Fuel or crush the luxury dream
‘’Do bloggers have an influence on the brand image of luxury brands?’’
University of Amsterdam
Faculty of Economics and Business
Msc. In Business Administration – Marketing track
Supervisor: dr. Mossinkoff
By Emma van der Straaten
11094095
Hand in date: 24th of June 2016
2
Statement of originality
Statement of originality
This document is written by Emma van der Straaten who declares to take full responsibility for the
contents of this document.
I declare that the text and the work presented in this document is original and that no sources
other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of
the work, not for the contents.
3
Acknowledgement
With the completion of this thesis, the final step towards achieving my Masters’ degree in Business Administration is completed. From the moment we had to choose a topic for our thesis, I knew I wanted to do something which related to bloggers. However, I did not expect the topic to be that interesting not only for me but also for others to read. The outcomes have surprised me not only because some weren’t as expected but also because some were exactly as I expected which made this thesis ‘’fun’’ to write. However, writing this thesis has not only been an interesting learning experience regarding to the topic but also for myself. By completing this thesis and my Masters’ I have learned that by hard work and dedication one can achieve a lot which helped me to gain more self-confidence. Furthermore it has taught me that sometimes you have to take a step back and let things come as they go.
I would like to thank first of all my supervisor for always giving me new motivation after our feedback sessions and for stimulating me to look in different corners with his expertise in the field. Also, a big thank you for my old university, Nyenrode for letting me execute my experiment for three days which consumed a lot of their time as well. And last but not least, to my family and friends for believing in me when I did not, making me less worried and especially for their patience in the process of dealing with me.
I hope you will enjoy reading my thesis and I hope that it will raise awareness and interest in this practical topic that is very valuable in today’s society.
4
Abstract
In the last decade, a new form of the traditional opinion leaders has emerged, the online opinion leaders, or in this research referred to as bloggers. The influence of bloggers has been covered in current literature, however this is relatively limited and no research has yet been performed, to our knowledge, where bloggers are linked to a key marketing concept: brand image. Due to the symbolic meaning and trend setting character of luxury brand, it is even more interesting to investigate whether bloggers would have an influence on the brand image of a luxury brand. Therefore this research is dedicated to answer the following question: ‘’Do bloggers have an influence on the brand
image of luxury brands?’’. This research extension is relevant since it brings awareness towards a
new field in marketing and helps to improve current literature about brand image and luxury brands. Furthermore it has many managerial implications for luxury brands related to their brand image and marketing tactics. First a Social Media study among the ten biggest bloggers worldwide has been conducted to investigate whether bloggers actively emphasize luxury brands on Instagram. Second, an experimental survey has been conducted to investigate if bloggers have an influence on the different aspects of the brand image of luxury brands. The experiment consisted of three different experiment groups and was targeted to three different luxury brands which have been carefully chosen in order to represent luxury brands as a whole. The main findings of this study indicate that bloggers do have an influence on the brand image of luxury brands but not on all aspects. Bloggers have an influence on the types of associations of luxury brands, user imagery, usage imagery, feelings evoked, perceived benefits, attitude and the actual perceived associations. The main influence appears to be that bloggers are able to portray a casual, younger image in which it was perceived that the brand could be worn in everyday situations and would evoke feelings of fun and excitement. However for the favorability and uniqueness of those associations no evidence was found that bloggers have an influence. After performing the social media analysis these results are especially interesting since here is concluded that bloggers do not actively emphasize brands, but they do have an influence.
5
Table of content
Statement of originality ... 2 Acknowledgement ... 3 Abstract ... 4 1. Introduction ... 7 2. Literature review ... 102.1 Web 2.0, social media and bloggers ... 10
2.1.1 Social media ... 10
2.1.2 Bloggers and social media ... 11
2.1.3 The influence of bloggers ... 12
2.2 Brands ... 14
2.3 Brand image ... 14
2.4 Luxury brands ... 16
2.4.1 The influence of bloggers on luxury brands ... 18
3. Conceptual framework ... 20
3.1 Hypothesis ... 21
4. Methodology ... 23
4.1 Research design ... 23
4.2 Experimental design ... 23
4.2.1 Treatments per group ... 24
4.2.2 Executing the experiment ... 25
4.3 Survey ... 26
4.4 Analytical strategy ... 28
4.5 Social media ... 30
5. Social Media analysis ... 32
5.1 Photo analysis ... 32 5.2 Text analysis ... 33 6. Results ... 35 6.1 Survey ... 35 6.1.1 Types of associations ... 35 6.1.2 Favorability of associations ... 40 6.1.3 Uniqueness of associations ... 42
7. Discussion and conclusion ... 44
6
7.2 Overall conclusion ... 51
7.3 Theoretical implications ... 52
7.4 Managerial implications ... 53
7.5 Limitations ... 55
7.6 Suggestions for further research ... 55
References ... 57
Appendix 1: Survey ... 62
Appendix 2: Experiment ... 65
Appendix 3: Bloggers included in Social Media research ... 69
Appendix 4: Measurement level Social Media research ... 70
Appendix 5: Social Media Analysis ... 72
List of tables Table 1: Descriptive analysis: mean and standard deviation hypothesis 1c ... 39
Table 2: Levene’s test of equality of error variances hypothesis 1c ... 39
Table 3: Repeated measure anova for hypothesis 1c. ... 40
Table 4: Descriptive analysis: mean and standard deviation hypothesis 2 ... 41
Table 5: Levene’s test of equality of error variances hypothesis 2 ... 41
Table 6: Repeated measure anova for hypothesis 2 ... 42
Table 7: Descriptive analysis: mean and standard deviation hypothesis 3 ... 42
Table 8: Levene’s test of equality of error variances hypothesis 3 ... 42
Table 9: Repeated measure anova for hypothesis 3 ... 43
List of figures Figure 1: Luxury brands versus masstige brands… ... 18
Figure 2: Conceptual Framework ... 20
Figure 3: Analytical model derived from Keller (1993) ... 21
7
1. Introduction
The both literary and practical definitions of traditional opinion leaders are widely known and adopted in literature of the early 20th century. Traditional opinion leaders are those individuals that are likely to influence other people in their immediate environment (Katz, 1957). They function as both a source and as a guide. However in today’s society, we can see a new version of these
traditional opinion leaders: online opinion leaders or also known as bloggers. The concept of bloggers has gained attention throughout the last year among a lot of different brands but most importantly, they have gained a lot of attention from consumers. Bloggers are perceived as potential powerful makers and breakers of brands (Hansen & Schneiderman, 2010). Therefore they can be perceived as early adopters of innovations in both products as services and are able to spread those innovations among others in both offline as online environments. However, their influence does not stop after the innovation phase, even throughout the whole product cycle bloggers advocate products which make them an important element of the customer engagement cycle. That what allows bloggers to execute their power in the degree they do, could be argued to be due to wide reach on social networks such as Facebook and Instagram and their own website/blog that they have (Wu & Wang, 2011). Social media functions as the carrying vehicle of the content allowing fans to interact and connect with other fans by using (e)WOM (Muniz & Schau, 2011). In a recent study among millennials it was found that 33% of the respondents mostly relied on blogs before making a
purchase (Luxury Society, 2016). Furthermore, even for luxury brands bloggers can be of influence. It is said that they are able to constantly fuel the luxury dream by posting content on a regular, if not daily, basis (Luxury Society, 2016). That bloggers have an influence is a fact, and that they are able to create brand awareness as well. However, no research has yet been dedicated to what the influence of bloggers is, if they have any, on a key marketing concept: brand image. Keller (1993, p. 3) defines brand image as ‘’ perceptions about a brand as reflected by the brand associations held in consumer memory. Brand associations are the other informational nodes linked to the brand node in memory and contain the meaning of the brand for consumers’’. Keller (1993) classified the types of
8 association into three categories of increasing scope: first, attributes, the descriptive features that characterize the product which can be both product related as non- product related. Second, are the benefits and lastly, the attitudes. These brand associations then vary in three dimensions: first in terms of favorability, strength (not included in this research) and uniqueness of brand associations. Especially for luxury brands, brand image is of high importance since luxury brands put much more effort in creating a symbolic meaning than other brands. Furthermore, luxury manufacturers often determine the fashion trends. This is also where the extraordinariness of a luxury product originates: from a different design or construction principle which is later adopted by mass market
manufacturers (Goody, 2006). Therefore it is even more interesting if these new forms of traditional opinion leaders are able to have an influence on the brand image of luxury brands. Would bloggers be able to influence all aspects of brand image such as the non-product related associations of a luxury brand while luxury brands are the ones who set the tone and are only worn by a limited extraordinary target-group? As indicated, no research has yet been dedicated, to our knowledge, to investigating whether bloggers have an influence on the brand image of luxury brands. It is therefore that this research will address this gap in the literature by aiming to add to existing literature of not only that related to online opinion leaders, but also to that of a relatively old but key marketing concept: brand image, and provide valuable insights to improve current literature about these concepts and that of luxury brands. However, also no research has yet been indicated to what bloggers exactly do on Social Media, most importantly, if they even actively emphasize luxury brands on Social Media. The implications of this research will therefore contribute to not only literature but will also have significant managerial implications which can be used globally by luxury brands whatever the outcome is regarding their brand image and marketing. The aim of this research is to investigate the following ‘’Do bloggers have an influence on the brand image of luxury brands?’’. Where also a study will be conducted related to whether bloggers actively emphasize luxury brands on Social Media to provide valuable managerial implications and insights. Accordingly the main research objectives are:
9 1. Investigate whether bloggers have an influence on the types of attributes of a luxury brands
and identify what this influence is, if there is any
2. Investigate whether bloggers have an influence on the favorability of associations of luxury brands and identify what this influence is, if there is any
3. Investigate whether bloggers have an influence on the uniqueness of associations of luxury brands and identify what this influence is, if there is any
4. Investigate whether bloggers actively emphasize luxury brands on Social Media
The research is structured as following. First a literature review will be provided which reviews the current literature related to the three concepts, online opinion leaders, brand image and luxury brands. Then in the next chapter the conceptual framework and hypothesis will be provided including a graphic representation of both the conceptual framework as analytical model. Then thirdly, the sample and research method will be explained followed by a detailed chapter about the results of both the Social Media analysis as the experiment. And lastly, in the final chapter an extensive discussion will be provided on the findings and questions that arise. This last chapter also provides an overall conclusion, theoretical implications, managerial implications, limitations and suggestions for further research.
10
2. Literature review
This chapter will provide a review of the literature about the key concepts of this research. The key concepts are bloggers, brand image and luxury brands. Furthermore a review will be provided of what is known about the concepts that will be used to test the effect of bloggers on consumers. Here the focus will be on brand image. Based on this review a research gap will be identified which follows to the research question.
2.1 Web 2.0, social media and bloggers
In the 21st century a dynamic and interconnected international environment has developed. The origin of the term Web 2.0 comes from the O’Reilly Media Web 2.0 Conference of 2004. Web 2.0 makes it possible for consumers to both create and consume information instead of just merely retrieving information from the World Wide Web (WWW). It facilitates inexpensive content creation and interaction which puts the laymen as center stage instead of the business in terms of the design, collaboration and communities on the WWW (Berthon, et al., 2012).
2.1.1 Social media
Social media is often referred to as user-generated content (UGC). The media are simply the vehicle that carry the content, such as Instagram, Facebook, Twitter and Pinterest and the content that is shared on these media in forms such as words, photos and videos is created by the consumer. The content shared by consumers can be perceived as value for businesses (Muniz & Schau, 2011) . Kaplan & Haenlein (2010, p.63) argue the following definition of social media: ‘‘a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of user-generated content.’’ Social media content consists of text, pictures and videos.
Social media began with text, which was initially in the form of a blog, where the word blog is short for the word ‘’web logs’’. Berthon, et al. (2012, p.264) defines blogs as: ‘’blogs are websites owned and written by individuals who maintain regular commentaries and diaries that may include text, graphics, videos, and links to other blogs and web pages and are usually presented in reverse
11 chronological order.’’ Social media networks are a different type of product, as social networks leverage the power of relationships and the collective wisdom by enabling a service where one can add and contact friends and update personal messages (Berthon, et al., 2012) . Nowadays the four types of social media content are nearly intertwined which increases the power of social media since it is highly accessible and scalable and calls for social media dialogues (many to many). As Hanna, et al. (2011, p.271) summarize correctly ‘’the real power of the social media ecosystem is that we are all connected’’.
2.1.2 Bloggers and social media
So what are bloggers and how do they make use of social media? In order to explain what bloggers are it is good to review the characteristics of traditional opinion leaders since common characteristics can be observed between the two. Opinion leaders act as intermediaries in the web of social
interactions. Katz (1957, p.64) states the following definition of an opinion leader: “the individuals who were likely to influence other persons in their immediate environment”. Where Weimann (1994) adds on by saying that the wide set of personal connections play a key influential role.
McQuail & Windahl (1993) elaborate further on this by stating that opinion leaders function as both a source and as a guide. Nisbet & Kotcher (2009, p.334) stress that it is not the formal power of the opinion leaders where their significance is relied on but rather on ‘’ the connective communication tissue that alert their peers to what matters among political events, social issues, and consumer choices”. Nowadays face-to-face contact is not a necessity for the presence of opinion leaderships anymore. Because of social media and the thereby created online communities, a new form of the traditional opinion leader is being acknowledged in recent literature: the online opinion leader. As stated above, common characteristics between bloggers and traditional opinion leaders can be observed. Both have a discussant role and are influential and valued among others. Therefore we can regard the new form of the tradition opinion leader, the online opinion leader, as a blogger.
12
2.1.3 The influence of bloggers
In recent literature the influence of bloggers is stressed. Hansen & Schneiderman (2010, p.86) state that bloggers are ‘’ potentially powerful makers and breakers of brands, political candidates, and news stories’’. Which is in line with the statement of Lyons & Henderson (2005, p.325) which states that ‘’bloggers have higher levels of enduring involvement, innovativeness, exploratory behavior, self-perceived knowledge, and more expertise within the product category’’. When looking into the innovation model of Rogers (2003) it can be argued that bloggers can be perceived as the early adopters of innovations in both products as services and that they then, due to their influence, have the ability to also spread the innovations among other adopters in both offline and online
environments. Therefore bloggers can take a significant role in providing opportunities for brands and thus can be perceived ‘’as the new force in the two-step flow of communication’’ (Uzunoglu & Kip, 2014, p. 394) . Sashi (2012) adds on by this with his customer engagement model of which the last stage before customer engagement can be argued to be a blogger activity: advocacy. Here he demonstrates that a blogger is not only an early adopter, but one who advocates the product further along in its product development cycle. Sashi (2012) furthermore developed a customer engagement matrix where he describes four types of customers: transactional, delighted, loyal and fans. Where both the delighted customer as the fan can be an advocate for a product, brand or company because these customers have an affective commitment which is positively related to generating (e)WOM (Harrison-Walker, 2001). Electronic type of Worth of Mouth (eWOM) are the shared photos, videos, experiences and opinions. (Cheung & Thadani, 2012) Social media, as the carrying vehicle of content it is, thus allows fans to interact and connect with other fans by using (e)WOM. However, more importantly, fans can also interact and connect via social media with potential customers and turning them into transactional customers which increases the possibility of their transition through the subsequent stages of the customer engagement cycle. All types of eWOM, the one directly created by a blogger or generated by interacting of multiple bloggers is more trusted by consumers than corporate messages because it is perceived to be what the consumer experiences by its own.
13 Therefore it is no surprise that marketers notice a major shift and change in practice nowadays. In a recent survey among millennials, 33% stated that they rely mostly on blogs before they make a purchase (Luxury Society, 2016). That this allows bloggers to execute their power in the degree that they do, could be argued to be due to their wide reach on social networks such as Facebook and Instagram and their own website/blog they have. (Wu & Wang, 2011).
The above stated findings have contributed to the literature about online opinion leaders, or as we will now refer to in this research, bloggers and their influence. However relatively little
research has been performed about this topic. What was specifically mentioned throughout the relatively limited literature about this topic is the importance of Social Media since this is basically the whole carrying vehicle for all the (e)WOM content that bloggers create and the platform which enables them to have the power they have. However, what bloggers exactly do on social media is not discussed while it is quite interesting to see whether bloggers even actively emphasize brands on Social Media. Furthermore it has not yet been researched what bloggers exactly do for these brands except creating brand awareness.
Keller (1993, p4) developed a conceptual model of brand equity from the individual
consumers’ perspective. In this model he defines, customer-based brand equity as ‘’the differential effect of brand knowledge on consumer response to the marketing of the brand.’’ Where brand knowledge consists of both brand image and brand awareness. Brand awareness consists of both brand recall and brand recognition. From the above stated findings we can conclude a significant influence of bloggers on creating brand awareness, among others. However no findings were found on what the influence of bloggers is on brand image, a key concept of marketing.
14
2.2 Brands
Before diving deeper into brand image, a short review about what a brand exactly is is needed. Stern (2006) developed a scheme which categorizes the construct of a brand according to four categories: nature, function, locus (world and mind) and valence (positive and negative). Jevons (2007, p.7) performed a historic meta-analysis and identified six key components of brand definitions which led him to develop the following definition ‘’A brand is a tangible or intangible concept that uniquely identifies an offering, providing symbolic communication of functionality and differentiation and in doing so sustainably influences the value offered’’. However Buchanan-Oliver, et al. (2008) conflate this classification to three components: symbolic, functional and psychological. Furthermore they argue that Jevons (2007) does not capture the experiential component enough and thereby they want to stress the need to address the experiential dimension of the brand as well.
2.3 Brand image
As indicated above, Keller (1993) developed a conceptual model of brand equity from the individual consumers’ perspective which consists of brand image and brand awareness. For this research, brand image is of particular interest.
Even though brand image has been recognized as a key concept in marketing, there is no universal definition of brand image. The most widely adopted definition is from Keller (1993, p. 3) who defines brand image as ‘’ perceptions about a brand as reflected by the brand associations held in consumer memory. Brand associations are the other informational nodes linked to the brand node in memory and contain the meaning of the brand for consumers.’’. However the favorability,
strength and uniqueness of these brand associations can vary and play an important role in
determining the differential response that makes up brand equity. Before elaborating on these three dimensions, it is useful to look into the different types of brand associations. Keller (1993) classified the types of association into three categories of increasing scope: first, attributes, the descriptive features that characterize the product which can be both product related as non- product related. Second, benefits, the personal value that a consumers attach to a product, consisting of functional,
15 experiential and symbolic benefits. And last, attitudes, defined by (Wilkie, 1986 in Keller, 1993, p. 4) as ‘’the consumers’ overall evaluations of a brand’’. Attitudes often form the basis for behavior. As stated earlier, these brand associations vary in three dimensions: first in terms of favorability, according to how favorable they are evaluated. Second, in terms of strength, this is related to the strength of connection to brand node. Depending on encoding and storage of the information inside the brain. Last, uniqueness of brand associations is related to whether the associations are shared with other competing brands. For this research the brand equity model of Keller (1993) and in particular his definition of brand image will be used as the basis of the analytical model since it’s the most widely adopted definition of brand image in current literature.
For the purpose of this research product related attributes are not included. The reason this is not seen as added value to the research is that bloggers can not transmit enough information through their content for one to judge about the product to make product related associations. One other part will also not be included in the research and that is the strength of associations. The reason for this is that the strength of associations is dependent on how information enters a consumers’ memory (encoding) and how it is then maintained as a part of the brand image in the consumers’ mind (storage). This is beyond the scope of this research due to its complexity.
As stated by Keller (1993, p.6) ‘’ the presence of strongly held, favorably evaluated associations that are unique to the brand and imply superiority over other brands is critical to a brand's success’’ and this will then lead to a positive brand image. Positive brand image increases customer loyalty and the probability of brand choice and decreases vulnerability to competitors’ marketing actions. Furthermore, positive brand image enables the brand to command larger margins and have more inelastic responses to price increases. Especially those consumers with a positive brand attitude are willing to pay a premium price for the brand (Starr & Robinson, 1978 in Keller, 1991, p.9). Also the effectiveness of marketing communications is increased with a positive brand image and high levels of brand awareness. Where even those consumers who are positively
16 predisposed towards the specific brand require less reinforcement through marketing
communications (Keller K. , 1993).
Especially for luxury brands, brand image is highly important because luxury brands put more effort in creating a symbolic meaning than other (even premium) brands and want people to believe in a certain luxury dream which makes them pay the maximum price level of a product in its category compared to other brands. Furthermore, luxury manufacturers often determine the fashion trends. This is also where the extraordinariness of a luxury product originates, from a different design or construction principle which is later adopted by mass market manufacturers (Goody, 2006, p. 344)
Would bloggers be able to influence all aspects of brand image such as the non-product related associations of a luxury brand while luxury brands are the ones who set the tone and are only worn by a limited extraordinary target-group? Looking at previous findings about their significant influence, this seems likely. Since we are interested in whether bloggers have an influence on brand image and the interesting and unique character of luxury brands, within this research the influence will be specifically tested on the brand image of luxury brands.
2.4 Luxury brands
The luxury retail market continues to thrive ever since 2011 after a short decline in global sales in 2010 due to the global economic crisis (Verdict, 2014). The most significant rise can be seen in Asia Pacific and a slight increase in the Middle East. In a time period from 2011-2015 the share of
expenditure on luxury branded products has risen over 6% in Asia Pacific (excluding Japan). However the biggest market for luxury products can be found in Europe, which accounts for 33% of global sales (Verdict, 2014) .
However what is luxury and what are luxury brands? The word luxury is derived from ‘’luxus’’ which means according to the Oxford Latin dictionary (1992) ‘’soft or extravagant living, (over)-indulgence and sumptuousness, luxuriousness and opulence’’. Marketing academics use ‘luxury’’ in
17 different ways and there is not really a consensus about what the definition of a luxury brand or product really is. However there is one basic definition that most academics share and that is that luxury is something more than necessary (Bearden & Etzel, 1982 p 184). The difference between necessity and luxury is mainly related to the availability, where necessities are available and
possessed by virtually everyone, while luxuries are that only to a very few people. Due to increases in living standards in the 19th century, it was added that luxury is also more than ordinary. Kemp (1998, p 592) added to the definition by stating that luxury is not only based on these two concepts but also on people’s desire for it. It can therefore be concluded that ’’luxury is anything that is desirable and more than necessary and ordinary’’ (Heine, 2012, p. 42). When adopting this definition of luxury, the following definition can be adopted for luxury brands ‘’Luxury brands are associated with products which exceed what is necessary and ordinary compared to other products of their category’’ (Heine, 2012, p. 51) However, these definitions are still vague regarding to what exactly are luxury products manufactured by luxury brands. A study has been performed by Heine (2012) of which the results suggested that luxury products have six primary characteristics which range from a minimum level to a maximum level that corresponds to the highest form of luxury and which are obviously not
independent from each other. These six primary characteristics are: high level of price, quality, aesthetics, rarity, extraordinariness and symbolic meaning (Heine, 2012). Luxury brands are highly associated with their core products (Kapferer, 2008, p. 193). Therefore the characteristics of luxury products accord to those of luxury brands: the image that is in the mind of the consumer about a luxury brand comprises the associations related to the six characteristics of luxury products. Any (potential) luxury brand can be evaluated by these six characteristics. It is important that the core product of a luxury brand is clearly associated with a high level of these six characteristics in order to achieve a luxury brand image and to expand the product portfolio.
Nowadays a lot of brands offer prestige to the masses: ‘’masstige brands’’. These brands focus especially on creating benefits and prestige. However due to their offering to the masses, they have a relatively lower level of rarity, prices and often have to compromise on quality,
18 extraordinariness and aesthetics (Keller K. , 2009, p. 295). Therefore, they can not be considered as a status luxury brand which have a high level in all characteristics. See figure 1 for a graphic illustration of the level difference between luxury and masstige brands. However, some luxury brands do apply this masstige strategy by creating a product range with masstige products. However, preserving the luxury image and exclusivity is a challenge. However, the evaluation of luxury brands should refer to the core product and therefore most luxury brands can afford to extend their product line with masstige brands as a way of generating growth.
Florin, et al. (2007) argue therefore that luxury items are now more attainable for more people, calling this ‘’democratization of luxury’’. This is due to not only the trading-down strategy of luxury brands but also due to for example rise in income, the trend that people spend less on certain goods so they can save for luxury items and globalization. On top of this democratization of luxury, they observe a constant focus on celebrity culture. Therefore people are argued to be more sophisticated, more up-to-date on trend about what is cool and what is fashionable.
Within this study the definition of luxury brands will refer to those brands which have high levels on all six of the primary characteristics: price, quality, aesthetics, rarity, extraordinariness and symbolic meaning.
Figure 1: Luxury brands versus masstige brands
2.4.1 The influence of bloggers on luxury brands
Even for luxury brands, some influence can be noticed of bloggers. Luxury brands are commonly seen in the content among the worlds’ biggest bloggers even though luxury brands can only be afforded by few people. For luxury brands bloggers can help fueling the ‘’luxury dream’’
19 among their readers. And since bloggers keep posting content on a regular, often daily, basis, they constantly refuel this dream towards their consumers. (Luxury Society, 2016) One could argue that this must be bad news for luxury brands, since as described earlier, rarity is one of the important characteristics that make a luxury product a luxury product. However, Veblen (1994) argues that those who purchase high priced items do so in order to communicate wealth and achieve a certain social status. This is in line with what Leibenstein (1950) calls the ‘’Veblen effect’’, whereby demand for a good rises because its price is high rather than low. He also refers to an interpersonal effect of ostentatious consumption: the ‘’bandwagon’’ effect, where demand rises when others in the reference group purchase the good. Knowing of the Veblen effect and bandwagon effect, bloggers communicating about luxury brands could be argued to rise the demand for luxury brands and therefore refuel the dream for luxury brands. Especially for ‘’star’’ luxury brands the above described effects are exactly what they are looking for. Star luxury brands strive for maximum brand awareness which goes far beyond their actual target group. Reason being that one of the major purchasing motives of the consumers of these brands is to demonstrate wealth to other people, often to those who can’t afford it. Generating brand awareness by their non-target group is needed in order to fulfill the need of their target group consumer for demonstrative consumption (Kapferer & Bastien, 2009).
No research has yet been dedicated to investigate what the influence is of those super powerful new online opinion leaders on the brand image of a luxury brand. Even though brand image for luxury brands is even more important than for any other type of brand. It is therefore that this research will address this gap in the literature by aiming to add to existing literature of not only that related to online opinion leaders, but also to that of a relatively old but key marketing concept: brand image. The implications of this research will contribute not only to literature but also has managerial implications which can be used globally by luxury brands. The aim of this research is to investigate the following ‘’Do bloggers have an influence on the brand image of luxury brands?’’.
20
3. Conceptual framework
Within this chapter the conceptual framework, hypothesis and analytical model of this study will be briefly discussed.
Figure 2: Conceptual framework
Figure 2 demonstrates an overview of the conceptual framework of this study. The overall purpose is to see whether on three main elements of brand image a change has occurred after blogger
exposure. As described in the literature review, bloggers are known to have an effect on the brand awareness together with other various effects. For example as Rogers (2003) indicates, they take a significant role in providing opportunities for brands by being early adopters, by which Sashi (2012) adds that bloggers advocate products throughout the whole product development stage and thereby increase customer engagement of others. Related to luxury brands, bloggers keep refueling the ‘’luxury dream’’ and therefore possibly maintain the luxury brand image. Taking this into account and the fact that nowadays 33% of the millennials rely mostly on blogs before making a purchase, I assume that bloggers do have an influence on the brand image of luxury brands. However, brand image consist of different aspects, as indicated earlier, therefore the influence needs to be tested on each of these aspects to see whether there is a difference of influence per aspect.
21
3.1 Hypothesis
As indicated above, I assume bloggers have an influence on the brand image of luxury brands based on the findings in current literature. However, whether the effect is positive or negative or any other direction is not measured since this is subjective and can therefore not be objectively measured. After removing the earlier indicated elements (product related attributes and strength of
associations) from Keller’s (1993) brand equity model, the following analytical model can be drawn which illustrates exactly what aspects of brand image is included in this study (see figure 3):
Figure 3: Analytical model derived from Keller (1993)
Hypothesis 1: bloggers have an influence on the types of associations of luxury brand H1a: Bloggers have an influence on the non-product related attributes
H1b: Bloggers have an influence on the perceived benefits H1c: Bloggers have an influence on the attitude
H1d: Bloggers have an influence on the perceived associations
Hypothesis 2: Bloggers have an influence on the favorability of associations of luxury brands Hypothesis 3: Bloggers have an influence on the uniqueness of associations of luxury brands
22 In order to make the concept ‘’luxury brands’’ measurable to test the above named hypothesis, three different luxury brands have been chosen which represent the luxury brand concept well. As
described earlier, there are six characteristics which make a luxury brand, a luxury brand: high level of price, quality, aesthetics, rarity, extraordinariness and symbolic meaning (Heine, 2012).However, there are two other criteria: they need to be globally well-known and not completely a like in order for the three brands to represent ‘’luxury brands’’. Furthermore, by choosing three brands the results will be more generalizable than if only one luxury brand is chosen.
Taking this in mind, the following three luxury brands have been chosen: Chanel, Louis Vuitton and Dior. Where Chanel positions itself as a classy brand with a strong heritage and focus on its founder Coco Chanel and her classy ideals of fashion. Louis Vuitton positions itself a little more contemporary than Chanel and focusses itself on a timeless, authentic style which also applies to the younger audience by offering an edgy touch (Chevalier & Mazzalovo, 2012, p. 23). Last, Dior positions itself as a cool, trendy but glamorous and luxurious brand and is known for its feminine designs (Okonkwo, 2007, p. 109). All three brands have in common that they have a strong heritage, score a high level on all six characteristics, are globally well known and represent a diverse set of strong luxury brands. Because they represent a diverse set of luxury brands, one can generalize the results of the hypothesis to luxury brands as a whole concept.
23
4. Methodology
In order to investigate whether bloggers influence the brand image of luxury brands, quantitative data has been collected. Before the hypothesis were tested it was first investigated if bloggers are even actively emphasizing luxury brands in their social media. Results of the Social Media research can found in chapter 5, results related to hypothesis in chapter 6. This chapter will describe the used samples, research design and procedure, the analysis method and measures and scales used to answer the proposed hypothesis. Furthermore at the end of this chapter the sample, procedure and analysis method of the Social Media research is discussed.
4.1 Research design
In order to investigate if bloggers have an influence on the brand image of luxury brands, a deductive quantitative research approach has been used. To be specific, an experimental survey has been conducted in order to gather structured data with a high internal validity due to the ability of manipulation within experiments.The reason for this approach is that this would provide an extensive data collection which allowed to compare the data of pre and posttest and therefore provide valuable insights in whether changes have occurred between the two time measurements. In other words, if an influence was apparent. This deductive quantitative research approach has led to the development of new insights related to the relation of influencers and brand image. This research can therefore be considered as explanatory research (Saunders & Lewis, 2012).
4.2 Experimental design
The experimental design was as follows: the experiment consisted of three groups in which each group would receive a different treatment, in which group one was the control group and group two and three the experiment groups. The experiment had a pretest-posttest design in order to compare the participant groups and measure if a change occurred as a result of the specific treatment each group received. In order to ensure a strong level of internal validity, all groups have been pretested before being post tested. This way it could be ensured that each group, most likely, started off from
24 the same foot and that therefore the groups were equivalent. This has been tested for each group during the data analysis in order to assure that the randomization process indeed was effective. (See chapter 5). The data collection method consisted of a survey which contained six identical surveys and was about three different luxury brands. Which meant that for each brand there were two surveys, one for the pretest and one for the posttest. See figure 4 on p.26 for a graphic illustration of the experimental design.
4.2.1 Treatments per group
As stated above, each group received a different treatment of which the effects were measured in the posttest. All three groups received the same survey and also the same pretest for each of the three brands. During the experiment the respondents in each group were presented a PowerPoint slide with four images related to the specific brand in question. For each of the three brands there were two slides: one slide for the pretest containing brand communication and one slide which contained the treatment. For each slide the respondents filled in an identical questionnaire consisting of ten questions.
The pretest consisted of four images of a specific brand, these images were campaign photos of the brand itself. In other words, this was the communication coming the brand itself. The reason for choosing images of the brand itself is that in order to investigate whether bloggers have an influence on the brand image of a luxury brand, one must first examine what the perceived brand image is the respondents have of the brand. Then when exposed to the treatment which contains images of bloggers, one can compare the results of the pre and posttest. The four images all looked somewhat alike to prevent confusion about the brand communication and on each image the brand name was shown. After the questionnaire was filled in for this specific slide, another slide was provided. This slide was the treatment slide which differed per group. Group one was provided with a slide which contained four images which were all from the same campaigns as the previous slide, so basically no treatment. This is because group one was the control group where no significant
25 difference in answers is supposed to arise after treatment. Since this research wants to examine whether bloggers have an influence on the brand image, for group two and three the treatment included four images of bloggers. However, the treatment differed between the two groups. Group two was shown four images of bloggers who wore the brand, in as far as possible, the same way as the brand tried to communicate with their own brand communication shown in the previous slide. Group three were shown four images of bloggers who wore the brand in a completely opposite way as the brand tried to communicate with their own brand communication shown in the previous slide. See appendix 2 for all slides for each group.
4.2.2 Executing the experiment
Before executing the experiment a pilot study was executed among six people in order to see if there were any ambiguous items, spelling mistakes or for example wrong scales, the weekend before the experiment would be executed. There were some aspects in the survey which were interpreted incorrectly and were therefore adjusted before the real execution of the experiment. Also, the images in the PowerPoint were pilot tested among these six people to see whether there were any critics or necessary improvements related to the image choice. This lead to the decision to change one image for one brand due to the fact that this image was seen as a misfit and was therefore perceived as confusing. By this pilot test of both the PowerPoint as the survey the reliability and validity of both aspects was tested.
The experiment was executed for two full days on Monday the 25th of April and Tuesday the 26th of April and for 3 hours in the morning on Friday the 29th of April at Nyenrode New Business School in Amsterdam. In total nine classes have been examined during these days, these were all the first and second year students. For each experiment group there were three classes that were randomly assigned to one of the three experiment groups. This did however have as a consequence that not all groups had the same amount of respondents. In total 114 respondents participated in the experiment. The reason for executing it all within one university is because this allowed me to limit
26 external influences between groups and therefore raise internal validity. Furthermore, this way it could be assured that all nine group were examined in a similar research setting. By excluding any external influences in each group and examine a relatively homogenous sample in terms of age, education and social status, it would be more likely that the randomization of the experiment would be effective and therefore that the experiment would succeed in the best possible way. Also by executing the experiment together with a group, one can assure that the respondents fill in each question so no missing values or wrongly interpreted answer would occur. Also this way the
respondents could ask questions when something was unclear and therefore the responses would be likely to be of value.
Before starting the experiment one slide was prepared with instructions for the survey in order to make sure no confusion or missing values would occur and to inform the respondents about their rights. No information was provided about the purpose of the study, which group the
respondents were in, the research question or other similar information since this might bias the respondents during the experiment.
Figure 4: Experimental design
4.3 Survey
As described earlier, there was one survey consisting of six identical questionnaires and each group was presented the same questionnaire. The questions asked were all related to the concept of brand image by Keller (1993). For each part of brand image certain questions were asked. See appendix 1
27 for the survey questions. All question were based on valid and reliable theoretical measures derived mostly from or sometimes slightly modified from the academic article of Keller (1993) or his book (Keller K. , 2013). Six questions were asked related to the types of associations. Firstly, the non-product related attributes were measured with four items, user imagery related to age, user imagery related to style, usage imagery related to situations and feelings evoked by the brand. All items related to user imagery and usage imagery were derived from Keller (2013), the item related to feelings evoked were derived from Kahle, Poulos and Sukhdial (Kahle, Poulos, & Sukhdial, 1988) . Secondly, the perceived benefits were measured with one item related to what need the respondent thinks the brand would fulfill. This item has been derived from Park, Jaworski, and MacInnis (Whan Park, Jaworski, & MacInnis, 1986). Third, attitude was measured via a 5-point likert scale with two items, one related to ‘’I love this brand’’ and one ‘’I would buy from this brand’’. This item was derived from Keller (1993). After these six questions were answered the respondents were asked to name three brand related associations on the basis of what they had just filled in. This was asked through an open question. This item was derived from Keller (1993). The favorability of the three associations were measured via a 5-point likert scale where the respondents had to indicate how favorable it is for the brand to have this specific association. This item was derived from Keller (1993). The uniqueness of the three associations has also been measured via a 5-point likert scale where the respondents had to indicate how unique the association is. This item has been derived from Keller (1993). On the last page of the survey two more question were asked related to the respondents gender and age which were both asked through a multiple choice question.
The survey was designed in Qualtrics (2016) and then exported into a Word document so it could be printed for the experiment. In the footnote via unrecognizable coding for the respondents the group was identified so that during the analysis it was known in which group the respondent was.
28
4.4 Analytical strategy
After gathering all the data, the data was coded into IBM SPSS. Since there were no missing values all the responses could be used.
The first 5 variables of the survey have a nominal measurement level. For these variables no descriptive statics such as mean and standard deviation could be calculated due to their nominal measurement level. Therefore these variables have been analyzed per brand. As described in the conceptual framework, the three brands represent luxury brands and therefore could be generalized and be referred to as luxury brands. In order to measure whether there is a significant relation between which group the respondents were in and their answer provided for the specific variable, cross tabs have been made including the chi-square test. In order to ensure the randomization was effective, for both the pretest as the posttest these tests have been run. If in the pretest there were no significant results, one could assume the randomization for that variable was effective. In this case, all pre-tests showed no significant results, therefore we can assume the randomization was effective. The exact significance levels of the pre-test can be provided on request as adviced by supervisor due to the extensiveness of the data. In case more than 20% of the cells had an expected count less than 5 a Fisher-exact test has been performed instead of a Pearson Chi-square. For one variable, user imagery style, a Monte Carlo correction had to be run for two brands since SPSS was unable to calculate the chi-square values due to not enough memory in the software of SPSS.
For the one item related to the three associations the answers had to be categorized in SPSS since this was an open question. Fortunately the instruction at the beginning of the experiment have had an effect since most answers were the same and the categorization process was quite easy. In order to then analyze this nominal data, the same steps have been taken as for the above described nominal variables.
Attitude, favorability and uniqueness have all been measured via a 5-point likert scale which means they have an ordinal measurement level. Attitude has been measured via two different
29 questions with a 5-point likert scale, ‘’I love this brand’’ (1 = not at all, 5 = a great deal) and ‘’I would buy from this brand’’ (1 = definitely not, 5 = definitely yes). For this variable a reliability check (Cronbach Alpha) has been run. Cronbach’s Alpha = .863 which indicates a high level of internal consistency. Also the corrected item-total correlations indicate that all items have a good correlation with the total score of the scale (all above .551). Also none of the items would substantially affect reliability if they were deleted (Δ<.10). In order to analyze this variable, two new variables have been created as a function of existing variables. Since each respondent answered the same two questions related to attitude in both the pretest as the posttest for each of the three brands, a scale mean has been calculated for each respondent: ‘Attitude pretest’’ and ‘Attitude posttest’’. Also, the favorability of associations has been measured via a 5-point likert scale (1 = extremely bad, 5 = extremely good). For this variable a reliability check (Cronbach Alpha) has been run. Cronbach’s Alpha = .744 which indicates a high level of internal consistency. Also the corrected item-total correlations indicate that all items have a good correlation with the total score of the scale (all above .345). Also none of the items would substantially affect reliability if they were deleted (Δ<.10). In order to analyze this variable, two new variables have been created as a function of existing variables: ‘’Favorability pretest’’ and ‘’Favorability posttest’’. The same steps have been followed for the uniqueness of associations which was also measured by using a 5-point likert scale (1 = not at all unique, 5 = extremely unique). For this variable a reliability check (Cronbach Alpha) has been run. Cronbach’s Alpha = .839 which indicates a high level of internal consistency. Also the corrected item-total correlations indicate that all items have a good correlation with the total score of the scale (all above .433). Also none of the items would substantially affect reliability if they were deleted (Δ<.10). In order to analyze this variable, again two new variables have been created as a function of existing variables: ‘’Uniqueness pretest’’ and ‘Uniqueness posttest’’. After these scale means have been calculated, for all variables descriptive statistics, skewness, kurtosis and normality tests have been run. In order to analyze whether there is a significant effect a repeated measure ANOVA has been run for all three variables.
30
4.5 Social media
A Social Media analysis of the ten biggest bloggers in the world, in terms of followers and exposure, was performed based on their Instagram posts (see appendix 3 for an overview of the bloggers chosen and their characteristics). As described earlier there is relatively little research performed regarding what bloggers exactly do on Social Media. However the importance of Social Media is stressed in almost all existing literature and since that is where bloggers communicate from, I have collected my own data by executing a social media research. This is in order to find more information on whether bloggers actively emphasize brands on Social Media as I perceive this as a valuable contribution to this research. Screenshots of the analysis in Excel can be found in Appendix 5. (The excel sheets including all calculations can be provided on request)
The reason for choosing Instagram is because this is the most widely adopted Social Media platform among bloggers and has the highest upload frequency per week among all platforms a blogger uses (Smith, 2016). The analysis started on the 5th of April and was finished on the 8th of April. Each blogger has been analyzed over the same time span, two months back from the date the analysis was started. Within this time span of each post that was related to each of the three brand an analysis has been performed, one for the picture and one for the text under the picture. The main variables that are interesting for this research are for the photo: is the brand recognizable, is the brand tagged, how central is the brand in the picture and are there other brands tagged in the picture. For text it is also interesting to know if the brand is tagged, what the brand centrality is in the text and if other brands are tagged. By combining the data of the analysis of both photo and texts some interesting insights can be provided. Since this data set has not been tested in SPSS in terms of significance effects and relations, no generalizable conclusions can be drawn and therefore all conclusions drawn from this hypothesis are based on this analysis only. See appendix 4 for all the measured variables and what their coding means in respect of the analysis. In total 190 photos and 190 texts have been analyzed, which makes a total of 380 items being analyzed. The numbers are not equal per brand since one brand was posted more frequently within the time span than of the
31 others. For Chanel the total number was 78, Louis Vuitton 53 and Dior 59. Since all accounts of the bloggers are open for public and I also follow them personally, no ethical issues arise. Also nothing has been screenshotted or saved, everything has been purely analyzed straight from the account.
32
5. Social Media analysis
As described earlier there is relatively little research performed what bloggers exactly do on Social Media. However since the importance of Social Media is stressed in almost all existing literature and since that is where bloggers communicate from, I have collected my own data by executing a social media research. This is in order to find more information on whether bloggers actively emphasize brands on Social Media. I believe that it would be a powerful contribution to the research question if we would know if bloggers even emphasize brands on their Social Media. It is likely to assume that they do, since like Hansen & Schneiderman (2010) state they are powerful makers and breakers of brands. But by investigating whether they really do emphasize brands actively, answers are provided which either way have implications for the outcome of the main research question: ‘’Do bloggers have an influence on the brand image of luxury brands’’. Imagine that the results show that bloggers do have an influence on all aspects or some of brand image and bloggers also actively emphasize bloggers than this has different implications than if other outcomes in each of the results would appear. The research methods for the results have been discussed in the previous (Methodology) chapter, in total 190 photos and 190 texts, so 380 items have been analyzed. The results will be briefly discussed below.
5.1 Photo analysis
Chanel: Most remarkable in the photo analysis of Chanel is that in 73% of the pictures the brand was
not tagged however on 91% of the pictures the brand was recognizable. Also the brand centrality was either moderate (51%) or high (49%) and in 64% of all the pictures the brand was the center of attention. Another remarkable aspect is that on only 21% of the pictures another brand was tagged. However 66% of the times there was no other brand tagged, Chanel also wasn’t tagged.
Louis Vuitton: Most remarkable in the photo analysis of LV is that in 51% of the pictures the brand
was tagged, however 26% of these tags come from the same blogger. Furthermore the brand was on 79% of the photos clearly recognizable while 62% of those pictures were not tagged. Related to
33 brand centrality, just as with Chanel, the brand either moderately central (57%) of highly central (43%) and in 57% of all the pictures the brand was the center of attention. Also in only 34% of the photos other brands were tagged. 40% of the times when no other brand was tagged, LV also wasn’t tagged, or in 28% of the cases where no other brand was tagged, there was only a tag of LV.
Dior: Most remarkable in the photo analysis of LV is that in 53% of the pictures the brand was tagged,
however most of these tags have to do with collaborations with the brand. In 75% of the pictures the brand was clearly recognizable, in 51% of the photos the brand centrality was high (2) and in 59% of the photos the brand was the center of the attention. Another remarkable aspects is that 76% of the photos did not have a tag of another brand. If other brands were tagged, in 71% of the cases Dior was also tagged.
5.2 Text analysis
Chanel: Most remarkable in the text analysis was that the brand was tagged even less than in the
photos, in only 19% of the analyzed posts the brand was tagged in the text. If the brand was tagged in the text, it was 47% of the times related to a Chanel Fashion Week show. 9% of the posts was tagged in both the text as in the photo and there the brand centrality was also in all cases high (2). In 18% of the posts the brand was tagged in the photo but not in the text. Furthermore related to brand centrality there are some remarkable points. In contrast to the photos, in 81% of the posts the brand centrality in the text was 0 and only 15% of the post had a high brand centrality (2). In this particular case where the brand centrality was 2, the brand was also always tagged in the text. In every case Chanel was not tagged in the text, the brand centrality in the corresponding photo was either moderate (1) or high (2). Last but not least, in 85% of the posts other brands were also not tagged of which 82% of the cases no other brands were tagged, Chanel also wasn’t tagged. 62% of all post did not include a tag of Chanel in both photo as text.
Louis Vuitton: Most remarkable in the text analysis was that the brand was tagged less than in the
34 brand was tagged, 44% of the texts were related to a Louis Vuitton fashion week show. 28% of the posts was tagged in both the text as in the photo and there the brand centrality was in 73% of the cases high (2) or else moderate (1). In every case LV was not tagged in the text, the brand centrality in the corresponding photo was either moderate (1) or high (2). In 75% of the analyses texts, no other brands were tagged of which 65% of the cases no other brands were tagged, Louis Vuitton also wasn’t tagged. The rest of the times Louis Vuitton was the only brand tagged in the text. 47% of all post did not include a tag of LV in both photo as text
Dior: Most remarkable in the text analysis was that the brand was tagged more than in the photos, in
56% of the analyzed posts the brand was tagged in the text and of all these text the brand centrality was high (2). 33% of those texts were Dior was tagged were used to express positive feelings about the product of the brand, 33% was related to a Dior Fashion Week show and 15% was related to collaborations between the blogger and the brand. In 51% of the texts the brand centrality was high (2). Remarkable for Dior is that in 93% of the texts, no other brands are tagged. If there were other brands tagged, than Dior was also tagged except for one single case. A last remarkable note is that in every case Dior was not tagged in the text, the brand centrality in the corresponding photo was either moderate or high. 25% of all post did not include a tag of Dior in both photo as text.
When looking into the data gathered one of the main findings of this analysis is that most pictures or text are not tagged. Where for Chanel this was 62%, Louis Vuitton 47% and Dior 25% of the posts that did not include a tag in either the photo or the text. In those cases that the brand wasn’t tagged, the brand was in a minimum of 75% of the cases clearly recognizable. This implies that bloggers especially for Chanel and Louis Vuitton aren’t really emphasizing the brand at all in terms of tagging but just wear or use the product(s) in their daily activities. Within the discussion (chapter 7) the results will be further discussed in relation to the results regarding brand image that are discussed in the upcoming chapter.
35
6. Results
In this chapter the main results will be briefly discussed of the data collected via the survey which all have been analyzed in SPSS. Each discussed result will conclude with whether the hypothesis is supported or not.
6.1 Survey
All data analysis for this survey was done via SPSS. Of the respondents 78.9% was male and 21.1% was female and 95.6% was between the age of 18 and 24. In total 114 respondents have participated in this research. Group one had 34 respondents, group two 38 and group three 42. Brand image can be split up into three sections: types of associations, favorability and uniqueness of associations. According to this division the results will be presented below.
6.1.1 Types of associations
For all questions related to the types of associations (question one to five) a cross table analysis has been performed as discussed in the analytical strategy chapter. For each question and each brand, a Chi-square or, if needed, Fisher-exact test has been performed related to the pre-test and showed no significant relations therefore we can assume that the randomization process was effective, as indicated earlier data is available on request. The result will be briefly discussed per variable per brand below.
I. User imagery: age
Chanel: There is a significant relation between which group the respondents were in and the age
group identified, where F = 36.19, p < .001. Most remarkable is that 71.4% of the respondents in group 3 identified an age group of 19-24 while only 11.8% of the respondents in group 1 gave this answer. Furthermore it is remarkable that 11.8% of the group one respondents identified an age group of 45+ while for both group two as three the percentage was 0.
36
Louis Vuitton: Again there is a significant relation where F = 33.53, p < .001. Most remarkable is that
73.8% of the group three respondents identified an age group of 19-24 while for the group one respondents this was only 20.6%.
Dior: Also for Dior there is a significant relation where F = 44.75, p < .001. Most remarkable is that
73.8% of the respondents in group three identified an age group of 19-24 while for the group one respondents this was only 14.7%. The most frequent identified age group of group one was 36-45 with 44.1%.
This supports hypothesis 1A: ‘’Bloggers have an influence on the non-product related attributes of
luxury brands’’
II. User imagery: style
For both Chanel as Dior a Monte Carlo correction test has been performed in order to perform the Fisher’s exact test due to not enough memory in SPSS.
Chanel: There is a significant relation between which group the respondents were in and the style
identified, where F = 62.84, p < 0.001. Most remarkable is that 57.1% of the group 3 respondents identified the style as being casual while for group one this was only 2.9% and group two 28.9%. Most frequently identified style for group one was Chic.
Louis Vuitton: Again there is a significant relation where F = 30.45, p < .0.002. Most remarkable is
that group one identified chic (23.5%) and edgy (20.6%) as style while for both group three the numbers are correspondently 0% and 9.5%. Group two identified edgy (31.6%) and casual (21.1%) as fitting style to the brand.
Dior: Also for Dior there is a significant relation where F = 85.86, p < .001.Most remarkable was that group two and three both identified casual (23.7%, 47.65%) and alternative (31.6%, 16.7%) as most fitting style to the brand while for both styles 0% of the group one respondents identified these styles. 41.2% of group one identified romantic while this was chosen 0 times among the respondents of group two or three.
37 This supports hypothesis 1A: ‘’Bloggers have an influence on the non-product related attributes’’
III. Usage imagery: situations
Chanel: There is a significant relation between which group the respondents were in and what
situation the respondents identified, where F = 59.68, p < .001. Most remarkable was that 81.6% of group two and 95.2% of group three chose ‘’everyday situation’’ while only 17.6% of the group one respondents identified this situation. Group one identified special occasions (29.4%) and formal work situations (26.5%) as situations they belief the user of the brand wears the product most in while for group three both situations were never chosen (0%).
Louis Vuitton: Again there is a significant relation, where F = 31.44, p < .001. Most remarkable is that
84.2% of group two and 71.4% of group three chose ‘’everyday situation’’ while only 35.3% of the group one respondents identified this situation. Furthermore, group one identified night time (20.6%) and formal work situations (17.6%) while for group three this has only been identified by 0% and 2.4% of the respondents correspondently.
Dior: Also for Dior there is a significant relation between the two variables, where F = 79.98, p < .001.
Most remarkable is that 63.2% of group two and 90.5% of group three chose ‘’everyday situation’’ while only 5.9% of the group one respondents identified this situation. 58.8% of the group one respondents identified special occasions while this situation was chosen by 0% of the group three respondents.
This supports hypothesis 1A: ‘’Bloggers have an influence on the non-product related attributes of
luxury brands’’
IV. Feeling
Chanel: There is a significant relation between which group the respondents were in and the evoked
feeling by the brand identified, where F = 43.83, p < .001. The most remarkable is that 44.1% of the respondents from group one identified a feeling of social approval while this was only identified by
38 7.1% of the group three respondents and 21.1% of the group two respondents. 69.0% of the group three respondents identified a feeling of fun which was only identified by 17.6% of the group one respondents.
Louis Vuitton: Again there is significant relation, where F = 24.17, p < .001. Most remarkable is that
again 44.1% of the group one respondents identified a feeling of social approval while only 10.5% of the group two and 14.3% of the group three respondents identified this feeling. Group two and three mostly identified most fun (21.1%, 31.0%) and excitement (31.6%, 19.0%).
Dior: Also for Dior there is a significant relationship, where F = 39.67, p < .001. Most remarkable is
that 41.2% of the group one respondents identified a feeling of warmth and 17.6% with self-respect. For both warmth as self-respect a maximum of 7.9% of the respondents of both group two and three identified any of these two feelings.
This supports hypothesis 1A: ‘’Bloggers have an influence on the non-product related attributes of
luxury brands’’
V. Benefit
Chanel: There is a significant relation between which group the respondents were in and the need
they believed the brand satisfied, where F = 18.55, p = .001. Most remarkable is 47.6% of the group three respondents identified an experiential need while this was only identified by 23.5% of the group one respondents. 44.8% of the group one respondents identified a symbolic need.
Louis Vuitton: Again there is a significant relation, where F = 10.93, p < .05. Most remarkable is that
63.3% of the group one respondents identified a symbolic need. 33.3% of group three identified with each need.
Dior: There is no significant relation.
