How do brand human characteristics affect brand evaluations?

(1)

Master Thesis

How do brand human characteristics affect brand evaluations?

University of Amsterdam Faculty of Economics and Business

Master of Science in Business Administration Track: Marketing

Supervisor: drs. Jorge Labadie MBM Second reader: drs. R.E.W. Pruppers

Student: Vyara Loevig Student number: 10825142

August 27th, 2015 Final version

(2)

Statement of Originality

This document is written by student Vyara Loevig 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

The current study examines the effect of brand age, brand gender and brand personality dimensions as a part of brand personality on consumer brand evaluations – brand attitude and purchase intention. Survey data are purchased from Qualtrics online research platform. The sample consists of 239 general population respondents from the United States of America. The results indicate that old brand personalities have a less positive effect on consumer brand evaluations in comparison to young brand personalities. Furthermore, study results indicate that a less strong brand gender might have a positive impact on brand attitude and purchase intention. Regarding brand personality dimensions it is found that when brands are defined as sincere, competent, sophisticated or exciting, they positively affect brand attitude and purchase intention. Rugged brands only positively impact brand attitude before presenting the stimuli (priming). However, they don't influence neither of the brand evaluations after priming. Furthermore, findings suggest that it is recommended that brands have a strong brand personality in comparison to a strong age or gender. Additional study findings show that there is a more positive attitude towards the ad when the gender of the endorsers in the ad contradicts the gender of the brand. In addition, there is a more favorable attitude towards the ad when the ad is perceived as a good fit for the brand. Lastly, brand familiarity is more important than any single brand personality trait in terms of brand evaluations. This study concludes with discussion of the results, theoretical contributions, managerial contributions and suggestions for further research.

Keywords: brand age, brand gender, brand personality, brand evaluations, brand attitude,

(4)

INDEX OF TABLES

Table 1 Correlation coefficients, variables in main test ... 41

Table 2 Model summary for linear regression model on brand attitude ... 42

Table 3 Model summary for linear regression model on purchase intention ... 45

Table 4 Repeated-measures design (GLM 4). Tests of within-subjects contrasts of old variable ... 49

Table 5 Repeated-measures design (GLM 4). Tests of between-subjects effects of old variable ... 49

Table 6 Repeated-measures design (GLM 4). Tests of within-subjects contrasts of young variable ... 50

Table 7 Repeated-measures design (GLM 4). Tests of between-subjects contrasts of old variable ... 50

Table 8 Repeated-measures design (GLM 4). Tests of within-subjects contrasts of gender variable ... 51

Table 9 Repeated-measures design (GLM 4). Tests of between-subjects contrasts of gender variable ... 51

Table 10 Summary of hypotheses testing ... 57

Table 11 One-way ANOVA. Descriptives for male score ... 79

Table 12 One-way ANOVA. Descriptives for female score ... 80

Table 13 One-way ANOVA. Descriptives for young score ... 81

(9)

Table 15 One-way ANOVA. Descriptives for sincere score ... 83

Table 16 One-way ANOVA. Descriptives for exciting score ... 84

Table 17 One-way ANOVA. Descriptives for competent score ... 85

Table 18 One-way ANOVA. Descriptives for sophisticated score ... 86

Table 19 One-way ANOVA. Descriptives for rugged score ... 87

Table 20 One-way ANOVA. Descriptives for familiarity score ... 88

Table 21 One-way ANOVA. Test of homogeneity of variances ... 89

Table 22 One-way ANOVA test results ... 90

Table 23 One-way ANOVA. Robust tests of equality of means ... 91

Table 24 One-sample T-test for male score ... 93

Table 25 One-sample T-test for female score ... 95

Table 26 One-sample T-test for young score ... 97

Table 27 One-sample T-test for old score... 99

Table 28 One-sample T-test for sincere score ... 101

Table 29 One-sample T-test for exciting score ... 103

Table 30 One-sample T-test for competent score ... 105

Table 31 One-sample T-test for sophisticated score ... 107

Table 32 One-sample T-test for rugged score ... 109

Table 33 Descriptives and one-sample T-test results from main study ... 128

Table 34 Key statistics for unprimed variables, all brands ... 129

Table 35 ANOVA test results for linear regression model on brand attitude ... 136

Table 36 Coefficient table for linear regression model on brand attitude ... 137

Table 37 ANOVA test results for linear regression model on purchase intention ... 138

(10)

Table 39 Description of variable codes used in priming models and additional findings ... 140 Table 40 Key statistics for primed variables, all brands ... 141 Table 41 Repeated-measures design (GLM 4). Between-Subjects factors for manipulation check of old variable ... 147

Table 42 Repeated-measures design (GLM 4). Descriptive statistics for manipulation check of old variable ... 147

Table 43 Repeated-measures design (GLM 4). Multivariate tests for manipulation check of old variable ... 148

Table 44 Repeated-measures design (GLM 4). Test of sphericity for manipulation check of old variable ... 148

Table 45 Repeated-measures design (GLM 4). Between-subjects factors for manipulation check of young variable ... 151

Table 46 Repeated-measures design (GLM 4). Descriptive statistics for manipulation check of young variable ... 152

Table 47 Repeated-measures design (GLM 4). Multivariate tests for manipulation check of young variable ... 152

Table 48 Repeated-measures design (GLM 4). Test of sphericity for manipulation check of old variable ... 153

Table 49 Repeated-measures design (GLM 4). Between-subjects factors for manipulation check of gender variable ... 156

Table 50 Repeated-measures design (GLM 4). Descriptive statistics for manipulation check of gender variable ... 156

Table 51 Repeated-measures design (GLM 4). Multivariate tests for manipulation check of gender variable ... 157

(11)

Table 52 Repeated-measures design (GLM 4). Test of sphericity for manipulation check of gender variable ... 157

Table 53 Mixed design ANOVA. Between-subjects factors for priming test on brand attitude... 160

Table 54 Mixed design ANOVA. Descriptive statistics for priming test on brand attitude (part 1) ... 161

Table 55 Mixed design ANOVA. Descriptive statistics for priming test on brand attitude (part 2) ... 162

Table 56 Mixed design ANOVA. Multivariate tests for priming test on brand attitude (part 1) ... 163

Table 57 Mixed design ANOVA. Multivariate tests for priming test on brand attitude (part 2) ... 164

Table 58 Mixed design ANOVA. Test of sphericity for priming test on brand attitude . 165 Table 59 Mixed design ANOVA. Tests of within-subjects contrasts for priming test on brand attitude ... 165

Table 60 Mixed design ANOVA. Tests of between-subjects contrasts for priming test on brand attitude ... 166

Table 61 Mixed design ANOVA. Marginal mean estimates by gender priming type for priming on brand attitude ... 166

Table 62 Mixed design ANOVA. Pairwise comparisons by gender priming type for priming on brand attitude ... 167

Table 63 Mixed design ANOVA. Univariate tests by gender priming type for priming on brand attitude ... 167

Table 64 Mixed design ANOVA. Marginal mean estimates by age priming type for priming on brand attitude ... 167

(12)

Table 65 Mixed design ANOVA. Pairwise comparisons by age priming type for priming on brand attitude ... 168

Table 66 Mixed design ANOVA. Univariate tests by gender priming type for priming on brand attitude ... 168

Table 67 Mixed design ANOVA. Between-subjects factors for priming test on purchase intention ... 168

Table 68 Mixed design ANOVA. Descriptive statistics for priming test on purchase intention (part 1) ... 169

Table 69 Mixed design ANOVA. Descriptive statistics for priming test on purchase intention (part 2) ... 170

Table 70 Mixed design ANOVA. Multivariate tests for priming test on purchase intention (part 1) ... 171

Table 71 Mixed design ANOVA. Multivariate tests for priming test on purchase intention (part 2) ... 172

Table 72 Mixed design ANOVA. Test of sphericity for priming test on purchase intention ... 173 Table 73 Mixed design ANOVA. Tests of within-subjects contrasts for priming test on purchase intention ... 173

Table 74 Mixed design ANOVA. Tests of between-subjects effects for priming test on purchase intention ... 174

Table 75 Mixed design ANOVA. Marginal mean estimates by gender priming type for priming on purchase intention ... 174

Table 76 Mixed design ANOVA. Pairwise comparisons by gender priming type for priming on purchase intention ... 175

(13)

Table 77 Mixed design ANOVA. Univariate tests by gender priming type for priming on purchase intention ... 175

Table 78 Mixed design ANOVA. Marginal mean estimates by age priming type for priming on purchase intention ... 175

Table 79 Mixed design ANOVA. Pairwise comparisons by age priming type for priming on purchase intention ... 176

Table 80 Mixed design ANOVA. Univariate tests by age priming type for priming on purchase intention ... 176

Table 81 Tests of between-subjects effects, ANCOVA with ad attitude as dependent variable ... 177

Table 82 Estimated marginal means for gender priming, ANCOVA with ad attitude as dependent variable ... 178

Table 83 Pairwise comparisons for gender priming, ANCOVA with ad attitude as dependent variable ... 178

Table 84 Univariate tests for gender priming, ANCOVA with ad attitude as dependent variable ... 178

Table 85 Estimated marginal means for age priming, ANCOVA with ad attitude as dependent variable ... 179

Table 86 Pairwise comparisons for age priming, ANCOVA with ad attitude as dependent variable ... 179

Table 87 Univariate tests for age priming, ANCOVA with ad attitude as dependent variable ... 179

INDEX OF FIGURES

(14)

Figure 2 Pretest results ... 32

Figure 3 β-values from linear model on brand attitude (third block) ... 44

Figure 4 β-values from linear model on purchase intention (third block) ... 47

Figure 5 Partial η2-values from mixed model on brand attitude ... 54

Figure 6 Partial η2-values from mixed model on purchase intention ... 56

Figure 7 One-sample T-test for male score ... 92

Figure 8 One-sample T-test for female score ... 94

Figure 9 One-sample T-test for young score ... 96

Figure 10 One-sample T-test for old score ... 98

Figure 11 One-sample T-test for sincere score ... 100

Figure 12 One-sample T-test for exciting score ... 102

Figure 13 One-sample T-test for competent score ... 104

Figure 14 One-sample T-test for sophisticated score ... 106

Figure 15 One-sample T-test for rugged score ... 108

Figure 16 Young variable histogram ... 130

Figure 17 Old variable histogram ... 130

Figure 18 Sincere variable histogram ... 131

Figure 19 Exciting variable histogram ... 131

Figure 20 Competent variable histogram ... 132

Figure 21 Sophisticated variable histogram ... 132

Figure 22 Rugged variable histogram ... 133

Figure 23 Brand attitude histogram ... 133

Figure 24 Purchase intention histogram ... 134

Figure 25 Absolute value of age variable histogram ... 134

(15)

Figure 27 Respondent age variable histogram ... 135

Figure 28 Brand familiarity variable histogram... 136

Figure 29 Primed male score variable histogram ... 142

Figure 30 Primed female score variable histogram ... 142

Figure 31 Primed young score variable histogram ... 143

Figure 32 Primed old score variable histogram ... 143

Figure 33 Primed brand familiarity variable histogram ... 144

Figure 34 Primed brand attitude variable histogram... 144

Figure 35 Primed purchase intention variable histogram ... 145

Figure 36 Primed ad attitude variable histogram ... 145

Figure 37 Primed ad fit variable histogram ... 146

Figure 38 Absolute value of primed gender variable histogram ... 146

Figure 39 Estimated marginal means of old variable for Apple (1 = unprimed, 2 = primed) ... 149

Figure 40 Estimated marginal means of old variable for Chanel (1 = unprimed, 2 = primed) ... 149

Figure 41 Estimated marginal means of old variable for Harley Davidson (1 = unprimed, 2 = primed) ... 150

Figure 42 Estimated marginal means of old variable for Nivea (1 = unprimed, 2 = primed) ... 150

Figure 43 Estimated marginal means of old variable for Red Bull (1 = unprimed, 2 = primed) ... 151

Figure 44 Estimated marginal means of young variable for Apple (1 = unprimed, 2 = primed) ... 153

(16)

Figure 45 Estimated marginal means of young variable for Chanel (1 = unprimed, 2 = primed) ... 154

Figure 46 Estimated marginal means of young variable for Harley Davidson (1 = unprimed, 2 = primed) ... 154

Figure 47 Estimated marginal means of young variable for Nivea (1 = unprimed, 2 = primed) ... 155

Figure 48 Estimated marginal means of young variable for Red Bull (1 = unprimed, 2 = primed) ... 155

Figure 49 Estimated marginal means of gender variable for Apple (1 = unprimed, 2 = primed) ... 158

Figure 50 Estimated marginal means of gender variable for Chanel (1 = unprimed, 2 = primed) ... 158

Figure 51 Estimated marginal means of gender variable for Harley Davidson (1 = unprimed, 2 = primed) ... 159

Figure 52 Estimated marginal means of gender variable for Nivea (1 = unprimed, 2 = primed) ... 159

Figure 53 Estimated marginal means of gender variable for Red Bull (1 = unprimed, 2 = primed) ... 160

(17)

1

1. INTRODUCTION

Brands, similar to people, have personalities. Brand personality is described as “a set of human characteristics associated with a brand’’ (Aaker, 1997, p.347). Aaker (1997) brand personality dimensions define brands as sincere, competent, exciting, sophisticated and rugged.

In addition to personality traits, brand personality consists of the demographic characteristics gender, age and class (Aaker, 1997) so that brands are perceived as masculine or feminine, young or old, upper or lower class.

General introduction: Phenomenon

Consumers’ perceptions of brand personality is a largely used phenomenon by marketers (Puzakowa, Kwak, & Rocereto, 2009). For example, brands can be exciting such as Mountain Dew and BMW (Aaker, Fournier, & Brasel, 2004) or one of a kind and amusing such as Dr. Pepper (Aaker, 1997). Moreover, Aaker (1997) illustrates brand personality with an example of Absolut vodka as young and modern and Stoli vodka as an intelligent, old-fashioned elderly man.

Brands have faces and names and human emotions are ascribed to them. Brand names can be associated with existing people (e.g. Mr. Kleen) and a first-person language is often used in brand communication as if, similar to people, the brand addresses consumers directly in a conversation (Aggarwal & McGill, 2012). Moreover, marketers frequently humanize brands in their brand positioning efforts. This marketing communication technique results in positive consumer reactions such as greater liking of products, more positive emotional responses as well as more favorable characteristics of brand personality (Delbaere,McQuarrie, & Phillips, 2011).

(18)

2

Specific introduction: Gap

Brand personality is considered a popular topic in the marketing literature for over five decades but there is a lack of evidence which factors form perceptions of brand personality (Maehle, Otnes, & Supphellen, 2011). Thus, these scholars investigate how consumers form their perceptions of the brand personality dimensions proposed by Aaker (1997), identify which brands consumers perceive as representatives of each of the brand personality dimensions and discover their common characteristics. However, no other brand personality traits are taken into consideration, for example whether the researched brands are perceived as young or old, upper or lower class. It could be of interest to find out whether brands are perceived as young or old, upper or lower class, because these findings could be useful to determine the actual consumer brand personality perceptions rather than those intentionally created by brand managers.

Brand personality has been researched in terms of congruity between consumers’ self-image and brands’ personalities (Swaminathan, Stilley, & Ahluwalia, 2009; Malär, Krohmer, Hoyer, & Nyffenegger, 2011; Mulyanegara, Tsarenko, & Anderson, 2009), attitude towards the brand, brand preference and brand satisfaction (Ivens & Valta, 2012), influence of metaphoric ads (Ang, 2006), personality transfer between brands and consumers after using the brands (Park & John, 2010). As a part of brand demographics and therefore brand personality, brand gender has been researched in terms of brand name (Wu, Klink, & Guo, 2013; Lieven, Grohmann, Herrmann, Landwehr, & Tilburg, 2015), color and logo (Lieven, et al., 2015), type fonts (Grohmann, 2014; Lieven, et al., 2015), gender dimensions of brand personality (Grohmann, 2009; Ulrich, Tissier-Desbordes, & Dubois, 2010; Azar, 2013), product gender perceptions (Fugate & Phillips, 2010), gender brand extensions (Lee, 2006; Ulrich, 2013), target gender transfer across products (Avery, 2012). Furthermore, there is an existing research on gender brand

(19)

3

personality in regard to the relation between masculine and feminine brand personality and brand equity (Lieven, Grohmann, Herrmann, Landwehr, & Tilburg, 2014), services brand gender (Ugolini, Cassia, & Vigolo, 2014).

As previously elaborated, brand personality traits and brand gender have been researched extensively. However, as of our knowledge, there is no evidence of a research on brand personality dimensions in a combination with, for example, brand demographics (e.g. age, gender), psychographics (e.g. attitudes towards self, others, etc.) or lifestyle (e.g. sporty or fashionable). It would be of interest to research the effect of brand personality traits and brand demographics on brand evaluations (e.g. brand attitudes, purchase intention), because it is important in branding context to find out whether and to what extent brand evaluations are influenced by brand human characteristics. For example, if Coca-Cola is perceived as a young female, how and whether these brand perceptions will influence brand evaluations. Coca-Cola is not a new brand, but its personality brings associations for youth and energy, because the ad campaigns clearly target the young generation. On the contrary, brands such as Old Spice might be perceived as an old male. Even though there are recent ad campaigns to appeal to younger men, the main target group consists of older men and brand evaluations among younger men might not be very favorable.

Problem definition

The thesis research question is: ’’How do brand human characteristics affect

brand evaluations?’’

Sub-questions

(20)

4

• How do brand personality dimensions (Sincerity, Competence, Excitement, Sophistication and Ruggedness) affect brand evaluations?

• How do they interact?

Delimitations of the study

The study will focus on the brand personality dimensions in combination with brand demographics and brand evaluations. It has been decided to not include psychographic and lifestyle factors in the study due to a possible expected weaker effect on brand evaluations compared to the expected effect of brand demographics.

Contribution

Theoretical contributions

The study will shed light on the effect of brand human characteristics on consumer brand evaluations in terms of attitude towards the brand and purchase intention. More specifically, it will be clarified which combinations of brand human characteristics will have more positive or negative impact on brand evaluations. In addition, possible interactions between any of the brand human characteristics will be explored.

Managerial contributions

By finding out the effect of brand personality traits and brand demographics on consumer brand evaluations, it would be possible for managers to create brand personalities for their brands based on consumers’ feedback rather than based on their own assumptions what would appeal to consumers. Managers would be able to understand how consumers perceive the personalities of the brands, what kind of brand attitudes they form and how likely consumers are to purchase the brands.

(21)

5

Outline

A literature review on the key concepts about brand personality, brand gender, brand anthropomorphism and brand evaluations is discussed in the next chapter, followed by a chapter discussing the conceptual model and hypotheses development. The thesis continues with a chapter discussing the research methodology, followed by a discussion chapter where study results, theoretical contributions, managerial implications and study limitations are discussed and further research is suggested. Finally, the study is completed with conclusion, references and Appendix.

2. LITERATURE REVIEW

2.1Brand personality

Brand personality is “a set of human characteristics associated with a brand’’ (Aaker, 1997, p.347). It can increase brand trust and elicit positive brand emotions which can lead to brand loyalty (Sung & Kim, 2010). Brand personality can be created and maintained through user imagery, pricing, advertisement, package design, symbols, public relations and celebrity endorsers (Aaker, 1997).

Brands can be considered responsible, active, aggressive, simple and emotional (Geuens, Weijters, & De Wulf, 2009). Fennis and Pruyn (2007) argue that brand personality traits influence the personality perceptions of the owner of the brand. This means that using a certain brand can influence how an individual is perceived by others. The authors indicate that the impact of the brand personality is stronger when both the brand and owner perceptions are congruent with the main brand associations. Thus, the transfer of brand personality traits to consumer personality is possible through impression creation.

(22)

6

Brand personality consists of five dimensions: Sincerity, Excitement, Competence Sophistication and Ruggedness (Aaker, 1997). Sincerity relates to sensible, honest, moral and joyful. Bold, enthusiastic, imaginative and modern characterize Excitement. Competence pertains to reliable, knowledgeable, intelligent, prosperous, Sophistication – to upscale, snobbish and charismatic. Finally, Ruggedness is characterized by sturdy and tough. Compared to the other brand personality dimensions, Excitement and Sophistication are most strongly related to perceived quality (Ramaseshan & Tsao, 2007). The authors investigate the moderating effect of brand concept on the correlation between brand personality and perceived quality and find that there is a positive relation between Excitement, Sophistication and perceived quality for brands with symbolic and experience-based functions. According to the scholars, Excitement is expressed through user imagery (Levi’s users are young) or packaging (Seiko has awesome package design) and Sophistication – through delightful design.

Sung and Kim (2010) find a link between brand personality dimensions, brand trust and brand affect. They argue that brand personality can increase brand trust and elicit brand affect. The scholars refer to brand affect as the possible evoked emotional response in the consumer as a result of using a brand. Thus, it is reasonable to make a distinction from brand attitude which is derived from brand beliefs and attitude towards the ad (Zhang & Zinkhan, 2006) and includes brand associations that comprise brand image (Keller, 1993). The study suggests that some brand personality dimensions impact brand trust more significantly, while other brand personality dimensions impact brand affect more strongly. Moreover, some brand personality dimensions influence consumers' functional and cognitive perceptions of a brand, while other brand personality dimensions affect consumers' feelings towards a brand. Brand trust is more likely to be influenced than brand affect when brands are perceived as having a sincere

(23)

7

brand personality. In contrast, exciting and sophisticated brand personalities influence brand affect more significantly than brand trust. Furthermore, rugged brand personalities demonstrate a significant influence on brand trust but not on brand affect, thus rugged brands may not evoke positive emotional response in consumers. Competent brand personalities influence brand trust and brand affect in a similar way. However, competent brand personalities influence brand affect more strongly than brand trust for apparel and perfume product categories.

Further in the context of brand personality dimensions, Klink and Althaide (2012) argue that one of many ways brand personality is created is with brand names through sound symbolism. For example, Ruggedness is created through brand names with back vowels, while Sincerity and Sophistication are created through brand names with front vowels.

Brand personality is stable over time and in relation with other brands and operates exactly as the human personality (Wee, 2004). By ‘’stable’’ the author means that brand personality is consistent over time as measured by Aaker (1997) brand personality framework.

Keller (1993) ascribes to brand personality functional, experiential and symbolic or self-expressive benefits. Functional benefits pertain to product or service consumption and to the product-related attributes, experiential benefits – to the feeling of using the product or service and the product-related attributes. Symbolic benefits refer to non-product-related attributes. By using visible products, consumers satisfy their need to express themselves and to be socially approved. Aaker (1997) suggests that fragrance, clothing and cosmetics serve symbolic functions and that toothpaste has utilitarian

(24)

8

functions, while computers and soft drinks are used in both functional and symbolic ways.

Through brands consumers express their identity and the way they want to be perceived by others (Graeff, 1998). The author argues that consumption situations and brand personality affect consumers’ brand evaluations. Furthermore, consumers use specific products to signal to others a suitable behavior and self-image. Thus, findings indicate that congruity between brand personality and self-image has a strong positive effect on attitude towards the brand and purchase intention. Similarly, Swaminathan, Stilley, & Ahluwalia (2009) argue that consumers’ self-image is important when searching for brands and brands’ personalities should be compatible with consumers’ ideal self-image and propose that the strength of consumers’ emotional attachment influences brand personality.

Similar to Swaminathan et al. (2009), Malär, Krohmer, Hoyer, & Nyffenegger (2011) suggest that the brand personality should match the consumer’s personality in order to achieve a strong emotional attachment with the brand. The study indicates that the emotional attachment with a brand is most strongly influenced by a an existing self- congruence rather than an ideal self-congruence with the brand personality which contradicts with Swaminathan et al. (2009) finding that brands’ personalities should be congruent with consumers’ ideal self-image.

Mulyanegara, Tsarenko, & Anderson (2009) discover that males and females prefer brands that are congruent with their personalities. The researchers measure how brand personality relates to consumer personality based on the personality dimensions of The Big Five model. The findings suggest that preferences for a trusted brand are driven by the personality dimensions for neuroticism and conscientiousness; extroversion and

(25)

9

being open to experience influence preferences towards an outgoing brand. The authors direct attention to the fact that there is a limited examination of the important effect of personality on preferences for particular brand personalities.

Brand personality can affect attitude towards the brand, preference for a brand and brand satisfaction (Ivens & Valta, 2012). The scholars investigate consumers’ perceptions of brands together with brand personality dimensions and the differences among these perceptions within and beyond consumer brands (e.g. Danone, Adidas, L’oreal, etc.). The study proves that brand personality has an impact on attitude towards the brand, brand preference and brand satisfaction. Furthermore, findings also show that brand personality perceptions differ in their magnitude and that there are distinct configurations of brand personality dimensions.

Brand personality perceptions can also be influenced by metaphoric ads and whether the product has functional or symbolic purposes (Ang, 2006). Brands that use metaphoric ads are seen as more sophisticated and exciting and less sincere and competent compared to brands that do not use metaphoric ads or illustrations. The author assumes that, in comparison to the rest of the brand personality dimensions, perceptions of competence are formed through multiple ad exposures. Furthermore, the researcher finds that metaphors in ads also impact attitude towards the ad, attitude towards the brand and purchase intention. Similar to brands, symbolic products are perceived as more sophisticated and exciting and less sincere and competent. Conversely, products of functional nature are perceived to be more sincere and competent but less sophisticated and exciting. However, when metaphors are used for utilitarian products, they are perceived as more sophisticated and exciting but less sincere. Thus, it can be inferred that metaphors could be used to highlight brand personality perceptions for utilitarian products.

(26)

10

An alternative view on brand personality is offered by Aaker and Fournier (1995): the brand as a relationship partner. The authors see the brand as a vigorous and collaborative partner in the existing relationship between the individual and the brand and consider its behaviors as an aggregate of the consumer’s perception of the personality of the brand. Further, the brand’s behaviors (e.g. change in the advertising campaign or reducing the package size) evoke consumer’s cognitive, attitudinal or behavioral responses. The scholars distinguish brand personality from user imagery in terms of public and private characteristics of brands. User imagery relates to the kind of person that consumes the product or the service (Keller, 1993). The authors argue that brand personality plays a more important role when consumers choose private brands compared to when they choose public brands and that user imagery is more important for the choice of public brands.

Based on their brand perceptions and associations, consumers consider brands as more attractive or less attractive. Brands with attractive personalities enhance self-perceptions in congruence with the brand personality and whether there is a personality transfer between brands and consumers after using the brands (Park & John, 2010). Thus, according to the authors, consumers who hold absolute beliefs about their personalities perceive themselves more positively on personality traits related to the brands they experience. That is, consumers' beliefs about their personalities play a significant role in the way consumers will react to using brands with attractive personalities. Furthermore, the impact of brands with attractive personalities on consumers' self-perceptions is not influenced by the length of the user experience with the brand – self-perceptions are unaffected if the brand experiences are not lasting.

Maehle, Otnes, & Supphellen (2011) examine how consumers’ perceptions of the various dimensions of brand personality in Aaker (1997) scale (Sincerity, Excitement,

(27)

11

Competence, Sophistication and Ruggedness) are formed and what type of products or brands affect these perceptions. There is a support for the idea that besides to brands, brand personality may be as well applied to product categories. For example, sincere brands are found to be related to high morals and to elicit associations in relation to family. Exciting brands evoke special and exciting feelings in consumers. Furthermore, competent brands are related to expertness and are perceived to be of high quality. Sophisticated brands are determined as feminine, while rugged brands are masculine. In addition, a link is found between consumers’ product performance expectations and perceptions of brand’s Sincerity and Competence. This relation is expressed through the user experience when the expected product performance is sufficient or even exceeds consumers' expectations. Lastly, the study findings indicate that company associations impact consumers’ perceptions of brand personality.

2.2 Brand anthropomorphism

Anthropomorphism is described by Epley, Waytz, & Cacioppo (2007) as ascribing a humanlike characteristics and traits to nonhuman object such as natural forces, mechanical or electronic devices, nonhuman animals and gods. The authors argue that anthropomorphizing nonhuman objects is more likely to occur when the anthropomorphism-related knowledge can be accessed and it is relevant, when people are aspired to be socially effective and when there is a missing sense of social relation to other individuals. This explains why anthropomorphism is common and pervasive. According to the authors, in order to understand the reasons behind anthropomorphizing nonhuman objects, it is necessary to take into consideration how individuals think about other individuals. Furthermore, anthropomorphism adds a moral aspect to nonhuman objects and thus suggests a respectful and careful treatment. Here the authors give an example with Mother Earth who should be protected from harm.

(28)

12

Brand anthropomorphism relates to perceiving a branded product as a human being and it occurs through two processes: perceived congruency between the appearance of branded products and human physical characteristics and to the degree to which consumers perceive such products as compatible with their actual, ideal or social self-concept (Guido & Peluso, 2015). Thus, the authors suggest that brand anthropomorphism consists of multiple dimensions. Moreover, they assume that brand anthropomorphism could be positively linked to brand loyalty, because consumers are more likely to repurchase branded products which are perceived as human beings. The study findings indicate that a unique brand personality can be created by employing design elements that resemble humanlike characteristics.

Despite their similarities, brand anthropomorphism and brand personality are distinct: brand anthropomorphism pertains to consumers’ perceptions of branded products as human beings, while brand personality relates to the consumers’ metaphorical perceptions of brands and products (Caprara, Barbaranelli, & Guido, 2001).

Congruence between self-concept and brand image is the driver of anthropomorphized brands, effectance motivation and sociality motivation are considered moderators (Puzakova, Kwak, & Rocereto, 2009). Thus, the authors argue that these findings serve as a sufficient prerequisite for developing significant relationships between consumers and brands.

Marketers purposefully ascribe human traits to products and thus humanizing of products suggests long-term business outcomes (Aggarwal & McGill, 2007). The scholars argue that the extent to which consumers perceive products as human is determined by whether there are humanlike product characteristics. Furthermore,

(29)

13

recognizing human traits in products affects consumers’ product evaluation. The authors indicate that products are less positively evaluated if they are presented as possessing humanlike characteristics but in fact they lack any human characteristics compared to products that are depictured as human and also have humanlike characteristics. However, more positive evaluations due to humanizing a product can be achieved only if the type of person that comes to mind is positively associated.

Aggarwal and McGill (2012) argue that anthropomorphized brands influence people’s strive for successful social interaction and this results in contrasting to the brand’s image behavior. The scholars investigate the effect of anthropomorphized brands on individuals’ behavior beyond the instantaneous brand consumption by presenting an imaginary brand personality, brand’s physical appearance, profession, etc. Furthermore, they suggest that the role consumers assign to a brand would affect their motivation to communicate with the anthropomorphized version of the same brand. The focus is on two brand roles: the role of a partner as a co-producer of the benefit and the role of a servant as an outsourced provider of the benefit. The role of brands as partners or servants is considered common in marketing by these scholars and thus they give an example with the financial organization Allianz which presents itself to the customers as a ‘’trusted partner’’. Other brands such as Scrubbing Bubbles from S.C Johnson and Co. are considered servants, because they work hard so that their customers don’t have to. The authors continue that an anthropomorphized brand influences the strive for successful social interaction. However, they point that this interaction is not related to a purchase or a consumption but rather to the indicated by the brand social interaction. The study findings show that, as a response to a brand prime, automatic behavior depends on whether it is anthropomorphized. Furthermore, the study indicates that the priming effects for anthropomorphized brands follow the motivation for interaction.

(30)

14

Brand anthropomorphism can carry negative consequences. Puzakova, Kwak, & Rocereto (2013) warn that humanizing brands can be problematic: consumers’ brand evaluations can be affected negatively if there is a negative publicity about the brand as a result of product misbehavior. As an appropriate illustration serves the suggested example of negative publicity about M&M’s in 2008 in regard of the poisonous substance melamine found in the candies. The authors’ findings indicate that consumers who believe in personality stability, so called entity theorists, respond less favorably to a negative publicity of product misbehaviors when the brand is anthropomorphized. In contrast, consumers who see personality traits as easily influenced, like the incremental theorists, do not evaluate a negative publicity about an anthropomorphized brand more unfavorably than nonhumanized brand. Thus, entity theorists accept product misdemeanor as more typical of a brand than incremental theorists. As of attitudinal and behavioral outcomes, entity theorists hold less favorable attitudes towards the anthropomorphized brand and have lower purchase intentions when a negative publicity about the brand is experienced. Furthermore, entity theorists react indifferently to an apology for product misbehavior when the brand is anthropomorphized. The negative effects of brand anthropomorphization on entity theorists diminishes only when the company provides a compensation for the product misbehavior. Lastly, Puzakova et al. (2013) argue that anthropomorphized brands are responsible for their negative brand performance because they are able to perform intentional actions.

2.3 Brand gender

As an important dimension of brand personality, brand gender has a significant impact on brand perceptions and associations. Brand gender is “the set of human personality traits associated with masculinity and femininity applicable and relevant to brands” (Grohmann, 2009, p.106).

(31)

15

Brands communicate their gender through elements such as brand name (Wu, Klink, & Guo, 2013; Lieven, Grohmann, Herrmann, Landwehr, & Tilburg, 2015), color and logo (Lieven, et al., 2015), and type fonts (Grohmann, 2014; Lieven, et al., 2015).

As a part of brand personality, masculinity and femininity haven’t been paid attention to in terms of how they are measured as well as they contribute to the branding theory (Grohmann, 2009). This scholar presents a way to measure brand personality dimensions: masculine brand personality (MBP) and feminine brand personality (FBP). The findings suggest that the favorability of consumer responses is increased by the congruence between brand personality and consumers' self-perception of own gender - masculinity and femininity. Further, androgynous brands, which are characterized as being both strongly masculine and feminine at the same time, evoke contradictory consumer expectations. It is found that a masculine extension category for a masculine brand results in greater brand fit and a feminine extension category increases the brand fit for a feminine brand.

Ulrich, Tissier-Desbordes, & Dubois (2010) classify brand gender as masculine, feminine, low masculine, low feminine, high masculine and high feminine. However, the study distinguishes brand gender from the masculine and feminine brand personality defined by Grohmann (2009). According to the scholars, consumers perceive brand gender and its dimensions depending on the gender of the brand user, the gender of brand personality traits, gender of the characteristics of brand communication, brand name gender in terms of grammar, gender of the characteristics of the logo and product gender.

Fugate and Phillips (2010) replicate and extend previous research on product gender perceptions and find that compared to women, men prefer to purchase products which

(32)

16

match their own gender. Individuals who focus on the congruence between product and self, use products as self-image, whereas individuals in households where the woman performs masculine chores and vice versa, do not emphasize on gender congruence. The study indicates that gender matching products might be less important for some individuals than another and that self-congruence with product personality suggests product gender congruence as well.

In congruence with Ulrich et al. (2010), Azar (2013) agrees that the perceptions of consumers of own gender impacts brand and product evaluations. The author measures brand masculinity and explores brand masculinity patterns (hegemonic, emerging, chivalrous and subaltern) and suggests that the gender of the parent brand is important in the evaluation of brand extensions across genders.

Unisex trend results in brand extensions across masculine and feminine products, thus based on consumers’ evaluations of gender extensions (extending the same brand in order to target the opposite gender), Lee (2006) attempts to identify conditions for successful gender brand extensions. It is found that the evaluation of gender extensions is impacted by the brand’s gender, consumers’ gender and the product type. Results show that a gender extension is accepted more positively if a masculine brand extends to target female consumers and that women accept gender extensions more easily than men. Ulrich (2013) agrees with Lee (2006) that extending masculine or feminine brands to the opposite gender (cross-gender brand extensions) is a growing marketing trend and notes that despite this fact, cross-gender brand extensions are not always successful. The study investigates how the consumers' complex gender and biological gender affect consumers’ evaluation of cross-gender brand extensions. Results indicate that consumers with traditional attitudes are less likely to accept cross-gender brand extensions

(33)

17

compared to consumers with more liberal attitudes and that biological gender, gender identity or sexual orientation have no significant impact on cross-gender extensions evaluations. Biological gender doesn’t significantly affect the acceptance of cross-gender brand extensions, thus this finding contradicts Lee's (2006) position.

A trend of gender transfer across brands (transfer of one gender targeting to another) shows that when brands of male gender target female consumers, male consumers react negatively and that men avoid products associated with femininity (Avery, 2012). However, men use shampoo, fragrances and cosmetics which are considered feminine product categories. These study results contradict Ulrich (2013) conclusion that biological gender doesn’t significantly affect the acceptance of cross-gender brand extensions and indirectly supports Lee (2006) that gender extensions are more positively accepted by women.

There is a limited empirical research on how brand personality perceptions relate to brand equity (Lieven, Grohmann, Herrmann, Landwehr, & Tilburg, 2014). These scholars investigate the relation between masculine and feminine brand personality and brand equity and find that strongly masculine and strongly feminine brands positively impact brand equity and this influence is not affected by the respondents’ gender. This study suggests that brands could be neither masculine nor feminine (brands that are not differentiated and are low in masculinity and femininity) and reveals that undifferentiated brands and androgynous brands are expected to be less attractive and thus associated negatively with brand equity. The finding in regard to androgynous brands supports Grohmann (2009) suggestion that androgynous brands can possess conflicting characteristics. Furthermore, masculine and feminine brands evoke more favorable associations compared to brands that are not differentiated.

(34)

18

There is no evidence of the impact of brand design on brand masculinity and femininity (Lieven, Grohmann, Herrmann, Landwehr, & Tilburg, 2015), thus this study examines the influence of brand design on brand masculinity and femininity perceptions as well as whether brand masculinity and femininity perceptions influenced by brand design are related to consumer preferences and brand equity. The authors elaborate on the influence of physical brand design characteristics on consumers’ perceptions of brand masculinity and femininity and find that brand masculinity and femininity increase brand preferences more specifically when the brand gender is congruent with the gender of the product category. Lieven et al. (2015) argue that this study is among the first to investigate design-related sources of brand gender perceptions. Results indicate that brand masculinity is enhanced by angular logo shapes and type fonts as well as back vowels, while brand femininity is influenced by round logo shapes and type fonts, front vowels. These findings support Wu, Klink, & Guo (2013) that brand names with back vowels are significantly associated with masculinity and brand names with front vowels are significantly associated with femininity. Brand names with front vowels are preferred if the brand targets women, whereas brand names with back vowels are preferred if the brand targets men (Wu et al. 2013). These authors find that congruence between brand gender target and brand name sound results in more favorable attitude-derived responses across products and that brand gender target and the interactions between these independent variables have no impact. Wu et al. (2013) consider this study to be the first to create a link between phonetic symbolism and gender brand personality.

Consumers rely on brand appearance such as logo or packaging to determine their perceptions of brand gender (Grohmann, 2014). The scholar argues that different type fonts influence consumers’ perceptions of brand gender such that script type fonts highlight brand femininity and display type fonts highlight brand masculinity. Moreover,

(35)

19

type fonts impact the possibility that a brand would be recommended by consumers to other consumers.

It is necessary to note that brand personality doesn’t apply to product brands only but to services brands as well. Ugolini, Cassia, & Vigolo (2014) examine whether gender dimensions of brand personality can make services brands within the same industry distinct and show that masculinity and femininity measured by Grohmann reveal dissimilarities in the positioning of services brands. According to the authors, this is the first study of brand personality gender dimensions in the service industry. The researchers suggest a further research in order to examine how consumers relate to the physical surroundings as well as to other consumers and this way specifically influence MBP and FBP. Moreover, they argue that it is necessary to set apart the effect of MBP and FBP related to the service industry or to a particular brand. The authors suggest a further research to investigate the influence of culture in perceiving MBP and FBP of global brands. The perceptions of MBP and FBP for services brands should be evaluated in terms of the respondents’ gender and the degree to which their gender and the brand gender are congruent as well as the extent to which respondents are involved with the brand.

Brand masculinity and femininity are brand gender dimensions which haven’t been explored despite their significance for the brand positioning of various product categories such as fragrances or personal care products (Lieven et al. 2015). The authors argue that modern marketers find new and different ways to use gender in targeting consumers, for example, generally perceived as feminine product categories such as diet soft drinks are extended to target male consumers.

(36)

20

Brand evaluations consist of consumer brand attitude and purchase intention (Graeff, 1998).

Brand attitude is defined as “a relatively enduring, unidimensional summary evaluation of the brand that presumably energizes behavior” (Spears & Singh, 2004, p.55) and originates from brand beliefs and attitude towards the ad (Zhang & Zinkhan, 2006). Brand attitude is a brand association that comprises brand image (Keller, 1993). The extent to which brand attitude is favorable is determined by consumers’ emotional attachment to the brand (Thomson, MacInnis, & Park, 2005). Brand attitude is also positively impacted by feelings elicited by ads (Pham, Geuens, & De Pelsmacker, 2013).

Purchase intention is the individual’s deliberate effort to make a brand purchase (Spears & Singh, 2004).

Congruence between brand personality and self-image has a significant positive effect on attitude towards the brand and purchase intention (Graeff, 1998). In his previous work, Graeff (1996) finds that when using promotional messages, brand image affects brand evaluations in terms of the extent to which consumers think about their self-image.

Brand attitude and purchase intention are positively influenced by ad format and information processing mechanisms so that the ad message becomes more convincing and brand evaluations become more favorable when ad format is congruent with a processing mechanism (Thompson & Hamilton, 2006). Attitude towards the brand and purchase intention are constructs within the attitude towards the ad framework: positive and negative feelings impact attitude towards the ad and brand attitude. Attitude towards the ad influences brand attitude; brand attitude impacts purchase intention (Spears & Singh, 2004). Additionally, brand attitude and purchase intention are affected by brand

(37)

21

familiarity and whether consumers are confident regarding the brand (Laroche, Kim, & Zhou, 1996). The authors argue that brand familiarity has an effect on consumers' confidence in terms of the brand which as a result influences purchase intention. Brand attitude is impacted by consumers’ familiarity with the brand.

Brand attitude is also impacted by brand personality congruence, brand user imagery congruence and brand usage imagery congruence (Liu, Mizerski, & Soh, 2012). Keller (1993) defines user imagery as the type of person who uses the brand and usage imagery – as the situations in which the brand is used. Liu et al. (2012) investigate the effect of brand personality congruence, brand user imagery congruence and brand usage imagery congruence on consumers’ brand attitude and loyalty in the context of luxury fashion brands. Results indicate a stronger influence of user and usage imagery on brand attitude and loyalty than brand personality congruence.

Purchase intention is positively affected by product-brand personality (Wang, Yang, & Liu, 2009). Additionally, the authors find that consumers' self-image congruence with the product- company-brand personality has a significant positive effect on purchase intention and serves as a moderator of the relationship between product-company-brand personality and purchase intention. Furthermore, purchase intention is significantly affected by brand awareness, brand loyalty and perceived quality (Chi, Yeh, & Yang, 2009) and by perceived value and store image (Grewal, Krishnan, Baker, & Borin, 1998). Moreover, Chi et al. (2009) find that perceived quality mediates the relationship between brand awareness and purchase intention, while brand loyalty is a mediator between brand awareness and purchase intention. Similar to Chi et al. (2009), Jalilvand, Samiei, & Mahdavinia (2011) find a significant positive effect of brand association, brand loyalty and perceived quality on purchase intention. Chi et al. (2009) and Jalilvand et al. (2011) argue that purchase intention is significantly influenced by perceived

(38)

22

quality. In this context, sophisticated and exciting brands are most related to perceived quality compared to sincere, competent and rugged brands (Ramaseshan & Tsao, 2007).

Morwitz, Steckel, & Gupta (2007) argue that purchase intention predicts sales. Findings indicate that there is a relation between purchase intention and existing products. Purchase intention is also correlated with purchases for durable goods and for short time periods. Moreover, purchase intention is correlated with purchases when respondents demonstrate intentions to purchase specific brands and when purchases are measured in terms of trial rates than when they are measured in terms of total market sales. Finally, correlation between purchase intention and purchases exists when a relative approach is used to gather purchase intentions.

As a consequence of rumors or a negative information about a brand, a brand crisis can affect brand evaluations negatively (Dawar & Lei, 2009). The study findings indicate that brand evaluations of consumers who are familiar with a brand are impacted by a relevant to the brand's benefit associations crisis rather by an irrelevant crisis. However, whether relevant or not, a brand crisis affects brand evaluations of consumers who are unfamiliar with a brand.

3. CONCEPTUAL MODEL AND HYPOTHESES

The conceptual model consists of the following key constructs presented in Figure 1. Figure 1 Conceptual model

(39)

23

The proposed conceptual model and its key constructs have been developed based on the literature review and the identified gap in the literature. It describes the variables and the relations and possible interactions between them.

The presented below hypotheses have been formulated based on the literature review and the identified gap in the literature as well as the proposed conceptual model in order to answer the research question.

3.1 Brand age

User imagery relates to the kind of user of the product or the service (Keller, 1993). In this context can be inferred that brand age is related to user imagery. Thus, it can be assumed that if a brand is perceived as young, then it will be positively related to attitude toward the brand and purchase intention.

Based on the proposed by Keller (1993) user imagery can be inferred that consumers will perceive brands as old depending on whether the brand users are old. Thus, old

(40)

24

brands are expected to be negatively related to attitude toward the brand and purchase intention.

According to Ivens and Valta (2012), brand personality impacts attitude towards the brand and perceptions of brand personality differ in their strength. Therefore, old brand age is expected to have a negative impact on attitude towards the brand and purchase intention.

Based on their brand perceptions and associations, consumers consider brands as more attractive or less attractive. Brands with attractive personalities enhance self-perceptions in congruence with the brand personality (Park & John, 2010). Therefore, it is assumed that brands perceived as old will be considered as less attractive and presumably these negative perceptions will also affect brand evaluations in a negative way.

H1: An old brand personality will result in a) more negative brand attitude and b)

lower purchase intention compared to a young brand personality.

3.2 Brand gender

Brand gender can be related to Keller (1993) user imagery. Therefore, it can be assumed that when a brand is perceived as masculine or feminine based on the users’ gender, it will be positively related to attitude towards the brand and purchase intention.

Grohmann (2009) defines brand personality as masculine and feminine and argues that consumer favorable responses are increased when brand personality matches consumer’s gender role identity. Consumer brand gender perceptions are derived from the gender of the brand user, gender brand personality traits, gender characteristics of brand communication, gender of the brand name in terms of grammar, gender

(41)

25

characteristics of the logo and product gender characteristics (Ulrich, Tissier-Desbordes, & Dubois, 2010).

Lieven, Grohmann, Herrmann, Landwehr, & Tilburg (2014) argue that feminine brands evoke more favorable associations compared to brands that are neither feminine nor masculine, thus it is expected that female brands will be positively related to brand evaluations.

Lieven et al. (2014) findings suggest that masculine brands evoke more favorable associations compared to brands that are neither masculine nor feminine, therefore it is assumed that male brands will affect brand evaluations positively.

H2: A high brand gender score will result in a) more positive brand attitude and b)

higher purchase intention compared to a low brand gender score.

3.3 Brand personality

Purchase intention is significantly impacted by perceived quality (Chi, Yeh & Yang, 2009; Jalilvand, Samiei, & Mahdavinia, 2011). In this context, sophisticated and exciting brands are most related to perceived quality compared to sincere, competent and rugged brands (Ramaseshan & Tsao, 2007). Therefore, sophisticated and exciting brands are most likely to have a strong positive impact on purchase intention because they are as well perceived to be of a higher quality compared to sincere, competent and rugged brands. From this can be inferred that brands with a defined brand personality will have a strong positive impact on purchase intention. Furthermore, from the abovementioned findings can be inferred that sophisticated and exciting brands will also likely affect brand attitude in a positive way, thus brands with a defined brand personality are as well likely to influence brand attitude in a positive way. Therefore, it can be assumed that brands with an undefined brand personality will influence purchase intention and brand

(42)

26

attitude less positively. Thus, the following hypotheses regarding sincere, competent, sophisticated, exciting and rugged brands are put forth:

H3: A high sincere brand personality score will have a stronger positive impact on

a) brand attitude and b) purchase intention compared to a low sincere brand personality

score.

H4: A high competent brand personality score will have a stronger positive impact

on a) brand attitude and b) purchase intention compared to a low competent brand personality.

H5: A high sophisticated brand personality will have a stronger positive impact on

a) brand attitude and b) purchase intention compared to a low sophisticated brand

personality score.

H6: A high exciting brand personality score will have a stronger positive impact on

a) brand attitude and b) purchase intention compared to a low exciting brand

personality score.

H7: A high rugged brand personality score will have a stronger positive impact on

a) brand attitude and b) purchase intention compared to a low rugged brand personality

score.

3.4 Interaction between brand age and brand personality

As previously mentioned, Park and John (2010) argue that brands can be viewed by consumers as attractive or less attractive, depending on the brand perceptions and brand associations they hold. In this context it was suggested that an old brand personality will lead to a lower purchase intention and a more negative brand attitude than a young brand personality – H1. Purchase intention is significantly impacted by perceived quality (Chi,

(43)

27

Yeh & Yang, 2009; Jalilvand, Samiei, & Mahdavinia, 2011) and according to Ramaseshan and Tsao (2007), brand personality dimensions influence brand quality perceptions: sophisticated and exciting brands are most related to perceived quality compared to sincere, competent and rugged brands. In this context it can be assumed that there might be an interaction between brand age and brand personality depending on whether the brand personality is perceived as old or young and whether the brand personality is strong (defined) or weak (undefined). This leads to the development of the following hypothesis regarding the possible interaction between brand age and brand personality:

H8: An old brand personality will result in a) more negative brand attitude and b)

lower purchase intention compared to a young brand personality when brand personality is strong/defined than when brand personality is weak/undefined.

3.5 Interaction between brand gender and brand personality

Regarding brand gender it was hypothesized that a high brand gender score will lead to a higher purchase intention and to a more positive brand attitude compared to a low brand gender score - H2.This assumption is based on the Lieven et al. (2014) findings that both masculine and feminine brands elicit more favorable associations compared to brands with an undefined gender. In terms of brand personality it was mentioned that Ramaseshan and Tsao (2007) consider sophisticated and exciting brands to be perceived to be of a higher quality compared to sincere, competent and rugged brands and that purchase intention is affected by perceived quality (Chi, Yeh & Yang, 2009; Jalilvand, Samiei, & Mahdavinia, 2011). It can be assumed in the context of brand gender and brand personality that there might be an interaction between these two independent variables. More specifically, the interaction between brand gender and brand personality

(44)

28

will depend on whether the brand gender is defined or undefined and whether the brand personality is strong (defined) or weak (undefined). Thus, the following hypothesis regarding the possible interaction between brand gender and brand personality is put forth:

H9: A high brand gender score will result in a) more positive brand attitude and b)

higher purchase intention compared to a lower brand gender score when brand personality is strong/defined than when brand personality is weak/undefined.

4. METHODOLOGY

4.1 Research design

In order to test the hypotheses and answer the research question, it was necessary to conduct an experiment. The research design used for the experiment was 5x2x2 factorial design. The experimental design included the following independent variables: brand personality (sincere, competent, sophisticated, exciting and rugged), brand age (old, young) and brand gender (male, female).

The brand personality dimensions Sincerity, Competence, Sophistication, Excitement and Ruggedness were rated on a 7-point Likert scale (1-strongly disagree; 7-strongly agree). Respondents were asked to indicate to what extent they disagreed or agreed with the following statements: ''This brand could be described as a sincere person ''; ''This brand could be described as a competent person ''; ''This brand could be described as a sophisticated person''; ''This brand could be described as an exciting person ''; ''This brand could be described as a rugged person ''.

(45)

29

Brand age was also rated on a 7-point Likert scale (1-strongly disagree; 7-strongly agree) and was measured through the statements ''This brand would be a young person'' and ''This brand would be an old person''.

Finally, similar to brand personality dimensions and brand age, brand gender was rated on a 7-point Likert scale (1-strongly disagree; 7-strongly agree). Here the respondents were asked to indicate to what extent they disagreed or agreed with the statements ''If this brand was a person, it would be male'' and ''If this brand was a person, it would be female''.

The dependent variables were presented by brand attitude and purchase intention (collectively brand evaluations). Brand attitude and purchase intention were measured on a 7-point Likert scale (1- strongly disagree; 7- strongly agree) and respondents were asked to indicate to what extent they disagreed or agreed with the statements ''I like the brand'' for brand attitude and “If I were to buy [product], I would consider this brand” (where [product] denotes a product relevant to the brand, for example Harley Davidson used the product “motorcycle” for purchase intention.

In addition to the independent and dependent variables, four possibly affecting the results control variables were included. These control variables were brand familiarity (respondents were asked to indicate to what extent they disagreed or agreed with the statement ''I am familiar with the brand'' on a 7-point Likert scale where 1 was strongly disagree and 7- strongly agree), brand (the goal here was to remove any systematic differences that arise from, e.g. price, product category, etc. and are correlated to each brand), respondents' age and gender.