Co-creation and luxury : the effect of co-creation disclosure on perceived luxury and brand attitude

(1)

Master Thesis

Rens Gruppelaar

Co-Creation and Luxury: The Effect of Co-Creation Disclosure on Perceived Luxury and Brand Attitude

Msc in Business Administration - Marketing Track Supervisor: Marcel Weber

Student number: 11139331

(2)

Statement of originality

This document is written by Rens Gruppelaar, 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)

Abstract

This study looks into the effect of co-creation disclosure on brand perceptions and evaluations. It focuses on the luxury landscape, more specifically, the luxury automobile industry. The use of co-creation increases nowadays, providing many advantages for firms and consumers. One implication of this development is the influence of these co-creation activities on consumers’ perceptions of the brand. Multiple studies looked into this effect of co-creation disclosure on the brand perceptions that non co-creating consumers have. In many categories, this effect appeared to be positive. However, the question rose whether this effect would be similar for the luxury industry. One study looked into the luxury fashion industry and found a negative effect. More knowledge was required in this field since it only covered the luxury fashion industry and did not provide considerable insight into the dimensions that form the luxury perceived by consumers. This study therefore aims at contributing to the luxury landscape by investigating the effect of co-creation disclosure on the brand attitude of non co-creating consumers and on their perceived luxury in the brand. The perceived luxury exists of multiple dimensions, which are all looked into in order to get a better understanding of how this concept is affected. Moreover, the relationship between co-creation disclosure and brand attitude has been researched, looking for a possible mediating role of perceived luxury. The results of a survey (N = 250) with an experimental approach do not demonstrate significant effects of co-creation disclosure on both perceived luxury and brand attitude. This absence of an (negative) effect is not in line with earlier research in the luxury industry and does therefore suggest that the effect of disclosure of co-creation might differ per luxury category. For the luxury automobile industry specifically, this means that firms can use and communicate co-creation without harming the perception of the brand.

(4)

Preface

Co-creation has fascinated me for a long time. This special interest started at my previous study, where this approach was introduced to me. I find it interesting that firms are increasingly recognizing the meaningful role consumers can fulfil in the creation and marketing of (new) products. When co-creation became the central topic of my thesis, I wondered how the majority of the market would perceive this. After thorough analysis of the extant literature I found a research gap, which was the starting point of this thesis.

Some people supported me significantly in the process of achieving my goal: finishing the Msc Marketing program. To those, I want to express my gratitude. First of all, special thanks goes out to my supervisor Marcel Weber for his counselling and mentoring throughout the thesis process. I would also like to thank my parents for supporting me, both mentally and financially. Furthermore, I want to show my appreciation to my girlfriend, brother, sisters and friends for their encouragement. I also feel grateful that more than 250 people completed my survey and thereby providing me the chance to complete this study.

(5)

List of Tables

TABLE 1: Scales, deleted items and Cronbach’s Alphas ... 37

List of Figures

FIGURE 1: Conceptual framework with hypotheses and variables ... 27

FIGURE 2: The total, direct and indirect effect of co-creation disclosure on brand attitude, through perceived luxury ... 43

(8)

1. Introduction

Companies are increasingly co-creating with consumers, as it provides many advantages (Van Dijk, Antonides, & Schillewaert, 2014; Lilien Morrison Searls, Sonnack, & Von Hippel, 2002; Fuchs & Schreier, 2011). Fuchs and Schreier (2011) mention pros as better product development and a decrease of costs and risks. Co-creation is especially valuable in New Product Development (NPD) (Hoyer, Chandy, Krafft, & Singh, 2010). According to Hoyer et al. (2010), co-creation can be used in all four phases of the NPD: ideation, product development, commercialization and post-launch. Witell, Kristensson, Gustafsson, and Löfgren (2011, p.154) emphasize the contribution of co-creation in this process: “Activating customers in the co-creation of a good or service can qualitatively enhance the resulting value-in-use”. Research has shown that co-creation in the early stages of the NPD can lead to more innovative ideas (Witell et al., 2011). The firm Procter & Gamble, for example, are therefore marketing more than 42 per cent of their products to the public that originate from ideas outside the firm (Russo-Spena & Mele, 2012). Users are even assumed to have a better ability for idea generation than the employees of a company (Kristensson, Gustafsson, & Archer, 2004). Other examples of brands using co-creation to innovate are Starbucks with ‘MyStarBucksIdea’, BMW’s ‘BMW Group Co-Creation Lab’ and Lego’s ‘Lego Mindstorm NXT’ (Russo-Spena & Mele, 2012).

Firms are thus increasingly using co-creation, which also creates new brand knowledge for the consumers that these co-created products are marketed to. Brand knowledge and brand associations affect brand equity (Keller, 1993), so it is essential for brands to find out what the consequences are of using and communicating co-creation to the public. Research reveals a positive effect on participants of co-creation, resulting in a stronger demand, overvaluation of products and improved brand attitudes (Brodie, Ilic, Juric, &

(9)

Hollebeek, 2013; Fuchs, Prandelli, & Schreier, 2010; Norton, Mochon, & Ariely, 2012). However, effects on non co-creating consumers have not been studied to a high extent, whereas this group outnumbers collaborators of co-creation. A few studies researched this matter. One of these is a study by Fuchs and Schreier (2011), which showed an increase in desirability of co-created products and in the attitude towards the brand. Furthermore, a study by Van Dijk et al. (2014) also implies a positive effect of co-creation knowledge on brand perceptions. Since these studies investigate the effect mostly using low-cost and noncomplex goods, the question arises if such an effect is also applicable in other contexts, like the luxury market. This is especially interesting since luxury marketing is often contradictive to classical marketing (Tynan et al., 2010).

The use of co-creation in the luxury market is increasing, mainly due to the economic recession that caused difficulties in the provision of satisfactory value compared to the high prices (Tynan et al., 2010). However, solely one study was found that has looked into the effect on brand perceptions of luxury brands when communicating co-creation to the public (Fuchs, Prandelli, Schreier, & Dahl, 2013). They found a contrary effect to Van Dijk et al. (2014) and Fuchs and Schreier (2011): the effect of labeling products as user-designed was negative on brand perceptons of non co-creating customers. Therefore, Fuchs et al. (2013) warn luxury brands for communicating or using co-creation. However, more knowledge was required in this field since the study only covered the luxury fashion industry (Fuchs et al., 2013). Moreover, the study did not provide considerable insight into the dimensions that form the perceived luxury of consumers.

This study therefore aims at contributing to this field of knowledge in the luxury landscape by investigating the effect of creation disclosure on the brand attitude of non co-creating consumers and on the luxury that these consumers perceive in the brand. Does this change if (non co-creating) consumers know about the co-creation of a luxury product? The

(10)

product category that was chosen as topic for this study is the luxury automobile industry. It is not only suitable because of its difference from the fashion category, but also provides co-creation practices that are integrated in this study. Examples are the BMW Co-Creation Lab (Russo-Spena & Mele, 2012) and “Your Ideas” by Ford (Williams, Gownder, & Wiramiha, 2010). Since luxury can be considered as a subjective construct (Tynan, McKechnie, & Chhuon, 2010), it might be particularly valuable for these luxury brands to use ideas of consumers. This might increase their ability to establish a luxury brand. To be in line with actual co-creation use and considering the value that it has for luxury marketers, co-creation in the ideation phase constitutes the basis of this study.

In order to get a deeper understanding of how the perceived luxury changes, a closer look was taken at the five dimensions this concept consists of, adopted from the BLI scale of Vigneron and Johnson (2004). Moreover, since a relationship was found between perceived luxury and brand attitude (Stegemann, Denize, & Miller, 2013) a possible mediating role of perceived luxury is researched as well.

Understanding the effect of co-creation information of luxury goods on consumer’s perceived luxury and brand attitudes, can be of great value to companies in this market. Firms might have to either hide or promote co-creation in their communication strategies. This insight is essential for marketers since these customers form the majority of the market (Fuchs & Schreier, 2011).

The effect of co-creation disclosure is measured using a survey with an experimental approach. The fact that this eventual insight is important for luxury brands, and more research is needed, leads to the following research question:

(11)

“How will the disclosure of co-creation of luxury goods affect the perceived luxury and brand attitude of non co-creating consumers and how are the perceived luxury and brand attitude in this case inter-related?"

This report continues with chapter 2 by reviewing the extant literature that is relevant for answering the research question. This chapter explains the central constructs of this study, the conceptual framework that forms the basis for the research that is conducted and concludes with the hypotheses that will be tested. Chapter 3 discusses the research design that is used for this study. It explains the pre-test, survey design and survey execution. Chapter 4 shows the results of the quantitative analyses that have been conducted to test the hypotheses of the study. Finally, chapter 5 provides the discussion and conclusion. This part goes into the explanations and implications of the results, the limitations of this study and the opportunities for further research on this topic. Furthermore it provides a short conclusion of the study.

(12)

2. Literature Review

This chapter provides an overview of the relevant literature for this master thesis. It results from the desk research that has been carried out and leads eventually to the hypotheses that are tested in this study.

2.1 Shift towards co-creation of value

The perspective of today on the role of the consumer in value creation is altering rapidly. The firm used to be the exclusive creator of value and the consumer was solely considered as target for the offerings of the company. Its role only came into play at the marketplace, which brought them together for the exchange of value (Prahalad & Ramaswamy, 2004). It can be concluded that there was barely any or no interaction at all between the firm and the consumer. According to Prahalad and Ramaswamy (2004), this has however changed since the consumer has become knowledgeable, connected, empowered and active. The marketplace is no longer exclusively for the exchange of value; it serves as a forum, which enables and encourages interaction between firm and consumer (Prahalad & Ramaswamy, 2004). One reason for this change is the growth of the Internet, which has significantly contributed to the way of (co-) creating value by enabling more communication between firm and consumer (Füller, Muhlbacher, Matzler, & Jawecki, 2009; Sawhney, Mohanbir, Verona, & Prandelli, 2005). The interaction between the firm and the consumer is now a source of value creation. Innovation with users might even replace innovation exclusively carried out by the firm (Baldwin & Von Hippel, 2011).

The consumer is of increased significance to the firm and the role of the consumer is thus becoming more active. Hoyer et al. (2010) state that co-creation is especially vital in new product development (NPD), giving consumers the chance to play a significant part in this process. Many firms develop new products that are not solely created by the firm but

(13)

co-created with consumers instead (Von Hippel, 2005). The products that are co-co-created are namely more adapted to consumers’ needs, which results in a higher likelihood of commercial success (Lilien et al., 2002). According to Gouillart (2014), the amount of value produced by the network of a firm will become more and more the basis of competition. He states that the forecast of leading academics is that co-creation will develop into a key source for a firm’s competitive advantage. This study focuses on co-creation in the NPD process since it appears to be vital for firm’s performances and since the growth of its use is expected to continue. In this way, the results of the study will be in line with the use of co-creation nowadays and thereby of greater value to firms.

Hoyer et al. (2010) identify four stages in the NPD process, namely ideation, product development, commercialization and post-launch. They state that co-creation can be of value in all these stages of the NPD process (Hoyer et al., 2010). Co-creation in the test and launch phases are fairly usual, but the use of co-creation in the early stages of innovation is growing fast (Bartl et al., 2010). These first phases of the NPD process are essential in attaining success (Cooper, 1993). Witell et al. (2011) illustrate this by showing that a firm’s (financial) performance can be significantly improved when products are developed based on ideas of co-creating consumers. This means that products that are based on consumer’s ideas have a more positive influence compared to products for which traditional market research was used (Witell et al., 2011). These findings are in line with similar results of a study by Lilien et al. (2002), who compared innovations that were developed with the ideas of lead users with innovations developed using traditional approaches. The engagement of consumers in the ideation stage is time- and cost saving plus it decreases the risk of failure (Hoyer et al., 2010). Therefore, many organisations have shifted these phases almost entirely to their consumer communities (Schreier et al., 2012). There are many brands that market the greatest ideas of their consumers (Schreier et al., 2012). Examples are Threadless, LEGO (Schreier et al.,

(14)

2012), BMW (Bartl, Jawecki, & Wiegand, 2010) and Ford (Williams et al., 2010). The increasing use and added value of co-creation in the ideation phase is reason for this study to use it as a basis for research. Moreover, this study concerns the luxury context, a subjective construct (Tynan et al., 2010). Co-creation in the ideation phase is regarded as highly relevant to luxury marketers since it might assist them in establishing and maintaining luxury in the brand.

Concluding, it appears that co-creation is a growing trend in NPD and provides multiple advantages such as an improvement of a firm’s offerings, resulting in increasing profits. However, the question how co-creation affects consumer’s perceptions of these firms is of logical interest as well.

2.2 Effect co-creation disclosure on non co-creators

The internal process and results of co-creation have been the focus of many studies but less research has been carried out concerning the external effects of co-creation. Taking part in co-creation influences collaborators of the co-creation process. Fuchs et al. (2010) demonstrate that customers that participate in the selection of products show a stronger demand for these products and demonstrate increased future loyalty intentions. Furthermore, the so-called ‘IKEA-effect’ implies that consumers (irrationally) overvalue their own creations compared to the ones of experts (Norton et al., 2012). It is also demonstrated that co-creating consumers have an increased positive attitude towards the product that they have helped to create (Brodie et al., 2013). However, limited knowledge is present on the effect creation has on brand perceptions of mainstream consumers that did not take part in the co-creation process (Van Dijk et al., 2014). Fuchs and Schreier (2011) stressed the importance of this group by emphasizing that this segment, in contrast to the co-creating consumers, is the majority of the market.

(15)

Multiple studies found an effect of consumers’ background knowledge on their product and brand evaluations. Stegemann et al. (2013) state that brand associations influence the brand attitude of consumers. Moreover, Keller (1993) mentions that the brand equity of a brand is built from brand knowledge and brand associations. Therefore, it is only logical to assume that brand knowledge of co-creation activities of a brand has its influence on the brand equity and consumers’ perceptions of a firm. The valence of this effect logically determines if a brand should either hide or communicate its co-creation activities to the public.

Extant research has indeed implied changes in the perception of a brand as a result of creation (Van Dijk et al., 2014). Fuchs and Schreier (2011) researched the effect of co-creation knowledge on non co-creating consumers and demonstrated an increase in desirability of co-created products that were either ‘created’ or ‘selected’ by customers. A more recent study by Schreier et al. (2012) revealed that consumers perceive a product as more innovative when knowing that it was user designed. Furthermore, in a study by Van Dijk et al. (2014), it is proven that information concerning co-creation with the consumer positively impacts product perceptions and behavioural intentions. These studies generalized the positive effect of co-creation disclosure to all products. However, they investigate the effect mostly using low-cost and low-complexity goods, like Pickwick tea (Van Dijk et al., 2014), breakfast cereal and T-shirt designs (Schreier et al., 2012). The question rises if such an effect also applies to other product categories like the luxury market, especially since “classical marketing is the surest way to fail in the luxury business” (Bastien & Kapferer, 2009, p. 2).

(16)

2.3 Luxury

2.3.1 Explaining luxury

Luxury is a complex, multi-layered concept, which is defined in many different ways in the existing literature. Cornell (2002, p. 47) explains: “Luxury is particularly slippery to define. A strong element of human involvement, very limited supply and the recognition of value by others are key components” (Cornell, 2002). Vigneron and Johnson (2004) mention a definition made by economists and marketing consultants (McKinsey, 1990, p. 486) that explains luxury brands as “those whose price and quality ratios are the highest of the market; that is, their price is significantly greater than the price of products with similar tangible features”. The importance of the psychological advantages, more than only utilitarian ones, is a major distinction of luxury goods compared to non-luxury goods that is widely agreed upon in current literature (Kim & Johnson, 2015; Nueno & Quelch, 1998; Vigneron & Johnson, 2004). It demonstrates that luxury products are symbolic (Kastanakis & Balabanis, 2012), capable of stressing the status and image of a consumer towards others (Nueno & Quelch, 1998). Vigneron and Johnson (2004, p. 486) state that literature agrees on the fact that the consumption of luxury products “brings esteem on the owner”. Tynan et al. (2010, p. 1157) say that “where the ordinary ends and luxury starts is a matter of degree as judged by consumers”. Vigneron and Johnson (2004) developed a theoretical framework for the luxury that consumers perceive in a brand. This is in line with the view of Tynan et al. (2010) of a subjective luxury concept. Luxury as a subjective construct is taken as starting point for this study. This subjective nature of luxury makes the use of co-creation in the ideation phase particularly interesting for luxury brands. It might be especially convenient to use the ideas of consumers in establishing luxury, since there is no objective standard for the luxury perception. Using ideation, the ability of brands to create this perception of luxury might

(17)

increase since the products can be more aligned with the needs of luxury consumers. Therefore, the elements that ‘make’ luxury might be easier to identify and add to the brand.

2.3.2 Perceived luxury

The perceived luxury of a brand or product differs per person, which was the motive for Vigneron and Johnson (1999) to develop their theoretical framework that identifies the dimensions that determine this perception. Later, they build a scale on this framework in order to measure the perceived luxury and establish a brand luxury index (BLI) (Vigneron & Johnson, 2004). This insight can be of value to marketing managers in the luxury industry by identifying the dimensions that must be sustained or influenced (Vigneron & Johnson, 2004). This is particularly useful because the luxury industry is a very competitive business where brands have to put great effort in differentiating themselves (Kim & Johnson, 2015).

The dimensions that form the perceived luxury of a brand are of a personal and non-personal nature (Vigneron & Johnson, 2004). The non-non-personal perceptions that perceived luxury is composed of are formed by conspicuousness, uniqueness and quality. Conspicuousness, based on the so-called ‘Veblen-effect’ (Lebenstein, 1950), refers to the external reason of purchase to signal wealth to others (Amaldoss & Jain, 2005). Vigneron and Johnson (2004) state that it means that a brand or product should be linked with social status that can be displayed by the consumer. According to them it can play an important role in luxury consumption (Vigneron & Johnson, 2004). Perceived uniqueness, also known as the ‘snob-effect’ (Lebenstein, 1950), relates to a need that is felt by luxury consumers and is formed by the assumption of exclusivity and rarity as drivers of desire for such a luxury brand (Vigneron & Johnson, 2004). Snyder and Fromkin (1977) mention that research shows that people feel the need to be unique instead of similar compared to others. They define it as “a positive striving for abnormality relative to other people” (Snyder & Fromkin, 1977, p. 519). Perceived quality is an essential aspect in the luxury industry. It determines if a brand is

(18)

perceived as superior (Kim & Johnson, 2015). Multiple researchers stress the crucial role that quality plays in the luxury perception (Godey et al., 2012; Quelch, 1987; Vigneron & Johnson, 1999). Vigneron and Johnson (2004) declare that consumers expect luxury brands to provide excellent quality in comparison with non-luxury brands. Prices can be a sign of quality to consumers (Rao & Monroe, 1989).

The personal perceptions of perceived luxury are formed by hedonic and the extended self (Vigneron & Johnson, 2004). Perceived hedonism is about sensory pleasure (Hirschman & Holbrook, 1982) and the ‘quest’ to find this (Hopkinson & Pujari, 1999). According to Hopkinson and Pujari (1999). It is a subjective concept, experienced by the individual. The second personal dimension is ‘perceived extended self’ (Vigneron & Johnson, 2004). This concerns the absorption of the symbolism attached to a luxury good, in the identity of the consumer. Consumers might use luxury (goods) to fulfil their wish to follow a wealthy life (Vigneron & Johnson, 2004) since people believe that their identity is partly formed by their belongings (Belk, 1988).

2.3.3 Democratization of luxury

In one of their articles, Vigneron and Johnson (1999) viewed ‘prestige’ as an important aspect of luxury brands, referring to the uppermost prestigious brands. In their 2004 study however, Vigneron and Johnson (2004) identify ‘prestige’ as the “extreme end of the luxury brand category”. This distinction is the consequence of the development of luxury to a broader industry. Luxury consumers are nowadays namely also present in the mass market. This development is referred to as the ‘democratization of luxury’ (Gardyn, 2002). According to Silverstein and Fiske (2003), these customers are looking for products that positively reflect their (desired) identity and assist them in managing the everyday life stress. These luxury products are called ‘new-luxury’ and can be produced in large numbers, despite the high prices (Silverstein & Fiske, 2003). This is supported by Florin, Callen, Mullen, & Kropp

(19)

(2007), who state that (expensive) luxury products are not exclusively attainable for the wealthy anymore. They state that nowadays, income levels do not keep people from buying luxury products. Consumers rather save on basic products and are willing to pay high prices for goods that are important to them (Florin et al., 2007). Moreover, the attainability of these goods has increased because of technological innovation, globalization, growing incomes and credit that it is easier to obtain nowadays (Florin et al., 2007). For the study to be in line with the current luxury context, the luxury market is not solely considered as ‘traditional luxury’, which exists of “very exclusive brands with the highest price tags” (Truong, Simmons, McColl, & Kitchen, 2008, p. 191), but also includes the new-luxury brands.

2.4 Brand attitude

According to Mitchell and Olson (1981) attitude is, in the marketing context, “an individual’s internal evaluation of an object such as a branded product”. Hence, brand attitude is “a consumer’s overall evaluation of the brand” (Faircloth, Capella, & Alford, 2001, p. 62). In the definition of Spears and Singh (2004, p. 55), “attitude towards the brand is a relatively enduring, unidimensonal summary evaluation of the brand that presumably energizes behaviour”.

Customers’ brand attitudes are central drivers of the brand equity of a company (Liu, Li, Mizerski, & Soh, 2012). Brand attitude is of great importance to organisations, as it relies on the individual’s brand perceptions which are viewed as a reliable predictor of consumer behaviour (Shimp, 2010). Moreover, according to Liu et al. (2012), a positive relationship between brand attitude and brand loyalty exists. Since brand loyalty drives customers to repurchase goods from the same brand, it can increase profits (Oliver, 1999). Attitudes to luxury brands are very significant for customers in their decision-making and are therefore essential for marketing managers in this industry (Stegemann et al., 2013).

(20)

2.5 Co-creation in the luxury market

The worldwide luxury market has grown dramatically over the last twenty years (Cicic, 2009; Truong et al., 2008). However, the economic recession made it a difficult challenge for firms to provide satisfactory value for the high prices (Tynan et al., 2010). This contributed to the use of co-creation by luxury brands as well, co-creating a superior value proposition (Tynan et al., 2010). For a long time, the luxury brands did not “ask consumers what they want; but told them what they should have” (Summers, Belleau, & Xu, 2006). However, since the sales went down, industry experts proposed to alter this attitude so they can meet the growing demands of the luxury consumers (D'Arpizo, Jilla, & Kamel, 2005). Tynan et al. (2010) state that luxury brands should attempt to access consumer ideas and knowledge since this will allow them to stay as innovative as possible.

2.6 Effect of co-creation disclosure in the luxury market

Co-creation is being used increasingly by brands in the luxury industry, but little is known about the effects it has on the brand perceptions of such a brand. Fuchs et al. (2013) researched one part of the luxury brand landscape by looking into the effect of disclosing user design of products in the luxury fashion industry on non co-creators’ demand for these products. In contrast to the positive effects that came forward in studies with a focus on products in the mainstream industries (Fuchs & Schreier, 2011; Schreier et al., 2012; Van Dijk et al., 2014), the results in this specific context demonstrate a negative effect of co-creation disclosure towards non co-creating consumers (Fuchs et al., 2013). This showcases a difference between product categories in effect of a co-creation claim on non co-creators. Fuchs et al. (2013) explains that the reason for this effect lies in the lower perceived quality and the failure of these user-designed luxury fashion items to signal high status to others (Fuchs et al., 2013); two important factors in luxury consumption (Truong et al., 2008; Vigneron and Johnson, 2004). More research is required on whether, by what factors and to

(21)

what degree the effect exists of labeling products as ‘co-created’ (Fuchs & Schreier, 2011; Fuchs et al., 2013; Van Dijk et al., 2014). Fuchs et al. (2013) propose that in the luxury market, this means that future research should be carried out in other categories than fashion. This seems only logical, since only one study has investigated this effect in the luxury industry. This study therefore contributes to this theoretical field. Moreover, the fashion context is only one small part of the luxury market and is very typical when it comes to communication of designers. It is for example very usual to put promotional emphasis on the designer of fashion items (Fuchs et al., 2013).

The two facets (perceived quality and the signaling of high status) researched by Fuchs et al. (2013) do not cover all of the determinants of perceived luxury by the consumer. When aligning these facets with the theoretical framework of Vigneron and Johnson (2004), one can identify perceived quality and conspiousness (i.e. signalling of high status). The other dimensions are however also essential in the luxury framework of Vigneron and Johnson (2004). Thus, the results of Fuchs et al. (2013) do not provide sufficient information to determine the influence on the perceived luxury of customers. This means that, besides the context other than fashion, the measurement of the role of perceived luxury is another gap present in the research concerning the effect of co-creation information on non co-creating customers in the luxury industry. Consumers’ perceived luxury is of great significance to marketeers in the luxury industry (Vigneron & Johnson, 2004).

2.7 Automobile luxury

Since this study focuses on the luxury industry beyond fashion, another category was chosen. One category within the luxury industry using co-creation is the automobile industry. BMW, which can be considered a luxury car brand, developed an online co-creation community in order to innovate with consumers. The BMW Group Co-Creation Lab is an opportunity for consumers to gather online and communicate ideas or thoughts with each

(22)

other and the brand. The platform invites consumers to engage in different stages in NPD in the form of idea contests, concept examinations and other innovation-related activities (Bartl et al., 2010). BMW uses co-creation especially in the ideation phase. This is demonstrated by a manager that has been quoted in an article by Bartl et al. (2010, p. 6): “The generated ideas added innovative and valuable input to the topics we are already working on and confirmed us that the overall direction we are following leads into the right direction.”

2.8 Hypotheses

The hypotheses serve to answer the research question of this study and are based on relevant research. The first two hypotheses concerning the perceived luxury and brand attitude of non co-creating consumers are based on several overlapping relevant findings of previous studies concerning co-creation disclosure.

2.8.1 Hypotheses 1 and 2

Outside the luxury context, many positive effects of co-creation disclosure were found (Fuchs & Schreier, 2011; Schreier et al., 2012; Van Dijk et al., 2014). Thus, these studies suggest that disclosing co-creation is a beneficial marketing approach. However, luxury marketing often requires the opposite of classical marketing (Bastien & Kapferer, 2009; Tynan et al., 2010). A study by Fuchs et al. (2013) is in line with this conclusion, since it implies a negative effect of co-creation disclosure within the luxury industry. Co-creation disclosure negatively affected the perceived quality and the ability to signal wealth to others. The study by Fuchs et al. (2013) was the first to research the effect for luxury brands and generalized its effect to the entire industry by using the results of only one luxury category, namely fashion. These results took however place within the luxury industry, as opposed to earlier studies. Since, as aforementioned, the marketing in the luxury industry is often contradictive to marketing in other industries (Tynan et al., 2010), the study by Fuchs et al. (2013) is considered as highly relevant for the expectations of this study. Furthermore, Fuchs

(23)

et al. (2013) predict a similar effect for other luxury categories than fashion as well. They explicitly mention the automobile industry and predict that the negative effect of their study will also be present for brands as Mercedes-Benz or Jaguar, which is applicable to this study considering the focus on the luxury automobile industry. The effect that was found, was specifically present for high-status luxury goods. Luxury cars can be seen as high-status goods (Sivanathan & Pettit, 2010). Moreover, the findings of Fuchs et al. (2013) apply to the early phases of NPD. An add-on study also suggests that the effect would hold for other co-creation phases or tasks, like selection (Fuchs et al., 2013). All of the above suggests that this study would also produce a negative effect of co-creation disclosure.

Another argumentation arrives at the same conclusion. Schreier et al. (2012) found that the positive effect of co-creation disclosure can overturn to a negative effect when it concerns co-created products with high complexity because the non co-creating consumers perceive these products as too difficult for creation with users. This study concerns co-creation disclosure for a luxury brand that produces cars, which can be considered as a highly complex product. Taking the results of Schreier et al. (2012) into account, the effect for this study should be negative.

Concluding, the arguments of luxury and complexity given above are the basis for the forming of the first two hypotheses concerning perceived luxury and brand attitude. The separate hypotheses are shortly elaborated on below, following the the line of reasoning above.

Forming expectations for the effect of co-creation disclosure on the perceived luxury of non co-creating consumers, the separate dimensions that form this concept are taken into account. Therefore, previous studies were used to determine the possible direction of this effect. In order to do so, especially the negative results of the study by Fuchs et al. (2013)

(24)

were taken into account Since perceived conspicuousness concerns the ability of a brand to show the social status of an owner (Kim & Johnson, 2015), signalling of high status can be seen as similar to the dimension of perceived conspiscuousness. Moreover, since a co-creation process includes multiple consumers, this could reduce the perception of a brand as ‘elite’. This leads to an expectation of a lowered perceived conspicuousness of people to whom co-creation is disclosed.

For the perceived uniqueness a similar expectation was present since the perception of exclusivity and rarity could decrease due to the engagement of customers in the creation process. It is problable that luxury consumers dislike if luxury brands are close to users, since these brands are build upon exclusivity (Kristal, Baumgarth, Behnke, & Henseler, 2016). The consumption of the product or brand could therefore be experienced as less ‘exclusive’ due to the involvement of others. Moreoever, it could be that consumers fear that the brand will be used by consumers that belong to undesired subcultures (Bothwell, 2005).

Another negative effect of the study by Fuchs et al. (2013) consisted of a lowering in the perceived quality. Schreier et al. (2012) furthermore suggest that with communicated co-creation for high product complexity, the perceived innovation ability of a brand reverses and becomes negative. This negative effect comes into existance because non co-creating consumers assume that co-creation is not suitable for such a complex product (Liljedal, 2016). The luxury automobile industry provides highly complex products (i.e. luxury cars) and since perceived innovation ability is closely related to the perceived quality (Schreier et al., 2012), this negative effect could also apply to this dimension.

The extended-self and hedonism dimensions are about personal rewards and not about interpersonal powers (Vigneron & Johnson, 2004). The effects on these dimensions depend on whether consumers like the idea of the co-creation of a product and whether it affects the way they experience the product.

(25)

The effects and expectations of former studies, as well as looking at the dimensions of perceived luxury seperately, lead to the hypothesis for the effect of co-creation disclosure on the perceived luxury of non co-creating consumers. This means:

Hypothesis 1: “Non co-creating consumers perceive less luxury in a brand when co-creation

is disclosed.”

Knowledge structures about a brand include information and associations (Aaker, 1991). These have an influence on the brand attitude (Keller, 1993), which implies that information or associations regarding co-creation can indeed have an effect on brand attitude towards that brand. Moreover, every brand experience can influence consumers’ attitudes (Hupp & Powaga, 2004), which can be the obtainance of information concerning co-creation. Fuchs and Schreier (2011) found a positive effect of co-creation disclosure on the brand attitude of non co-creating consumers. However, this effect considered products that were not luxurious and not complex. This is a significant distinction since this study concerns a luxury brand that produces cars, which is congruent with both luxury and high complextity. The result is a similarly negative expectation for the effect of co-creation disclosure when it comes to the brand attitude of non co-creating consumers. This leads to the following hypothesis:

Hypothesis 2: “The brand attitude of non creating consumers is less positive when

co-creation is disclosed.”

2.8.2 Hypothesis 3

Lutz (1993) mentions the effect of perceptions on attitudes. Within the luxury context, Vigneron and Johnson (2004) suggested a relationship between the perceived luxury and the formation of attitudes. Stegemann et al. (2013) indeed demonstrated that consumers’ perceived luxury of a brand has a positive influence on the attitude of consumers towards the

(26)

luxury brand. However, they stress the need for further analysis of this relationship (Stegemann et al., 2013). The findings of the studies mentioned above are taken as basis for the formation of an expectation in this context. This relationship is therefore researched by testing the following hypothesis:

Hypothesis 3: “Perceived luxury has a positive influence on the brand attitude of non

co-creating consumers.”

Fuchs et al. (2013) found that perceived quality and agentic feelings (similar to conspicuousness), mediate the effect of co-creation disclosure on consumer demand. These two mediating dimensions are also present in the perceived luxury scale and consumer demand is related to brand attittude. The question arises if perceived luxury can mediate the effect on consumers’ brand attitude in the context of this study. Since a positive effect has been found of perceived luxury on brand attitude (Stegemann et al., 2013), this study will therefore look into the possible mediating role of perceived luxury on the relationship between co-creation disclosure and brand attitude. Formally:

Hypothesis 4: Perceived luxury mediates the relationship between co-creation disclosure and

brand attitude.

2.9 Conceptual framework

The hypotheses explained above, lead to the conceptual framework of this study. This framework, displayed in Figure 1, shows the variables and hypotheses that are investigated in this study.

(27)

Figure 1. Conceptual framework with hypotheses and variables

Summarizing, the focus of the study is how co-creation disclosure affects perceived luxury and brand attitude, which are the dependent variables. The independent variable here is the co-creation disclosure (co-creation vs. no co-creation), which will be the subject of manipulation in the survey. The relation between perceived luxury and brand attitude is also looked into, since it is implied that perceived luxury has an effect on brand attitude (Stegemann et al., 2013). Moreover, co-creation has an effect on multiple dimensions of this construct, which could point to a possible mediating effect of perceived luxury in this case.

The next chapter explains how the research of this study is conducted, clarifying the research design of this study.

(28)

3. Research Design

This chapter explains the research design that is used for this study. Firstly, the pre-test is described. Secondly, the survey design is clarified which goes into the measurement of the variables. Finally, the survey execution is explained in detail.

3.1 Pre-Test

In order to obtain a neutral and objective picture about what luxury brand to select for the survey, a pre-test was conducted (N=20). The goal of the pre-test was to select a brand that is seen as both luxurious and familiar. These aspects were also measured in pre-tests of Vigneron and Johnson (2004) and Kim and Johnson (2015). The respondents were asked about four different brands, all from the automobile category. Some of these brands, namely Mercedes-Benz and BMW, were included since they came forward as (new-) luxury brands in relevant literature (Florin et al., 2007; Fuchs et al., 2013; Russo-Spena & Mele, 2012; Vigneron & Johnson, 2004). The other brands that were included in the pre-test were Audi and Lexus.

The pre-test was conducted among 20 respondents of which 55% was female. The age range was between 21-52 years (M = 26.20, SD = 6.73). The participants were asked on a 7-point Likert scale to indicate if they considered the brand as luxury and if they were familiar with the brand. The pre-test demonstrated that Mercedes-Benz was considered as most luxurious (M = 6.1, SD = .97), an important requirement for this study. It was also considered as a familiar brand (M = 6.25, SD = .97). Therefore, this brand was selected as luxury brand for the survey.

(29)

3.2 Survey design

The hypotheses of this study were tested using a survey with an experimental approach. This was opted for since, according to Saunders and Lewis (2012), an experiment shows “whether a change in one independent variable produces a change in another dependent variable”. This matches with the focus of this study, which is the effect of co-creation disclosure (independent variable) on perceived luxury and brand attitude (dependent variables). It also enables the researcher to analyze the data in search for a mediating effect of perceived luxury by looking into the relation between perceived luxury and brand attitude.

In the survey, a between-subjects design was used. This means that people belonged to one of the groups and could not participate in multiple groups. The participants were randomly assigned to one of the groups, which prevented selection bias. The experimental approach includes that only the independent variable is to be manipulated, keeping the other variables, apart from the dependent variables, constant (Saunders & Lewis, 2012). In this way, respondents of each group were exposed to the same information, except the manipulation. This means that the addition of the disclosure of co-creation can be seen as the explanation for changes in the dependent variables.

The participants received questions measuring the dependent variables of this study. All the questions that were present in the survey can be found in Appendix A. The scales are explained below.

3.2.1 Perceived luxury

Participants had to answer questions that measured the five dimensions of perceived luxury, as determined by Vigneron and Johnson (2004). The scale they developed is closely related to their framework, which consists of personal and personal dimensions. The non-personal dimensions are conspicuousness, uniqueness and quality, whereas the non-personal dimensions are hedonic and the extended self. The dimensions were measured on a 7-point

(30)

semantic differential scale, which is considered the best method to measure abstract concepts for brands (Mindak, 1961). The items that were used consisted of word pairs of bipolar adjectives, based on the article by Vigneron and Johnson (2004). The word pairs contained one word indicating ‘extreme luxurious’ and the other indicating ‘fairly luxurious’ (Kim & Johnson, 2015). One critique on their scale was the fact that it was tested on Australian business students, who might be more acquainted with the concepts and words used in the scale (Christodoulides et al., 2009; Kim & Johnson, 2015). It cannot be expected that the convenience sample of consumers that are researched in this thesis will be able to respond as precise as the Australian business students. This increased the likelihood that people would not understand some of the concepts and that the participants would not notice the bipolarity of the two words that belonged to a pair of adjectives as designed by Vigneron and Johnson (2004). Therefore, the scale was slightly adapted in order to measure the dimensions correctly. For this adaption of the scale, the article by Kim and Johnson (2015) was used. This article made use of both qualitative and quantitative approaches to revise the BLI scale of Vigneron and Johnson (2004). The eventual revision for this study of the scale’s word pairs was focussed on making the words more distinguishable, increasing clarity and emphasising the bipolarity in the word pairs. This was especially carried out by integrating words as ‘fairly’ and ‘very’, an approach that was already used in some items of the original scale. One item was deleted, namely ‘luxurious vs. upmarket’, which Kim and Johnson (2015) identify as unnecessary since measuring luxury is already the purpose of the scale. Moreover, two items were clarified by providing the participants a short explanation derived from the Vigneron and Johnson (2004) study, as well as the Kim and Johnson (2015) article.

Examples of word pairs that were used are the ones that measured perceived hedonism. The word pairs used here were ‘exquisite’ versus ‘tasteful’; ‘extremely attractive’

(31)

versus ‘attractive’; and ‘very stunning’ versus ‘fairly stunning’. Some of the items were reverse-scored in order to increase the validity of the measured items.

3.2.2 Brand attitude

The attitude towards the brand has been measured using the scale developed by Spears and Singh (2004). This scale is in line with the conceptualisation of brand attitude (“an individual’s internal evaluation of the brand”) and is generalizable to different kind of products and services (Spears & Singh, 2004). Therefore, it is assumed that the scale can also be used for the luxury market in the context of this study.

The scale measured the brand attitude towards Mercedes-Benz with five items, asking for the personal evaluation of the participant. All these items are measured on a 7-point semantic differential scale. One example of a wordpair that was used is ‘unfavorable’ versus

‘favorable’.

3.3 Survey Execution

The survey execution is explained in detail by providing explanations concerning the sample, data collection, procedure, stimuli and manipulation check.

3.3.1 Sample

There were originally 332 online responses to the survey. Incompletion was the reason for taking out 77 of the respondents. Since non co-creators are the focus of this study, five more responses were deleted because these people indicated to have co-created with Mercedes-Benz. This resulted in an eventual number of 250 respondents that were included in the sample for this study. Of this group, 56% was female. The participants’ ages ranged from 17-76 (M = 27.43, SD = 7.78). The sample exists of people from different countries, with most participants coming from Holland (83%). The treatment groups, randomly assigned to

(32)

the participants, were evenly divided. The co-creation group contains 124 completed responses, whereas the non co-creation group exists of 126 respondents.

3.3.2 Data collection

The participants of the survey were recruited through a convenience sample. The chosen language of the survey was English, which means that the survey was accessible to multiple nationalities. The survey was online and active between the 14th and 28th of May 2016. The researcher contacted potential respondents by social media where more than 800 people could be reached. Moreover, emails were sent to other contacts of the researcher. The survey took approximately 5-7 minutes to complete. Respondents were encouraged to take the survey by providing an incentive in the form of a chance to win a cinema voucher. Furthermore, the researcher informed and ensured the participants about the confidentiality of their (personal) data. In order to convince participants of this approach, the personal questions were limited to topics as gender, age and country of origin. There was only an optional question, which allowed the participants to leave their email address. It was explained that this email address would be exclusively used to contact them in case of winning the voucher mentioned earlier.

3.3.3 Procedure

The survey started with a short note of gratitude for taking the survey and a promise of confidentiality. The goal of the survey was explained as gaining the views of participants on “a well-known” brand. This contributes to the avoidance of possible biases that can occur if participants realize that co-creation is the central unit of analysis in the survey. Participants were then shown a short explanation of the procedure, followed by the text about Mercedes-Benz together with their logo. The texts the participants saw depended upon the group they were in. The assignment of the participants to one group or the other was random.

(33)

The texts that the participants received were equal to a certain extent: for one group, co-creation information was added. For the first group, this means the general description about the brand and the short announcement of a new car with new features. For the second group this means the same text with a disclosure of co-creation activities of the brand. After this text, the perceived luxury of the brand was measured, as well as the consumers’ brand attitude. Next, participants were asked to indicate if they ever co-created with Mercedes-Benz, since the study concerns non co-creating consumers. Moreover, a manipulation check was integrated.

The last question provided the option to leave an email address in order to get a chance of winning the cinema voucher. This was followed by a short explanation of the actual goal of the study and the fictional nature of the text concerning Mercedes-Benz.

3.3.4 Stimuli

The stimuli in the experiment were descriptions that contained a short text concerning a luxury brand and the announcement of a new product. As mentioned earlier, Mercedes-Benz was selected as luxury brand for this study after conducting a pre-test (N = 20).

The text, accompanied by the brand logo, was identical for both groups to a certain extent: for one group co-creation activities involving the new product were added. The text for the group with no co-creation disclosure was formulated as follows: “Mercedes-Benz is

one of the most well-known automobile manufacturers in the world. This established global brand focuses on delivering their promise to provide ‘the best or nothing’. Next year the organization will introduce a new car, for which Mercedes-Benz developed new features, which will be announced soon.” In the description for the group exposed to the manipulation,

information about the use of co-creation was added: “For this car, Mercedes-Benz

collaborated with consumers to develop new features, which will be announced soon. Through a virtual platform, Mercedes-Benz invited people from all over the world to

(34)

contribute their suggestions and ideas. Moreover, people were asked to evaluate concepts by voting for, and commenting on them. Mercedes-Benz used this consumers' input in developing the car.” The texts as presented to the participants can be found in Appendix A.

The texts of the stimuli were developed using information from the worldwide web (e.g. Williams et al., 2010), as well as using articles regarding the use of co-creation in the automobile industry (Bartl et al., 2010; Russo-Spena & Mele, 2012). The general information concerning Mercedes-Benz was derived from different websites, whereas the co-creation activities were based upon the actual use of co-creation in the automobile industry. The cases that were used here originate from BMW (“About the BMW Group Co-Creation Lab”) and Ford (Williams et al., 2010). According to Russo-Spena and Mele (2012), there are five different forms or phases of co-creation, of which BMW particularly uses the ideation- and evaluation phase. Reflecting the reality of the use of co-creation, the stimuli embody the ideation phase. In this way, an authentic text was created which originated from real-life co-creation. Since these specific activities were adapted from other brands, and thus fictional for Mercedes-Benz, the participants in the no disclosure condition could not possess the same co-creation related information as the other group.

3.3.4 Manipulation check

A manipulation check was integrated in an attempt to measure the effectiveness of the manipulation. Unfortunately, the conclusion was drawn from the results of the manipulation check that it was not properly executed. The answers that were given did often not correspond with the assigned group of the respondent. However, it was assumed that the manipulation check was not representative and that the manipulation was effective. The text disclosing the co-creation activities of Mercedes-Benz were clearly explained and it is assumed that people absorbed this information when answering the questions. This conclusion is supported by the comparison of means between the groups, which shows a clear distinction in a consistent

(35)

direction. This means that respondents were not deleted, since it would unnecessarily reduce the total number of respondents of the study and thereby reducing the reliability of the results.

(36)

4. Results

This chapter displays the results of the survey that were analysed using IBM SPSS. Firstly, the data preparation and preliminary analysis are explained. Secondly, the testing of the hypotheses is presented.

4.1 Preliminary analysis and data preparation

Before the testing of the hypotheses, a reliability analysis was conducted. Moreover, the normality and the homogeneity of the data were analysed. The results of these steps are explained below.

4.1.1 Reliability analysis

A reliability analysis was done for all variables in order to detect the consistency of the measures (Field, 2009). This demonstrated that for three variables, items had to be deleted in order to improve the reliability of the scale. This information was used to compute the items into the variables with the most ideal possible reliability. For conspicuousness, this means the removal of one item, which resulted in a Cronbach’s α = .44. This is considered a low Cronbach’s α. For uniqueness, two items were deleted which lead to an improved Cronbach’s α = .74. This is considered as sufficient according to the recommended minimum of .70 (Field, 2009). Also for the variable hedonism one items had to be deleted, demonstrating a Cronbach’s α = .62. This is not very high, but can be considered sufficient when looking at the bare minimum of Cronbach’s α = >.50 (Field, 2013). The variables quality, extended-self and brand attitude showed immediate sufficient Cronbach’s Alphas >.70 which could not be improved significantly by deleting any items. Table 1 provides an overview of the number of items that were deleted and the eventual Cronbach’s α of the variables.

(37)

Table 1. Scales, deleted items and Cronbach's Alpha

Scale Items deleted Cronbach’s α

Conspicuousness 1 .44

Uniqueness 2 .74

Quality Non .72

Hedonism 1 .62

Extended-Self Non .74

Brand Attitude Non .92

The variable conspicuousness showed a Cronbach’s α that was insufficient. However, the variable is an essential part of the total perceived luxury. Since this is a central dimension in this study, it was decided to retain conspicuousness. Also hedonism demonstrated a low Cronbach’s α, but was above the bare minimum (Field, 2013). Therefore, this variable is not eliminated from the analyses that follow.

4.1.2 Normality

The data were checked for normality. The histograms of the variables demonstrated bell-formed shapes, which points at normal distribution among the variables. The skewness of the data were checked and showed to be within the limits considered acceptable, namely between -1.5 and +1.5 (Tabachnick & Fidell, 2013). This does not point towards a non-normal distribution among the data. Moreover, the central limit theorem states that for any dataset with a sample >30, it can be assumed that the data is normally distributed (Field, 2003). Since the sample of this study exists of 250 participants, this assumption is indeed made.

(38)

4.1.3 Homogeneity of variance

In order tot test the homogeneity of variance the Levene’s test was carried out, which tests if the variances among the groups (co-creation vs. non co-creation) are equal. Data is considered homogeneous when the Levene’s test is not significant (p >.05). This means that the variances are generally similar. Only for ‘conspicuousness’, the variances differed for the non co-creation group and the co-creation group, F(1,98) = 4.67, p < .05. For the other variables, the variances appeared to be equal for both groups. For the total luxury that participants perceived in the brand, including all dimensions of the BLI scale (Vigneron & Johnson, 2004), the variances were equal for the non co-creation group and the co-creation group, F(1,248) = 3.19. For perceived conspicuousness, the variances were equal for the non co-creation group and the co-creation group, F(1,98) = 4.67. For perceived uniqueness the variances are equal for both groups also, F(1,248) = .50. Perceived quality demonstrated equal variances for the groups, F(1,248) = .34. Also perceived hedonism indicated equal variances for the two groups, F(1,248) = 3.47. For perceived extended self the variances for the groups were equal as well, F(1,248) = .78. Also the last variable, brand attitude, showed equal variances for the groups, F(1,126) = .11.

4.2 Hypotheses testing

As mentioned above, the normality and homogeneity were checked. These are both assumptions of the independent samples t-tests that were conducted to test the first hypothesis. The variables that measure perceived luxury are reversed in order to make the amount of luxury that is perceived going upward from left to right. In this way, the scale ranges from ‘fairly luxurious’ (left) to ‘extremely luxurious’ (right). After testing the first hypothesis, concerning the perceived luxury, a closer look has been taken at the dimensions of this concept by analysing them separately. The second hypothesis was tested using an independent samples t-test as well. A Pearson correlation test was used to test the third

(39)

analysis. The testing of the fourth hypothesis was carried out with the PROCESS Macro tool, developed by Hayes (2012).

The first analysis tests if non co-creating consumers perceive less luxury in a brand when co-creation is disclosed. This has been carried out using the independent samples t-test. As mentioned above, the normality and homogeneity were checked, which are both assumptions of this test (Field, 2013). The independent samples t-test compares the means of the groups for the total perceived luxury, computed of all the dimensions present in the framework of Vigneron and Johnson (2004). This means that the dimensions of conspicuousness, uniqueness, quality, hedonism and extended-self are all included. Thus, the amount of luxury respondents perceive in the brand in either the non creation or creation condition. On average, participants perceived more luxury in the brand when co-creation was disclosed (M = 4.30, SE = .07) than when co-co-creation was not disclosed (M = 4.24, SE = .07). This difference was not significant t(248) = -.62, p >.05. This result means that non co-creating consumers do not perceive less luxury in a brand when co-creation is disclosed. There is even an, although not significant, slightly positive difference in the means when co-creation is disclosed. Therefore, it can be concluded that the hypothesis is rejected:

Hypothesis 1: “Non co-creating consumers perceive less luxury in a brand when co-creation

is disclosed.” = Rejected

Since the research question of this study focuses on how the effect of co-creation disclosure works, all dimensions of perceived luxury were analyzed seperately. This provides an insight in if, and how these perceptions are influenced by co-creation disclosure. The results of the analyses, generated with the analyses of independent samples t-tests, can be found below.

(40)

Conspicuousness

The first dimension is perceived conspicuousness. Testing for the effect of co-creation disclosure on this perceived conspicuousness dimension, the means of the groups are compared. On average, participants perceived more conspicuousness in the brand when co-creation was disclosed (M = 5.04, SE = .08) than when co-co-creation was not disclosed (M = 5.01, SE = .09). This difference was however not significant t(239.75) = .436, p >.05. Despite the slightly higher mean for co-creation disclosure, it can be concluded that perceived conspicuousness is not significantly affected by co-creation disclosure.

Uniqueness

The analysis of perceived uniqueness demonstrates the amount of uniqueness participants perceive in the brand in either the non co-creation or co-creation condition. On average, participants perceived slightly more uniqueness when co-creation was disclosed (M = 3.68, SE = .11) than when co-creation was not disclosed (M = 3.66, SE = .10). However, this difference was not significant t(248) = -.16, p >.05. Therefore, the conclusion is that also perceived uniqueness is not significantly affected by co-creation disclosure.

Quality

The statistical analysis of perceived quality has also been carried out, by comparing the means of the two groups (co-creation vs. no co-creation). On average, participants perceived more quality in the creation condition (M = 4.16, SE = .09) than in the non co-creation condition (M = 4.13, SE = .09). However, this difference was not significant t(248) = -.22, p >.05. This means that the perceived quality is not significantly affected by co-creation disclosure.

(41)

Hedonism

The independent samples t-test demonstrates the amount of hedonism participants perceive in the brand in both conditions. On average, participants perceived more hedonism when co-creation was disclosed (M = 4.11, SE = .10) than when co-creation was not disclosed (M = 3.89, SE = .11). This difference was not significant t(248) = -.1.49, p >.05. Therefore, it can be concluded that co-creation disclosure does not significantly affect the hedonism perceived in the brand.

Extended-Self

The last dimension that has been analysed is perceived extended-self. On average, the perceived extended-self was higher with co-creation disclosure (M = 4.51, SE = .09) than without co-creation disclosure (M = 4.44, SE = .09). This difference was however not significant t(248) = -.58, p >.05. Therefore, also the dimension of extended-self is not (significantly) affected by co-creation disclosure.

The second hypothesis contains the expectation that the brand attitude of non co-creating consumers is less positive when co-creation is disclosed. This has been tested using the independent samples t-test. This test compares the means of the groups for brand attitude, thus the attitude towards the brand in both the non co-creation and co-creation condition. On average, participants showed a better brand attitude when co-creation was disclosed (M = 5.12, SE = .09) than when co-creation was not disclosed (M = 4.98, SE = .11). This difference was however not significant t(248) = -1.5, p >.05. It can be concluded that the brand attitude of non co-creating consumers is not significantly less positive when co-creation is disclosed. Therefore, the hypothesis is rejected.