Thesis (Marketing Track)
Measuring the extent to which brand personalities affect customer-based brand equity.
University: University of Amsterdam Product: Thesis Supervisor: A. C. J. Meulemans Track: Marketing Name: Martijn MarinusStatement of originality
This document is written by Student Martijn Marinus 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.
Acknowledgement
As a final step of obtaining my master degree in Business Administration at the University of Amsterdam I have written my thesis into the area of brand personalities and brand equity. This is the first time writing a thesis of this nature, which I have experienced as a challenging but foremost an enlarging task. The continuing search for predicting consumer behaviour and responses has always been one of my interests. It was not long after learning about the concepts of customer-based brand equity and brand personality during my master courses till I knew what my thesis topic was going to be. And now, after encountering many articles and interesting research fields it got my attention even more. The enormous amount of interesting topics made me realize how short five months actually is for a full study. Due to time restraints I had to limit the extent to which I was going to research and test the concepts. This resulted in less attention for the mediating variables as I would have wanted to. In my opinion, this mediation could have been a thesis topic apart.I would like to take this moment to thank some people. Firstly, I would like to thank my supervisor Toon Meulemans for all his support, advice and guidance. Secondly, I would like to thank my family, friends and in specific MSc. J.A.W. Veldman for additional guidance and feedback in the process of creating this thesis. Last, above all, I would like to thank the Master himself, the author of all knowledge and wisdom, for His countless love. I sincerely hope you enjoy reading this thesis, as much as I did writing it. Kind regards, Martijn Marinus 24th of June 2016 University of Amsterdam, The Netherlands
Abstract
In the competition of brands, firms search for unique, key values to function as building blocks for brand identities. Customers interpret these brand identities and translate these into a brand image, which subsequently influences the customer preferences (Keller, 1993; 2003). One dimension of the brand image is that of brand personality, defined as ‘the set of human characteristics associated with a brand’ (Aaker, 1997). Aaker (1997) presented five brand personalities, namely: Sincerity, Excitement, Competence, Sophistication, and Ruggedness. While the general effects of brand personality have been measured, this has not yet been done for the brand personalities individually. Based on this the research question for this research has been formulated: “To what extent does customer-based brand equity (CBBE) differ across brand personalities?”. We expect different shopping motivations to have an effect on the level of CBBE. Therefore, is the mentioned relation tested for mediation of utilitarian- and hedonic shopping motivations.
The findings of this research can help managers tuning brands’ strategies with the desired audience using the concept of brand personality. Improved insights in this concept provides new ways for differentiation and helps predict consumer responses. Thereby, the knowledge on the mechanisms behind utilitarian and hedonic shopping motivations can help managers build an environment where marketing creates more value without necessarily adjusting the marketing strategy itself.
The results of this research indicate that several brand personalities exceed in one or more equity dimensions. The sophisticated personality is perceived as the highest quality. Even though the Cronbach’s alpha for brand uniqueness didn’t exceed the minimum value for reliability, were the tests still executed. The results indicate that the sophisticated personality is perceived as the most unique, with sincerity being the least unique. Competent brands are significantly more often considered for purchases than other brands. Thereby, the results seem to indicate that a utilitarian- and hedonic mind-set is to a certain extent malleable resulting in more effective marketing, however further research will be required to support this indication.
Table of contents
ACKNOWLEDGEMENT ... 3 ABSTRACT ... 4 CHAPTER 1 INTRODUCTION ... 8 CHAPTER 2 THEORY ... 10 2.1 BRAND IMAGE ... 10 2.2 BRAND PERSONALITY ... 12 2.3 BRAND EQUITY ... 14 2.4 CUSTOMER-BASED BRAND EQUITY (CBBE) ... 14 2.5 CONCEPTUAL DOMAIN OF CUSTOMER-BASED BRAND EQUITY ... 16 2.6 HEDONIC AND UTILITARIAN PERCEPTIONS ... 18 2.7 CONCEPTUAL FRAMEWORK ... 20 2.8 LITERATURE GAPS AND RESEARCH QUESTION ... 20 2.9 THEORETICAL AND MANAGERIAL CONTRIBUTION ... 21 2.10 HYPOTHESES ... 22 2.10.1. Expected differences and biggest impacts ... 23 CHAPTER 3 RESEARCH DESIGN AND METHOD ... 26 3.1 SAMPLE ... 26 3.2 RESEARCH DESIGN ... 27 3.3 PRE-TEST: BRAND GENERATION ... 28 3.3.1. Method ... 28 3.3.2. Results ... 28 3.4 SURVEY AND MEASURES ... 29 3.5 PROCEDURE DATA COLLECTION ... 29 3.6 MAIN STUDY ... 30 CHAPTER 4 RESULTS AND ANALYSIS ... 31 4.1 PRELIMINARY ANALYSIS ... 31 4.2 RELIABILITY AND SCALE MEANS ... 31 4.3 ANALYSIS ... 32 4.3.1 Covariance variables ... 355.2 LIMITATIONS ... 40 5.3 CONTRIBUTION ... 40 CHAPTER 6 CONCLUSION ... 41 APPENDIX ... 43 LITERATURE ... 43 ATTACHMENTS ... 47 ATTACHMENT 1: PRE TEST ... 47 ATTACHMENT 2: SURVEY MAIN RESEARCH ... 48 2.1 Normal condition ... 48 2.2 Utilitarian condition ... 49 2.3 Hedonic condition ... 50 2.4 Demographic questions ... 51 ATTACHMENT 3: CRONBACH’S ALPHA ... 52
Table of figures and tables
Figures
FIGURE 1. BRAND PERSONALITY SCALE ... 13 FIGURE 2. CUSTOMER-BASED BRAND EQUITY PYRAMID ... 16 FIGURE 3. CONCEPTUAL FRAMEWORK ... 20 FIGURE 4. CONCEPTUAL RELATIONS BRAND PERSONALITY & CBBE ... 22Tables
TABLE 1. HYPOTHESES ... 25 TABLE 2. PRE-TEST RESULTS: BRANDS, TRAITS & PERCENTAGES ... 28 TABLE 3. VARIABLES, DIMENSIONS, MEANS AND STD. DEVIATION: NORMAL, UTILITARIAN AND HEDONIC CONDITION ... 34 TABLE 4. OVERALL MEAN: NORMAL, HEDONIC AND UTILITARIAN CONDITION ... 38 TABLE 5. SUPPORT HYPOTHESES... 39Chapter 1 Introduction
We interact daily with brands requiring our attention and consideration. Increasing competiveness and daily introductions of new brands require firms to search for new strategies to strengthen their position. (Lee, Lee, Wu, 2011). Key values are chosen as building blocks for brand identities, these are transformed into one or more components like logo, price and design. The difference between brand identity and brand image is, that the brand identity describes how a company sees itself whereas the brand image describes the customers’ thoughts and feelings towards the brand (Roy and Banerjee, 2007). Customers interpret the firm's identity and translate this into a brand image (Keller, 2003), which influences the customer preferences. (Keller, 1993; Martínez & De Chernatony, 2004).
It can be said that a firm's most valuable asset, for improving marketing productivity, is the knowledge that has been created on the brand, in customers' minds from the firm's investment, in previous marketing programs (Keller, 1993). This brings us to the widely studied concept of brand equity. Keller (1993) defined brand equity as unique marketing effects resulting from the brand name. For example, the results of marketing a product or service because of its brand name that would not occur if the same product or service did not have that brand name. In this same article, Keller proposes different ways to measure these marketing effects according to one or several equity dimensions e.g. loyalty, brand awareness and perceived quality. When a firm is seeking to build strong brand equity, it is essential for this firm to understand the core dimensions of the brand image and what causes these dimensions to increase in value (Lee & Oh, 2006). A way of doing this is according to the construct of brand personality.
Aaker (1997) defined brand personality as ‘the set of human characteristics associated with a brand’. A considerable amount of attention in research has focused on how the brands’ personality enables a customer to express his or her own self (Belk, 1988), an ideal-self (Malhotra, 1988), or specific dimensions of the self (Kleine & Keman, 1993) through the use of a brand (Aaker, 1997). Both scholars and researchers acknowledge the role that brand personalities can play for differentiating a brand in e.g. its product category (Plummer, 1984), enhancing the customers’ preference and loyalty towards the brand (Fournier, 1998), and the creation of brand equity (Keller, 1993). Brand personality also influences brand recognition
and brand beliefs such as: perceived quality (Ramaseshan & Tsao, 2007), and brand associations (Freling & Forbes, 2005).
Widely accepted research on brand equity has focused on developing measurements tools (e.g., Keller, 2003; Park and Srinivasan, 1994; Yoo & Donthu, 2001), but only little empirical research has focused on understanding the effects and the process of brand equity formation through e.g. the concept of brand image. Although brand image has been considered as the driving force of brand asset and brand performance, few studies have elaborated on the relationship between brand image and brand equity (Zhang, 2014).
This research focusses on a brand personality as a dimension of brand image. The purpose of this research is not to measure the effect of brand personality on brand equity as a whole, but to measure the level to which brand personalities outcome differ across one and another.
A concept that may mediate the effects of brand equity on shopping behaviour is that of shopping motivations. One explanation of shopping motivations, is that consumers are driven by hedonic- and utilitarian motivations (Park-Poaps, 2010). This concept perceives strong brand associations and perceived quality as hedonic drivers, whereas task-oriented and functional shopping behaviour are considered utilitarian (Park-Poaps, 2010). It this context, it may appear that brand equity effects become stronger when driven by hedonic motivations. Then again, utilitarian motivations may decrease the effects of brand equity when the purchase is a product efficient and rationally highlighted process.
This research is structured as following. The next chapter, chapter two, describes the current state of literature in regarding topics such as brand personality and brand equity. This chapter ends with the hypotheses and theoretical and managerial contributions. Chapter three continues with the methods and tools used to gather data and test these hypotheses, followed by the results which can be found in chapter four. Chapter five discusses the results, and mentions the contributions and limitations of this study. The final chapter summarizes the research question, approach, arguments, and highlights, once again, the contribution of this research.
Chapter 2 Theory
2.1 Brand image
Every day we interact with brands requiring our attention and consideration. Increasing competiveness and daily introductions of new brands require firms to search for new strategies to strengthen their position (Lee et al., 2011). One of the most valuable assets for any firm is the intangible asset represented by its brands (Keller & Lehman, 2004). A key component of this intangible asset can be the associations consumers have with a brand, for example: young, refreshing and caffeine for a cola brand. These associations can be both concrete in a functional way, as well as abstract related. This cognitive representation one has with a brand is often referred to as the ‘brand image’. To prevent confusion with the concept of brand identity, this research describes brand identity as: how a company sees itself, whereas the brand image describes the consumers’ thoughts and feelings towards the brand (Roy & Banerjee, 2007). Numbers of studies have been conducted on the topic of brand image. Gardner and Levy (1955) describe, in their classical article, that the overall character or personality may even be more important than mere technical facts about a product. Later researchers (i.e. Herzog, 1963) regarded brand image not as the individual traits of a product rather as the sum of impressions consumers receive from different sources, all of which combine to form a brand personality.
The construction and maintenance of brand image is a prerequisite to the brand management of a firm (Park, Jaworski, & Maclnnis, 1986). Up to now, scholars have not come to an agreement on the definition of brand image. This definition varies mainly between four perspectives: (1) blanket definitions, (2) meanings and messages, (3) personification and (4) cognitive or psychological elements. For example: (1) “Customers’ general perception and impression of a brand” (Herzog, 1963). (2) “The symbolic meaning embedded in the product or service” (Noth, 1988). (3) “Brand image reflects customers’ characteristics, and they purchase the brand to express themselves” (Martineau, 1957). (4) “Brand image is consisted of the customers’ opinion, attitude and emotion toward a brand, which reflects the cognitive or psychological elements of the brand” (Gardner & Levy, 1955).
According to Nandan (2004) the positioning of brands in the market, customers’ preferences are often based on brand image rather than the product. Whereas brand identity
originates from the company’s perspective, the brand image relates to the consumer’s perception of the brand. This presents marketers with the challenge to find consistency between the individual's belief about him- or herself and the brand image the consumers have of the brand (Graeff, 1997). The congruence between the brand image and customers’ self-image would enhance customer satisfaction and customers’ preference for the brand (Jamal & Goode, 2001). A brand image is formed by consumers based on the associations they remember in respect to that brand. This view on brand image consisting of a set of associations is in a sense similar to the definitions used by Keller (1993), Aaker (1991; 1992) and Park & Srinivasan (1994). These will be later described within the construct of brand equity. As previously mentioned, the sum of the brand impressions experienced by consumers be can combined to form a brand personality. Throughout the years considerable attention has been given to this construct. The most frequently used definition of brand personality is that of Aaker (1997) which refers to ‘the set of human characteristics associated with a brand’. Research indicates several views on the relation between brand personality and brand image. Some research explains the relationship between brand personality and brand image (e.g., Patterson, 1999) while other research argues that the two constructs can be considered interchangeable (e.g., Hosany, Ekinci, & Uysal., 2006). Others refer to the brand image as a wide umbrella that also includes brand personality as a dimension (e.g., Patterson 1999; Plummer, 1985). In the Customer-Based Brand Equity (CBBE) model by Aaker (1992), brand personality is defined as an association type. This type of associations is a non-product related attribute, defined as an external aspect of the product or service that relate to its purchase or consumption (Keller, 1993).
In this research brand personality is considered as a type of association, similar to the description by Aaker and Keller. In this sense, brand personality is considered a dimension of the brand image with the latter being a wide umbrella. Brand personality as a tool which brands can leverage to differentiate themselves from competitors. The concept of brand personality is highlighted further below.
2.2 Brand Personality
Brand personality describes the ‘set of human characteristics associated with a brand’ (Aaker, 1997). To visualize this, this research refers to a classic example with the clear use of brand personalities. In the television advertisement ‘PC vs. Mac’ run by Apple featuring two spokesmen both personifying Microsoft and Apple (2006). Apple being young, hip and trendy while the other being old, stiff and old-fashioned. This campaign has been so successful for Apple that it forced Microsoft to have a counter advertisement refreshing its personality perception among consumers. The ‘PC vs. Mac’ advertisement describes the relevance of an existing brand personality and emphasises the importance of establishing and managing this to make the brand more appealing to consumers.In a research conducted by Aaker (1997), five types of concepts to best represent the brands are determined: Sincerity, Excitement, Competence, Sophistication and Ruggedness. Both sophistication and ruggedness tap a dimension that individuals desire but do not necessarily have (Aaker, 1997). Sophistication is often linked to upper-class brands such as Mercedes Benz, being charming and good-looking. Research indicates that consumers are expected to experience more difficulties with stepping down from upper class to a so called lower-class than the other way around. Ruggedness is visualized more masculine with brands like Marlboro or Harley Davidson, both described as though and outdoorsy brands. Sincere brands capture the idea of warmth and acceptance enhancing a positive feeling. Excitement is often presented with youthful, exciting brands like Red Bull and Coca Cola, being imaginative, social and energetic. The last concept of competence personalities like the Wall Street Journal or Microsoft are being presented as reliable, market leader brands encapsulating responsibility, dependability, and security (Aaker, 1997; Freling, Crosno and Henard, 2010). Figure 1 represents an extended overview of the Brand Personality Scale and can be found on the next page. Consensus among scholars on the impact of brand personalities has been growing. Brand personality has influences on consumer preference and usage (Sirgy, 1982), increases levels of trust and loyalty (Fournier, 1998), elicits consumer emotions (Biel 1993), stimulates active information processing (Biel, 1992), encourages self- expression and association (Belk 1988), provides a basis for product differentiation (Aaker, 1992), and influences brand attitudes and cognitive associations (Freling and Forbes, 2005; Freling, et al., 2010). With existence and increasing impact of such brand personalities, certain questions as “What
makes one brand personality better, or more impactful than the other” remain unanswered. By understanding this scale of brand personality appeal, managers and marketers may learn how brand personalities can influence attitude, behaviour and the relevance of particular personalities (Freling et al., 2010).
Aaker’s definition of brand personality as previously presented has been critically reviewed by other authors (e.g. Azoulay & Kapferer, 2003). This definition of brand personality is considered being too broad hereby, resulting in the embracement of concepts beyond those of brand personality. Defining brand personalities to loosely can result in the interpretation that nearly everything related to a human being, can be applied to brands (Azoulay & Kapferer, 2003). Even though researchers debate the presented definitions and the interchangeable character of brand personality (or not), the scale has been validated and refined in a variety of cultures and contexts (Aaker, 2000; Aaker, Garolera, & Benet-Martinéz, (2001). This scale currently stands as the dominant brand personality scale in the marketing literature (Freling et al., 2010). Figure 1: Brand Personality Scale Source: Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34, 347–356.
2.3 Brand equity
Building strong brands is one of the main goals of product and brand managers. Literature has shown that strong brands result in higher revenue streams, both in the short and long term (Aaker, 1991; Keller, 2003; Kapferer, 2004). Scholars have developed several conceptualizations of brand management and how brands affect consumer behaviour. Two of these concepts from the past are brand equity and customer-based brand equity (Aaker, 1991, 1992; Keller, 1993). These concepts focus on how knowledge structures affect consumers' awareness, brand image and brand personality. Since the term “brand equity” emerged in the 1980s, there has been a growing interest in the subject among marketing researchers and practitioners (Cobb-Walgren, Beal & Donthu, 1995). Financial motivations have been the main reason for studying brand equity. In other words, estimating the value of the brand for accounting purposes, and strategy-based for improving market productivity (Keller, 1993). Brand equity as a strategic instrument is considered as the driving force for increased market share and profitability of the brand which is based on the market’s perceptions (Christodoulides & De Chernatony, 2010). Research has been done to answer outcome- and process related questions like: ‘What makes a brand strong?’ and ‘How do you build strong brands?’. As a result, logical, well-integrated and grounded models have been developed reflecting state-of-the-art thinking about branding from both an academic as practical point of view (e.g. Keller, 2002). One of the most used concepts representing Brand Equity is that of Keller (1993, 2002) and Aaker (1991). The models and concepts can be described as Customer-Based Brand Equity (CBBE). CBBE describes in essence that the strength of a brand lies within the mind of the customer. The linkages between the brand and what a customer learns, feels, sees and hears about the brand over time. In this perspective, the big challenge for marketers is to create the right knowledge structures consisting of the desired thoughts, feelings, images, beliefs, perceptions, opinions, and so on and become linked to the brand (Keller, 2002).
2.4 Customer-Based Brand Equity (CBBE)
Customer-Based Brand Equity (CBBE), brand equity from the customer’s perspective, describes the differential effect of brand knowledge on customers’ response to the marketing of the brand. It describes customers’ familiarity, associations and favourability towards the brand (Washburn and Plank, 2002). Four main dimensions of CBBE will be presented.Keller (1993, 2001) outlines the CBBE model to assist managers in their brand-building efforts. According to Keller (1993, 2001) brand knowledge is the central component of the CBBE model. Brand knowledge describes customers’ cognitive representation of the brand (Peter & Olson, 2001)
The first two distinct types of brand knowledge Keller (1993) describes brand awareness and brand image. Brand awareness can be defined as the level to which customers are able to identify the brand under different conditions (Rossiter & Percy, 1987). Brand image is defined as the perception about a brand reflected by the brand associations held in the customer’s memory. These two types of brand knowledge have often been confirmed by different scholars in prior marketing research (Agarwal & Rao, 1996; Mackay, 2001; Pappu, Quester, & Cooksey, 2005). Keller (1993) divides the concept of brand awareness in two dimensions: brand recall and brand recognition. Brand recall as the customers' ability to retrieve the brand when given the product category, whereas brand recognition is described as the ability of customers to confirm prior exposure to the brand when given the brand as a cue. The relative importance of brand recall and recognition depends on the level to which product decisions are made in the store at the moment of purchase (where they potentially may be exposed to the brand) versus outside the store before the purchase, among other factors (Bettman, 1979; Rossiter & Percy, 1987).
According to the CBBE model, brand image can be described as the customers’ associations with the brand. Keller (2003) refers to the brand image as the strength, favourability and uniqueness of brand associations held in memory, which result in perceived quality, a positive attitude and overall positive affect (Esch, Langner, Schmitt, & Geus, 2006). According to Keller (2001) brand personality can be seen as non-product related attributes as a type of associations.
In the research of 2001, Keller presents the Customer-Based Brand Equity Pyramid which can be seen in figure 2. According to this theory, significant brand equity can only be created when the right four building blocks are in place. From bottom to top is (1) Identity, (2) Meaning, (3) Responses and (4) Relationship. This research focuses mostly on the second and third building block. With meaning describing brand performance and imagery through
creating strong, favourable and unique associations, and responses describing customer judgements and feelings expressed through positive, accessible responses. Figure 2: Customer-Based Brand Equity Pyramid Source: Keller, K. L. (2001). Building Customer-Based Brand Equity. Marketing Management, 10(2), 14-19.
2.5 Conceptual domain of customer-based brand equity
The present research conceptualizes CBBE according to the literature of Aaker (1991) and Keller (1993) based on customer perceptions. This section provides an overview of the four dimensions of CBBE examined in this study. Two often mentioned and empirically validated dimensions of CBBE which are not being used are brand associations and brand awareness (e.g. Pappu et al., 2005). Brand associations can serve as good indicators of the meaning a brand has for consumers (Keller, 1993), however, measuring associations with brands doesn’t add much value when considering the purpose of this research. The dimensions used to measure CBBE are brand uniqueness, brand loyalty and the related purchase intention, brand attitude and perceived quality. These dimensions are explained below. 1. Brand Uniqueness Brand uniqueness is defined as the degree to which customers feel the brand is different from competing brands (Netemeyera et al., 2004). The essence of brand positioning is to establish a sustainable competitive advantage often defined as the “Unique Selling Proposition”. Product attributes and associations can build the brand to the level at which it is perceived as unique, giving customers a compelling reason for buying a particular brand(Aaker, 1982). Brand uniqueness can help to understand how unique a brand is perceived in its environment and if this is in any sense related to the brand personality.
2. Brand Loyalty & Purchase Intention
Brand loyalty is defined as “the level of attachment that a customer has to a brand” (Aaker, 1991). Oliver (1997) defined it as: “a deeply held commitment to rebuy or repatronise a preferred product or service consistently in the future, despite situational influences and marketing efforts having potential to cause switching behaviour”. Whereas brand loyalty can be viewed from a behavioural perspective through the form of actual purchases and favouring a brand, can it also be seen from an attitudinal perspective. This attitudinal perspective on brand loyalty can be described as the tendency to be loyal to a focal brand, demonstrated through the intention of buying the brand as a primary choice (Yoo & Donthu, 2001) and considering it in the evoked set. As previously mentioned, brand equity is conceptualised from a customers’ perspective. Hence, this research extent the conceptualisation of brand loyalty from an attitudinal- and customer perspective as described with ‘Purchase Intention’.
3. Brand Attitude
Brand attitude can be described as the customers’ overall evaluation of the brand (Wilkie, 1986). One widely accepted approach to brand attitude is based on a multi-attribute formulation in which brand attitude is the function of the associated benefits and attributes that are salient to the brand (Keller, 1993). One of the most influential proposed multi-attribute model is that of Fishbein and Ajzen (1980). This expectancy-value model views brand attitude as a multiplicative function of (1) the salient beliefs a customer has about the brand/product and (2) the evaluative judgement of those attributes and benefits. This research focuses on the last part of this model, using brand attitude as an indication whether the overall evaluation of the brand is either positive or negative. 4. Perceived Quality The last dimension in the CBBE is perceived quality. Perceived quality is not the actual quality, rather the customers’ subjective evaluation of the products quality (Zeithaml, 1988). Like associations, perceived quality provides a reason for the customer to purchase products and does it differentiates the brand from competitors. It can be stated that all the CBBE dimensions correlate with one another. The higher
attitude of a customer is towards the brand, the more likely he or she develops favourable quality perceptions and brand loyalty (Pappu et al., 2005). The level of brand equity may have different impacts on customers shopping behaviour. For example, a customer may be more or less sensitive for certain associations. One concept that can mediate the impact of brand equity, is the level to which customers are driven by hedonic- or utilitarian motivations.
2.6 Hedonic and Utilitarian perceptions
When discussing hedonism and utility, scholars often refer to customers’ shopping motivations. These are mainly categorized as utilitarian and hedonic drives (Babin, Darden, & Griffin, 1994; Childers, Carr, Peck, & Carson, 2001; Kim, 2006). In the past, shopping has often been referred to as a rational process where the customer is in need for a specific product acquisition (Forsythe and Bailey, 1996; Fischer and Arnold, 1990; Sherry et al., 1993). Other research has abandoned the perspective that shopping is only build out of cognition and integrated hedonic- and emotional values referring to shopping as a way of leisure and recreation (Hirschman & Holbrook, 1982; Halvena & Holbrook, 1986; Bagozzi & Heatherton, 1994; Hoffman & Novak, 1996).
The impact of hedonic- or utilitarian shopping motivations can be experienced in the following two examples. A study by Forsythe and Bailey (1996) elaborated that shopping enjoyment is positively related to the amount of time spent shopping. In line with these results is the finding that recreational shoppers spend more time shopping than economic shoppers, although they spent less time deliberating before purchase (Bellenger & Korgaonkar, 1980). Another study examining the differences in roles between hedonic and utilitarian shopping motivations, found that customers who are highly motivated by hedonism tended to buy on a more frequent basis compared to those with a high level of utilitarianism. These two examples clearly indicate the impact of hedonic or utilitarian shopping motivations on the effect on brand loyalty, purchases and time spent. In terms of the means-end-chain as proposed by Wagner (2007) the utilitarian interpretation would describe concrete attributes and functional consequences with the hedonic interpretation describing psychosocial consequences and instrumental values.
Literature outlines hedonic brands as reflecting experiential values of shopping that include fantasy, arousal, sensory stimulation, enjoyment, pleasure, curiosity, and escapism (Scarpi, 2006; Hirschman and Holbrook, 1982). In this sense, are strong associations and perceived quality be seen as hedonic drivers. Utilitarian brands present the customer as being task-oriented, rational, and cognitive (Babin et al., 1994), with the intentions or desires to purchase a product efficiently and rationally highlighted.
This research considers two perceptions to take into account. On the one hand customers are driven by certain shopping motivations resulting in behaviour where they base their evaluations on hedonic- or utilitarian drivers. On the other, customers base their evaluations on how the brands position themselves as rather hedonic or utile focused, where the latter would function as a spectrum. An example of this spectrum could be a brand offering and promoting purely its core business for the lowest price possible or a brand wrapping its core business with a satisfying, full-service experience for a premium price. Many scholars concluded that utilitarianism and hedonism are essentially complementary and intertwined (Babin et al., 1994; Scarpi, 2006) and that customers use both concepts and benefits in different contexts.
When integrating these concepts in this research, it may appear that brand equity differences may play a lesser role in a utilitarian perspective when comparing core businesses than it would in a hedonic setting.
2.7 Conceptual Framework
Figure 3: Conceptual framework Figure 2 is the conceptual framework for this research. It has been developed to give an overview of the hypotheses which have been tested, and is further explained in the following subchapter. In addition, this framework has been developed to give insight in the CBBE dimensions that have been measured. Also, this figure provides mediating conditions on the which CBBE would differ under hedonic and utilitarian conditions or positioning.2.8 Literature gaps and Research question
As noted previously, brand image has been considered as one of the driving forces of brand asset and brand performance. Both academics and practitioners acknowledge the importance of creating a brand image and increasing the brand equity. Studies have been conducted proposing new ways on measuring brand equity, presenting multidimensional scales and discussing which variables influence a positive brand equity.Where brand equity is described as the differential response on marketing efforts, brand image can be seen as an influencing variable on brand equity outcomes. Few studies
elaborated on the relationship between brand image and brand equity being presented face-to-face, even fewer studies suggested or measured differences in a dimension of brand image e.g. brand personality. Therefore, this research utilizes a survey design to examine the effects of different brand personalities on customer-based brand equity and provide empirical evidence for these differences or not. Research Question: To what extent does Customer-Based Brand Equity differ between brand personalities, and is this relation mediated by hedonic or utilitarian positioning? This research incorporates several literature streams within the marketing literature. Key in this research is the personality dimension’s theory as presented by Aaker (1997) and the customer-based brand equity model as presented by Keller (1993). As previously described, this research measures the effects of a potential mediator based on the scale to which the brand is perceived as hedonic or utilitarian.
2.9 Theoretical and managerial contribution
This research contributes to the literature in the following manner. Firstly, this research gives insights in the level of the independent variable on how brand personalities differ from one another. The effects of brand personality, presented as a subpart of ‘brand knowledge’, has been empirically proved and the topic received more attention throughout the years (e.g. Keller, 1993; Aaker, 1997; Freling et al., 2010). Even though, no further exploration has taken place to see how brand personalities differ with their effects on CBBE. Empirical evidence on this detailed relationship does not yet exist in current literature. Secondly, this research further improves and gives sight in the importance per CBBE dimensions. Thirdly, while much CBBE literature emphasizes measuring equity dimensions they are often not linked to the underlying causes. Linking brand personalities with CBBE dimensions is a way to predict customer responses based on these brand images.
The results of this research can be used in several ways. Firstly, in a highly competitive environment, managers may seek for many ways to differentiate the brand from opposing brands in the industry. Understanding the appeal of brand personalities and its underlying
preferences for one or more brand personalities relate to positive or negative affected CBBE in general. Knowhow on positive and negative affected CBBE dimensions can increase the effectiveness of the brands positioning. When lacking progress on one or more equity dimensions, comparison between brand personalities can help understand how to adjust the brand formula in a detailed way. Thirdly, this research points out which differences exist when a brand is highly perceived as a utilitarian or hedonic brand. This customer’s perception may result in purchases based on the CBBE results of either hedonic or utilitarian positioning.
2.10 Hypotheses
This research focusses on a relation between an independent and dependent variable. The independent variable is the concept of ‘brand personality’ with the dependent variable indicating ‘customer-based brand equity’. The mediator included in this research is the buying motivation, whether brands are perceived from a utilitarian- or hedonic perception. Figure 4: Conceptual relations As shown above, the independent variable brand personality can be brought down to the personalities sincerity, excitement, competence, sophistication and ruggedness. CBBE in this research consist of brand uniqueness, brand loyalty and the related purchase intention, brand attitude and perceived quality. In this sense CBBE describes the customer responses to the different personalities. This all is visualized in figure 4.2.10.1. Expected differences and biggest impacts
According to Aaker (1997), the brand personalities Sophistication and Ruggedness tap a dimension that individuals desire but do not necessarily have. This is in line with sophisticated personality traits as upper-class, glamorous and sexy in which aspirational associations come forward or the glamorized masculine, strength and western ideals. While consumers desire and glamorize these associations, we expect consumers to perceive these brands at a higher quality. Therefore, we hypothesize that perceived quality of sophisticated and rugged brands are higher compared to other brand personalities. Hypothesis 1: Sophistication and ruggedness are perceived as better quality, compared to sincerity, excitement and competence.
As noted in the literature, perceived quality can be seen as a hedonic motive. We hypothesize that desire and aspirational associations are linked towards pleasure and enjoyment, with the latter being associated with hedonic perceptions (Hirschman & Holbrook, 1982; Scarpi, 2006). Building on the first hypothesis the second hypothesis states that we expect sophistication and ruggedness to perform better on CBBE outcomes when perceived from a hedonic state of mind. Hypothesis 2: Sophistication and ruggedness perform better on CBBE when perceived from a hedonic than a utilitarian perspective. One of the traits sophisticated brands are perceived by is upper-class. We expect it to be more difficult for consumers to step down from ‘upper-class’ to a so called ‘lower-class’ than the other way around. Therefore, we expect consumers to be more loyal towards sophisticated brands.
Hypothesis 3: Sophistication performs better on brand loyalty, compared to sincerity, excitement, competent and ruggedness.
Sincerity is closely related to personality traits as cheerful, honest and family-friendly. We expect that these are traits one can easily identify themselves with and positively associate with. As mentioned in the presented theory, the equity dimension brand attitude relates to positive or negative evaluation of the brand. We expect that associations from sincere brands are more likely to result in a positive evaluation of the brand. Hypothesis 4: Sincerity performs better on a positive brand attitude, compared to excitement, competent, sophistication and ruggedness
Excitement is associated with being imaginative, trendy, contemporary and young. Because consumers can find their imaginative, up-to-date image we expect this to lead to personal, unique brand images. Hypothesis 5: Excitement performs better on brand uniqueness, compared to sincerity, competence, sophistication and ruggedness. With competent brands being experienced as market leaders and success, we expect that these brands are quite known and address the needs of the population in general. Being picked up by the mainstream, we expect that these brand are experienced more as the least unique. Hypothesis 6: Competence performs worse on brand uniqueness, compared to excitement, competence, sophistication and ruggedness.
With competent brands being experienced as market leader, being reliable and intelligent, the chances are high that consumers expect this brand to successfully fulfil in the basic needs of the consumer and be consequently considered in the evoked set. We expect that competence performs better on purchase intention than other brand personalities.
Hypothesis 7: Competence performs better on purchase intention, compared to sincerity, excitement, sophistication and ruggedness.
Following up on this personality, being reliable and successful, we expect that consumers favour competence when perceived from a utilitarian perception. Hypothesis 8: Competence performs better on CBBE when perceived from a utilitarian perspective than from a hedonic perspective. Table 1: Hypotheses H1 Sophistication and ruggedness are perceived as better quality, compared to sincerity, excitement and competence. H2 Sophistication and ruggedness perform better on CBBE when perceived from a hedonic perspective than from a utilitarian perspective. H3 Sophistication performs better on brand loyalty, compared to sincerity, excitement, competent and ruggedness. H4 Sincerity performs better on a positive brand attitude, compared to excitement, competent, sophistication and ruggedness H5 Excitement performs better on brand uniqueness, compared to sincerity, competence, sophistication and ruggedness. H6 Competence performs worse on brand uniqueness, compared to sincerity, excitement, sophistication and ruggedness. H7 Competence performs better on purchase intention, compared to sincerity, excitement, sophistication and ruggedness. H8 Competence performs better on CBBE when perceived from a utilitarian perspective than from a hedonic perspective.
Chapter 3 Research Design and Method
To test the previously discussed hypotheses, as well as measure the effects of utilitarian and hedonic perceptions and additional demographic factors, data has been collected. This chapter explains these sections of the research. The first section of this chapter explains the sample, and follows up with the research design. The second section of this chapter presents the survey and input that has been used for this survey e.g. the pre-test that is held. The third section describes the used measures and the procedure for collecting data.
3.1 Sample
The population for this study is the Dutch, shopping consumer. Normally, the most representative way of sampling in this situation would be random sampling. While we don’t have access to a customer base, this is not an option. Therefore, this research uses convenience sampling. A convenience sample is a type of non-probability sampling where the researcher addresses a group of consumers that are readily accessible and convenient. This form of sampling is convenient, relative low in cost and a fast way of gathering data (Fricker & Schonlau, 2002). A disadvantage in using convenience sampling is that the results possibly are not representative for the entire population (Boxill, Chambers, & Wint, 1997). This research tries to overcome this disadvantage by searching for consumers in many different environments, e.g. different demographic, social and economic backgrounds. In statistics the word “sample” can be used in a variety of contexts. In this research, sampling relates to the selection of a subset of elements representative for the population. The population is defined as the collection of elements, which represent the research group (Salant, Dillman, & Don, 1994). These elements can be a group of brands, consumers and more. The first sample is that of the consumer. With this research addressing a universal phenomenon, the goal is to use a generic sample covering the shopping and consuming, of the Dutch population. This means consumers with varying ages, gender, interests, and educational levels. These consumers can be find both offline and online.The second sample used in this research is the set of brands. These brands need to represent all consumed brands. Aaker (1997) proposes three criteria, which need to be
addressed while choosing the set of brands. Firstly, these brands have to be salient, well-known brands. Secondly, the brands should represent a wide variety of brand personalities. Thirdly, the brand sample should address a wide range of product categories, both hedonism and utilitarian. Using a large set of brands can increase the generalizability and robustness of the measurement scale. A disadvantage however can be an increase in boredom and fatigue under the respondents resulting in response bias. Therefore, the brand sample for the pre-test consists out of 25 brands from fifteen different categories. Ten brands will be used in the main research, evaluated on their allocated personality and the relative percentage with which they are selected.
3.2 Research Design
An online survey is the chosen method for this research. In specific a questionnaire construct. The main reasons for choosing a quantitative, online survey are the direct data control, attractive pricing of the method and that respondents stay in their familiar environments at home increasing external validity of this study (Reips, 2000). A disadvantage can be a quality issue relating to non-response (e.g. boring, long) or errors across data (Kooiker, Broekhoff, & Stumpel, 2011). This is minimized by using two brands per personality resulting in a shorter survey and continuous expectation management throughout the survey.
Before the main research takes place, a pre-test is held. This pre-test has been conducted to determine the right set of brands to include in the brand sample. Sub-chapter 3.3. will further elaborate on the goals, method and results of this pre-test. To cover both the normal situation as well as the mediators, three surveys are being created and spread. The choice for three separate surveys is made to prevent respondents bias in one of the three conditions. For the normal condition this research strives to collect between 100 and 120 completed surveys. With time being a limitation to fully exploit the ins-and-outs of the topic, the goal is to gather between 30 and 40 responses for both surveys researching the mediating conditions. In the case of a normal distribution, roughly bell shaped spread, 30 respondents are enough to execute statistical analysis (Saunders & Lewis, 2012). Otherwise these results will be used as an indication for potential differences. To increase the understanding in the reasoning behind the concepts and to prevent misinterpretation of the results a few unstructured, qualitative interviews will have been held
3.3 Pre-test: Brand generation
A pre-test has been executed to generate a set of ten brands for the main research. Each personality would in this way have two brands that are evaluated and compared. After careful considering a cross-category set of 25 brands was selected to be included in the pre-test. These brands were chosen from either previous research (i.e. Aaker, 1991) or personally by the researchers’ judgement. 3.3.1. Method The pre-test was executed by using an online survey, where respondents divided the brands into five personality dimensions together with fifteen personality traits as proposed by Aaker (1991). The first set of brands were selected in such a way, that each personality would have five potential brands that could fit with this personality. A full overview of the study and brands included in the pre-test can be found in attachment 1. 3.3.2. Results The end sample included 36 respondents. Except for “Excitement”, all personalities had two or more brands that were allocated to its personality by at least 50% of the respondents. Respondents experienced more difficulties with allocating brands to the personality excitement. The lowest percentage of allocation for excitement brands was at 45% of the respondents. The selected brands per personality that is used in this research, together with their allocation percentages, is listed in the table below (table 2). Table 2: Pre-test resultsBrand Personality Brand 1 Allocation Brand 2 Allocation Traits
Sincerity Dove 75% Toyota 68%
Family-oriented, Honest & Friendly
Excitement Pepsi 47% Red Bull 45% Daring, Cool & Imaginative
Competence Microsoft 67% Sony 56% Reliable, Intelligent & Confident leader Sophistication Rolex 91% Nespresso 66% Upper-class, Charming & Glamorous
3.4 Survey and measures
Three surveys have been used to gather data for this research. All surveys include the ten brands that re chosen from the pre-test results: Dove, Toyota, Pepsi, Red Bull, Microsoft, Sony, Rolex, Nespresso, Jack Daniels and Jeep. The survey itself consists out of ten brands with ten evaluative statements per brand. This means two brands per personality, two statements per brand equity dimension. All brand pages showed only the brand name together with the brand’s logo retrieved from the internet. To make sure respondents are aware, they are asked a question whether they are thinking about the product, brand or associations of the presented brand. Each CBBE dimension has one statement positively framed and one counter-indicative statement negatively framed. These statements can be answered with a Likert-point-scale of five points, with “strongly disagree” or “strongly agree” as the two extremes. The statements are adapted from previous surveys and researches found in the literature. An overview of all questions per condition is presented in in attachment 2. The three surveys differ in the way questions are asked and how the statements are formulated. In the normal condition, respondents are asked to think of the brand in general. In the hedonic condition, respondents are asked to think of the associations related to the brand. In a utilitarian setting, respondents are asked to think of the products relating to the brand and their functional benefits. In each situation, a few examples (i.e. associations; functional benefits) are given in the introduction. Thereby, the statements are adjusted to each condition, e.g. “Brand X is my favourite brand” for the general condition and “Their products are my favourite products” for the utilitarian condition. The survey concludes with several demographic questions. These questions are about age, gender and education. These so-called ‘control variables’ are used to measure how much random variables differ from one another.3.5 Procedure Data Collection
As explained in the subchapter 3.2 the data has been collected through an online survey. The survey is created with Qualtrics on www.qualtrics.com. Qualtrics is easy to use, provides a wide range in types of questions and other tools e.g. randomization of questions.
3.6 Main Study
As mentioned above, the surveys have been created and hosted by Qualtrics. The normal condition survey was online from the 17th of May till the 31st of May, fifteen days in
total. The survey was spread using social media (e.g. Facebook; LinkedIn), email and by asking people to fill it in. The snowball technique (asking respondents to share the survey) was used to gather more responses. It is unclear how many people had seen the survey, but 179 people started with survey and 103 respondents finished the survey completely. The hedonic and utilitarian positioned survey was online from Monday 6th of June till Monday the 13th of June, in total 7 days. Similar to the normal condition, were these surveys spread by using social media, email and personal invitation. For the utilitarian condition 30 finished surveys were gathered. The hedonic situation gathered 31 filled-in surveys. Filling in the survey was in all situations voluntarily. The respondents were assured of complete anonymity and all surveys received were kept strict confidential.
Chapter 4 Results and Analysis
This chapter covers the empirical part of the thesis and explains the models and tests used for statistical analysis. Descriptive statistics of the dataset are presented, reliability of the scales are measured and the hypotheses are tested with SPSS Statistics.4.1 Preliminary analysis
As mentioned in subchapter 3.6 179 people started the survey. From all surveys, 103 were completely finished and usable for statistical analysis. The other 76 surveys had too much data missing, and were therefore removed from the data set. After removing the 76 surveys there were no missing values in the dataset anymore. Several items in the dataset were counter indicative to prevent response bias. These statements (4, 5, 6, 9 and 10) were recoded. To test the covariance variables are the scales recoded into dummy variables. The variable gender was recoded from 1 (male) and 2 (female) to 0 (male) and 1 (female). With the same reason, age is recoded using the following scale: 0 (<18), 1 (19-25), 2 (26-35), 3 (36-45), 4 (46-55), 5 (56-65) and 6 (> 65), and educational level to 0 (High School), 1 (Intermediate Vocational Education – MBO), 2 (University of Applied Sciences – HBO), 3 (University of Scientific Education – WO) and 4 (other). From the 103 respondents in the dataset, 55,3% were men and 44,7% women. 72,8% of all respondents are between the age of 18 and 25, and most of them (40,8%) study or studied at the University of Applied Sciences (HBO). Slightly less, 33%, study or finished their study at the University of Scientific Education (WO).4.2 Reliability and scale means
The level of reliability describes the overall consistency of the measure (Saunders et al., 2009). In other words, measuring the consistency if this research were to be repeated a second or third time. All questions were checked for reliability with the Cronbach’s alpha, the most common scale for testing the reliability (Field, 2013). In this study all variables are scale items. The reliability has been tested by using two statements per dimension. The final means will be a combination of the presented brands per brand personality as covered in subchapterAs presented in table 3 are some scales not reliable. The Cronbach’s alpha for brand uniqueness for all brands does not exceed the .7 scale. This dimension won’t be removed, but tested for indicative purposes. All other CBBE dimensions not exceeding the .7 alpha are dropped for statistical testing, except for two scales with an alpha of >.69. When one of the two brands (-dimension) is reliable while the other brand in the same personality is not, the reliable brand scale is kept as representative for the brand personality. All dimensions were checked on Cronbach’s Alpha (> .7) and Corrected Item-Total Correlation (>.3). Cronbach’s Alpha If Item Deleted has not been taken into account, while the scales were all two-item measures. As a final step of data preparing before executing the different tests are scale means computed from those dimensions which are tested reliable. In this part brand dimensions belonging to the same brand personality are combined. The means and standard deviations can be seen in table 3 together with the amount of brands with reliable scales used.
4.3 Analysis
Most of the hypotheses are based on the assumption that the mean of one variable is significantly lower or higher than the means of other variables within the same group of respondents. Therefore, to analyse and test these assumptions, a “Paired Samples T-test” is run. Because the mediating effects are measured through separate surveys, these means can be easily compared with a Paired Samples T-Test as well. All data is tested with a 95% reliability, giving a significance level of p < .05 meaning that hypotheses with a p-value higher than .05 are rejected.For hypothesis 1 all scales tested reliable for perceived quality. The results of the test show that perceived quality of the brand personality sophistication (μ = 4.36; σ = .48) is significantly different than excitement (μ = 3.41; σ = .72), sincerity (μ = 3.62; σ = .61), ruggedness (μ = 3.75; σ = .58) and competence (μ = 3.96; σ = .49). All p-values are below .05. As can be read above, is the mean of ruggedness lower than both sophistication and competence and can therefore not be perceived as higher quality than the other brand personalities. No further tests are needed.
The results of testing hypothesis 2 show that a hedonic view on the brand personality sophistication (μ = 3.80; σ = .60) results in higher brand uniqueness when perceived from a
utilitarian perspective (μ = 3.33; σ = .45). The other dimensions are not (significantly) higher evaluated. This difference does not exist when compared to the normal condition.
A similar effect is found with the personality ruggedness. When the personality ruggedness is perceived from a hedonic perception (μ = 3.64; σ = .53), brand uniqueness is evaluated significantly higher than when perceived from a utilitarian perception (μ = 3.18; σ = .54). The other dimensions are not (significantly) higher evaluated. This difference also exists when compared to the normal condition.
For hypothesis 3 are three brand loyalty scales tested reliable, which are that of sophistication, competence and excitement. The test results show that the dimension brand loyalty for sophistication (μ = 2.77; σ = .90) is not significantly different than competence (μ = 2.69; σ = .84) and excitement (μ = 2.67; σ = 1.09). All p-values are higher than .05. For hypothesis 4 are all personality scales tested reliable for brand attitude. The results show that brand attitude for the brand personality sincerity (μ = 4.11; σ = .47) is significantly higher than both brand attitude towards excitement (μ = 3.66; σ = .79) and ruggedness (μ = 3.82; σ = .72). These values have a p-value lower than .05. The difference in brand attitude between sincerity and sophistication (μ = 3.99; σ = .79) is not significantly (p > .05). Competence is tested with a higher mean on brand attitude (μ = 4.19; σ = .70), but the difference with sincerity is not significant as well (p = .254).
While this research didn’t provide enough reliable scales to test hypothesis 5 and 6, these results are used for indicative purposes. The results show that sophisticated brands score highest on brand uniqueness (μ = 3.54; σ = .66), followed by ruggedness (μ = 3.22; σ = .58), competence (μ = 3.10; σ = .62), excitement (μ = 3.03; σ = .56) and at last sincere brands (μ = 2.78; σ = .54). When not considering the reliability of the scales, both sophistication and sincerity are significant higher and lower than the other brand personalities in their brand uniqueness.
With hypothesis 7, four brand personalities are used which were tested reliable on their scales for purchase intention. These four brand personalities are competence, excitement, sophistication and ruggedness. The results of the test show that purchase intention for competent brands (μ = 3.71; σ = .88) is significantly higher than purchase intention for sophisticated brands (μ = 3.26; σ = .68), ruggedness brands (μ = 3.03; σ = .66 and
The test results of hypothesis 8 do not show any significant differences in CBBE for the personality competence, when comparing the utilitarian condition to the hedonic and normal condition. Table 3: Variables, dimensions, means and std. deviations across conditions
Normal Utilitarian Hedonic
Variable Dimension Mean Std. Dev. Mean Std. Dev. Mean Std. Dev.
Sincerity Brand Uniqueness 2.8* 0.54 2.7 0.43 2.8 0.49 Brand Loyalty 2.5* 0.60 2.5 0.56 2.2 0.79 Purchase Intention 3.25* 0.57 3.1 0.42 3.2 0.61 Perceived Quality 3.6 0.61 3.2 0.34 3.7 0.48 Brand Attitude 4.1 0.47 3.5 0.49 3.7 0.44 Excitement Brand Uniqueness 3.0* 0.56 2.7 0.83 3.4 0.68 Brand Loyalty 2.7 1.08 2.4 0.64 2.0 0.57 Purchase Intention 3.0 0.70 3.1 0.44 2.9 0.80 Perceived Quality 3.4 0.72 3.2 0.60 3.1 0.54 Brand Attitude 3.6 0.79 3.1 0.74 3.2 0.76 Competence Brand Uniqueness 3.1* 0.62 3.1 0.43 3.3 0.45 Brand Loyalty 2.7 0.84 2.9 0.71 2.6 0.72 Purchase Intention 3.7 0.87 3.4 0.73 3.4 0.40 Perceived Quality 4.0 0.49 3.5 0.53 3.5 0.49 Brand Attitude 4.2 0.70 3.6 0.67 3.6 0.41 Sophisticated Brand Uniqueness 3.4 0.89 3.3 0.45 3.8 0.60 Brand Loyalty 2.8 0.90 2.5 0.32 2.2 0.63 Purchase Intention 3.3 0.68 3.4 0.41 3.1 0.44 Perceived Quality 4.4 0.48 4.2 0.54 4.2 0.42 Brand Attitude 4.0 0.79 3.8 0.38 3.7 0.58 Ruggedness Brand Uniqueness 3.2* 0.58 3.2 0.54 3.6 0.52 Brand Loyalty 2.4* 0.60 2.7 0.46 2.3 0.75 Purchase Intention 3 0.66 3.2 0.50 2.9 0.82 Perceived Quality 3.8 0.58 3.6 0.49 3.3 0.74 Brand Attitude 3.8 0.72 3.6 0.52 3.6 0.45 * = Not reliable according to Cronbach’s Alpha (< .7), only for normal condition
4.3.1 Covariance variables
The covariance variables for this research are gender, age and educational level. Covariance variables are used to measure how much random variables differ from one another.
For the variable gender an Independent-Samples T-test was run to check if there are any eye-catching differences between these groups. Gender is in this situation the independent variable and the brand personality equity dimensions the dependent variables. In a few situations male and female significantly differed in their evaluation of equity dimensions from brand personalities. Females (μ = 3.78; σ = .59) evaluated the perceived quality of sincere brands significantly higher than males (μ = 3.50; σ = .61). A similar effect was found in their evaluation of sincere brands on brand attitude. Females (μ = 4.22; σ = .47) evaluated their attitude towards sincere brands as more positive than males (μ = 4.02; σ = .47). Finally, did females evaluate their brand attitude towards competent brands (μ = 4.36; σ = .48) as more positive than males (μ = 4.06; σ = .82). For the second and third covariance variables, age and educational level, a one-way ANOVA test was run. Age and educational level as the independent variables with the equity dimensions per brand personality as dependent variables. Before looking at the factorial ANOVA results, the Levene’s test was used to check whether the assumption of equal variances was not violated. The Levene’s test for homogeneity of variances showed that indeed several variables had a significance level greater than .05. After checking the results of the factorial ANOVA, significant differences in both age as educational groups were found in the brand personality excitement. These differences, however, are quite small and look rather like a coincidence than a useful statistic. Therefore, it will not be further mentioned.
Chapter 5 Discussion
In this chapter the most significant findings are discussed. The research question is answered and the results are explained in respect to the existing literature. Furthermore, the limitations of this study are described together with a number of suggestions for future research.
5.1 Discussion of the results
The test results show that each brand personality has its unique characteristics and related CBBE dimensions where it exceeds other personalities. For the first hypothesis the CBBE effects of sophistication and ruggedness were measured. Results show that sophisticated brands are perceived as higher quality than the other brand personalities. This partially supports the hypothesis where it’s expected both sophistication and ruggedness to be evaluated as higher quality. Due to the upper-class, glamorous and sexy traits functioning as aspirational associations we expected sophistication to be experienced as higher quality. The quality of ruggedness was evaluated as the third highest, leaving sincerity above him.
The mediating conditions have interesting effect for both sophistication and ruggedness as well. While not necessarily the perception of the quality improved, did the evaluation of brand uniqueness become significantly more positive when the brand was perceived from a hedonic perspective. This is in line with the literature, where hedonic shopping motivations are linked with strong aspirational associations (Scarpi, 2006; Hirschman & Holbrook, 1982). These associations may give the brand, in a hedonic environment, more ways to differentiate themselves and make them perceive as more unique. When consumers perceive the brand more rational, as utilitarian shopping motivations describe, the core business of the brand may seem more similar to that of other brands. These results indicate as well that either a utilitarian or hedonic perception of the brand can be stimulated. This is useful for brands that differentiate themselves with functional benefits or more abstract associations. In this way they can strengthen their positioning and increase their marketing value without necessarily adjusting the marketing strategy. Hypothesis 3 predicted differences in the evaluation of loyalty between personalities. It’s expected that consumers find it more difficult to switch from the glamorous, upper-class brands to other brand personalities with less aspirational associations. The results showed
that consumers seem to perceive themselves as to be more loyal towards sophisticated brands, however this difference was not tested significant.
All brand personalities were tested and compared on the brand attitude dimension. With sincerity being perceived as rather cheerful, honest and family friendly, it was hypothesized that consumers would positively associate themselves with these traits and therefore evaluate the brand attitude more positive. The results indicated that the personalities sincerity and competence are both evaluated as the most positive on brand attitude with no significant differences between these two. The personality sophistication closely followed as the third most positive brand attitude, indicating no significant difference with sincerity as well.
As Netemeyera et al. (2004) proposed, is brand uniqueness defined as the degree to which customers feel the brand is different from competing brands. Unique brand attributes and associations are mentioned for giving consumers compelling reasons to buy these brands (e.g. Aaker, 1992). While we predicted excitement to score well on this dimension, being young, trendy and imaginative resulting in unique images of the brand, sophistication scored highest on brand uniqueness. This can be related back to the strong associations sophisticated brands possess, in line with the literature as proposed by Aaker (1997). As previously mentioned does the evaluation of brand uniqueness become more positive when perceived from a hedonic perspective. Because the brand uniqueness scales are not tested reliable, can these differences not be tested on their significance level.
Sincerity indicated to be the least brand unique of all brand personalities. This is not in line with our prediction, where we expected competence to be perceived as the least unique brand personality. Consumers explain this difference with their consideration of sincere brands to be “one-of-many” when considering the product categories. Competent brands, which we predicted to be the least unique, are much more seen as a brand leading and developing the market, having lots of expertise and know-how. Because of the latter, competent brands are seen as more outstanding than sincere brands. Similar as with hypothesis five, are the brand uniqueness scales not tested reliable and can the results not be tested on their significance level.
