=0.758). As there was no mediating effect of PBA between a brand’s communication style emphasizing its roots and WOM, there was also no moderating effect. For the mediation of PBA between a real story communication style and WOM, a significant moderating of marketing skepticism effect was found (path coefficient=-0.182 with p-value=0.008). However, as there is only one moderating effect, this hypothesis is rejected.
33
Table 2. Hypotheses summaries
Nr Hypothesis Accepted/
rejected? H1a. Brand scandals decrease perceived integrity. Rejected H1b. Brand congruent boundary spanning agent behavior
increases perceived credibility.
Accepted
H2a. A brand's communication style emphasizing its virtue
increases perceived integrity.
Accepted
H2b. A brand’s real story communication style increases
perceived credibility.
Accepted
H2c. A brand’s communication style emphasizing its roots
increases perceived continuity.
Accepted
H3. Brand anthropomorphism increases perceived symbolism. Accepted
H4. Skepticism toward marketing weakens the relationship
between iconic cues and the PBA dimensions continuity, integrity and credibility.
Rejected
H5. Perceived brand authenticity has a positive effect on
emotional brand attachment.
Accepted
H6a. Perceived brand authenticity has a positive effect on word
of mouth
Accepted
H6b. Emotional brand attachment mediates the relationship
between perceived brand authenticity and word of mouth.
Rejected
H7a. The effect of authentic cues on (a) emotional brand
attachment and (b) word of mouth is mediated by perceived brand authenticity.
a. Partially accepted b. Partially accepted
H7b. Marketing skepticism moderates the mediation between (a)
authentic cues and emotional brand attachment and (b) authentic cues and word of mouth.
a. Rejected b. Rejected
34
Figure 2. Model with path coefficients
35 5. Discussion
5.1 General discussion
The first hypothesis could not be proven. Although this hypothesis was supported by the article of Morhart et al. (2015), this study indicates that there is no direct relationship between brand scandals and perceived integrity. A potential explanation for this may be the different cultural background of the respondents, or because the brands chosen for this study did not appeal to brand scandals. For the other authentic cues, there was a strong relationship between these cues and the PBA dimensions (H1b, H2a+b+c and H3). This indicates that brands can positively influence the PBA of their brand via these cues, while the only negative authentic cue (brand scandals) did not have an effect on PBA in general.
Marketing skepticism does not moderate the relationship between the iconic cues (a brand’s communication style emphasizing its virtue, a brand’s communication style emphasizing its roots and a real story communication style) (H4). This hypothesis was also supported by the article of Morhart et al. (2015). A potential explanation for this may be that the marketing communications used in the survey were more emotionally oriented than informationally oriented. According to the above mentioned study of Spangenberg & MacLachlan (2005), customers react more positively to emotional claims than to informational claims. Therefore it could be that while customers are skeptic about marketing, they still consider the brands used in this study as integer, continuous and credible. Besides that, it might be that the respondents in this study had lesser persuasion knowledge or that the ulterior motives of the brands were not that easy accessible. Subsequently, H7b-a and H7b-b were rejected, as there was no significant moderating effect on the mediation between the three iconic cues (a brand’s communication style emphasizing its virtue, a brand’s communication style emphasizing its roots and a real story communication style).
For emotional brand attachment, it appears that both brand anthropomorphism and a brand’s communication style emphasizing its roots are important indicators (H7a-a). Consequently, the PBA continuity and symbolism are important, as these mediate the relationships between these authentic cues and emotional brand attachment (H5). The other authentic cues and PBA dimensions did not have any effect on emotional brand attachment. This is in contrast with the article of Morhart et al. (2015), since they suggest that there is a direct effect of symbolism, credibility and integrity on emotional brand attachment. According to this article, there is no effect of continuity on emotional brand attachment.
