Strength of the gender image of the mother brand : the effect in a cross-gender brand extension context

(1)

1

Master Thesis Marketing

“Strength of the gender image of the mother brand: the

effect in a cross-gender brand extension context”.

Author: Giulietta Fantacci Student number: 10826742

Supervisor: Dr. Karin Venetis University of Amsterdam (UvA)

Faculty of Economics and Business (FEB) MSc Business Administration – Marketing Track August 2015

(2)

2

Acknowledgements

Hereby I would like to thank my thesis supervisor Dr. Karin Venetis for her help and patience during all the different steps related to the thesis process.

Furthermore, I would like to thank all the respondents for their cooperation and willingness to participate in this study.

Finally, I would like to thank my family and my friends. Their role was fundamental during the thesis period, thanks to their moral support and encouragement I have always felt motivated by them. Furthermore, they both helped me to gather sufficient data, thus, I will always be grateful to all of them.

Kind regards, Giulietta Fantacci

Statement of originality

This document is written by Student Giulietta Fantacci 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)

3

Abstract

The subject of this master thesis is the effect of the strength of the gender image of the mother brand on consumers‟ evaluation of cross-gender brand extensions and on the evaluation of the parent brand post-extension launch.

The evaluation of cross-gender brand extensions does not show any differences based on the perceived strength of the initial gender image of the mother brand regardless of consumers‟ gender. Indeed, this research shows that when looking at the outcome gender plays an important moderating role. Furthermore, gender is not the only variable that influences the evaluation of the extension. According to the findings, brand image fit is a mediator between the extension from a strong or weak gendered brand and the evaluation of it. Overall, cross-gender brand extensions made from a weak gendered brand are more accepted and better evaluated by women compared to men. However, when the extension is introduced from a strong gendered brand, men and women do not show meaningful differences in the evaluation and acceptance of the extension.

This research shows that in terms of evaluation of the mother brand post-extension launch there are no differences between weak and strong gendered brands. Furthermore, gender does not play a moderating role between the introduction of the extension and the post-extension evaluation of it. Overall, when evaluating the mother brand post-extension launch no differences are shown between genders or the strength of the image of the parent brand. Indeed, the extended brands result to be similarly affected by the extension in terms of attitudes towards it and brand equity, leading to possible dilution effects.

(4)

4

3.3 Conceptual framework………..……..26 Chapter 4: Methodology………...………28 4.1 Pre-test……….……….……...28 4.2 Experimental design………..………...……….33 Chapter 5: Results……….……….38 5.1 Sample overview……….……….38 5.2 Condition similarity……….……….….42 5.3 Normality check……….. 43 5.4 Reliability check………...……….…...44 5.5 Manipulation check……….…...44 5.6 Hypotheses testing………..………47

(5)

5

Chapter 6: Discussion………..65

6.1 Interpretation of the results………..………65

6.2 Theoretical implications………...………..68

6.3 Managerial implications………...…..70

Chapter 7: Conclusion………..…………..71

7.1 Summary………...……….71

7.2 Limitations and suggestions for further research………74

Reference List……….…………76 Appendix I: Pre-test questionnaire

Appendix II: Main survey questionnaire Appendix III: SPSS Output group differences Appendix IV: Descriptive Statistics

(6)

6

Chapter 1: Introduction

1.1 Background

The impact of a brand extension on the master brand is often overlooked but can be strategically critical to an organization. Companies can benefit from brand extension strategies; however, this practice can also be harmful for them. Indeed, while using a strong brand name can minimize the risk of introducing a product in a new market by providing consumers the familiarity of and knowledge about an established brand, the inaccurate extension could create damaging associations that may be costly and difficult to mutate (Aaker and Keller 1990). For instance, in the 1990s Harley Davidson launched new products such as aftershave and perfumes. The company extended the brand too far and even the loyal fans did not like the idea, as it did not resonate with the tough brand identity. Thus, companies need gaining insights into consumers' evaluations in order to understand the success of a brand extension. Even though the concept of brand extension has been widely studied, research on this topic has poorly taken into account the fact that brands possess different personalities which can affect consumers' evaluations of both the mother brand and the extension.

Brand personality is not only associated to personality characteristics, indeed, researchers include also demographic characteristics, such as gender, age and class which are inferred directly from the brand's user imagery, employees, or product endorsers and indirectly from other brand associations (Aaker 1997; Grohmann 2009). Focusing on gender characteristics, Grohmann (2009) has developed a scale that measures gender dimensions of brand personality, underlying their importance when brands have symbolic value for consumers striving to strengthen their own masculinity or femininity. In practice, marketers support consumers‟ necessity for self-expression by creating masculine or feminine brand associations, for example using different colors, fragrances and design (Grohmann 2009), targeting one biological gender only, and consequently acquiring a gendered image (Ulrich 2013). Gendered brands can have the advantage of leveraging on their

(7)

7 masculine or feminine associations in order to attract males or females‟ consumers. However, brands which try to extend beyond their traditional market segment can face downsides due to the strong association with a particular gender. As a consequence, companies may choose to target the opposite gender segment by using different brand names (Jung and Lee 2006). Nevertheless, all the aspects of brand building involve significant costs thus, the creation and support of a new brand is expensive and difficult (Aaker and Joachimsthaler 2000). As a result, organizations may consider an extension strategy in order to overcome the cost disadvantage in developing a new brand. However, gendered brands extending to target the opposite biological sex may have problems in developing strong positions, making it difficult to evaluate whether this strategy is a successful one. For instance, Dior scores less well as “Dior Homme” compared to the feminine skincare market and the feminine underwear brand Aubade failed with its male extension “Aubade for Men” after two seasons (Ulrich 2013).

Extending the same brand name to target the opposite sex requires for a major understanding of the conditions for consumer acceptance and evaluation of the extensions (Ulrich 2013). Findings from previous researches on cross-gender brand extensions are inconsistent. For example, Jung and Lee (2006) found that women are significantly more receptive towards cross-gender brand extensions than men. On the contrary, according to Ulrich (2013) biological sex has no significant impact on the acceptance of cross-gender brand extensions, indeed, men result as receptive as women. Therefore, further research is needed to gain more insights on the topic and to overcome some of the limitations of previous studies (Jung and Lee 2006; Ulrich 2013). For instance, additional understanding is necessary to investigate if the acceptance of cross-gender brand extensions is influenced by the strength of the gender image of the parent brand. In other words, since the initial gender image of the mother brand can be perceived to be strongly/weakly masculine or strongly/weakly feminine, this might have an impact on consumers' evaluation of the extension, perceived fit and evaluation of the parent brand.

(8)

8 So far, research on cross-gender brand extensions has been scarce. Past studies on this topic focused more on consumers' evaluations of the extension and on attitudes towards the mother brand post-extension launch. As a consequence, research has not investigated yet the impact that this strategy can have on the equity of the parent brand. Thus, beyond consumers' acceptance and evaluation of cross-gender brand extensions, further research is necessary in order to examine to what extend the brand equity of the parent brand can be affected by extensions which are inconsistent with the brand gender of the mother brand (Lieven et al. 2014).

1.2 Research goal

Cross-gender brand extensions have become increasingly more popular among companies (e.g. Triumph swimsuits, Chanel perfumes). Consequently, a major understanding of the effects that this strategy will cause on several dimensions can help organizations to more effectively manage the extensions. While previous research has determined the importance of consumers‟ characteristics in the acceptance of cross-gender brand extensions, the subject of this study is the effect that the strength of the initial gender image of the mother brand has on consumers' evaluation of the extension. In other words, the study will contribute to the existing literature (Ulrich 2013; Jung and Lee 2006) by focusing on how the perceived strength of the initial gender image of the parent brand influences consumers in the acceptance and evaluation of a cross-gender brand extension. Finally, studies have not been conducted yet with respect to cross-gender brand extensions and the impact that they have on the equity of the parent brand. Therefore, this research will address one of the gaps underlined by Lieven et al. (2014) and will investigate the impact of a cross-gender brand extension on the evaluation of the mother brand by including measures of brand equity for both strong and weak gendered identities.

(9)

9

1.3 Research question and sub questions

This study will attempt to answer the following research question:

“How does the perceived strength of the initial gender image of the mother brand influence consumers in the acceptance and evaluation of a cross-gender brand extension? And what is the impact of a cross-gender brand extension on the evaluation of strong and weak mother brands

post-extension launch?”

In order to answer this research question, the literature review will explain the main concepts related to the topic and will answer the following sub questions:

 Does a brand have gender? How can the gender of a brand be determined?

 What are brand extensions? Are there any differences between men and women in the perception of fit in a brand extension context?

 Are there any differences between men and women in the evaluation of a cross-gender brand extension in terms of attitudes and purchase intent?

 What is brand equity? Why is it important to study the effect of a brand extension on the equity of the mother brand?

1.4 Contribution

Theoretical

Research literature concerning cross-gender brand extensions has been scarce, probably because the phenomenon seems quite new. While previous studies have focused mainly on consumers‟ characteristics as a factor influencing the evaluation and acceptance of cross-gender brand extensions, this research will look at the outcome by analyzing the issue from a brand image perspective. Thus, this study will contribute to the existing literature (Ulrich 2013; Jung and Lee 2006) by looking at the topic from a different perspective, that is, the evaluation and acceptance of a

(10)

10 cross-gender brand extension may be influenced by the strength of the gender identity of the mother brand and not only by consumers‟ characteristics. Furthermore, this research will shed light on the inconsistent results of past studies regarding the evaluation of cross-gender brand extensions by men and women. Finally, studies have not been conducted yet with respect to cross-gender brand extensions and the impact that they have on the evaluation of the mother brand post-extension launch in terms of brand equity. Therefore, this research will provide a contribution to the existing literature addressing one of the gaps underlined by Lieven et al. (2014) regarding how brand equity is affected by brand extensions that are inconsistent with the brand gender of the parent brand.

Managerial

In modern society traditional gender roles are evolving over time, as a consequence, a single gender brand positioning strategy may be a limit for a company. Therefore, more and more organizations tend to introduce brand extensions in order to target the two biological sexes at once, using a dual-gender positioning strategy. However, managers should be cautious when executing extensions. Indeed, several brands that have been successful with their initial target have experienced obstacles in developing strong positions with their cross-gender extensions. Thus, this study will contribute to the practice by helping marketers to understand whether a cross-gender brand extension can be successful based on the strength of the parent brand's gender image and if the overall evaluation will differ depending on the biological sex of the targeted consumer. Furthermore, managers will gain more insights regarding the potential harmful effects of a cross-gender brand extension, based on the brand gender identity, in terms of both attitudes towards the mother brand and equity. In this regard, companies can have a better understanding of the consequences that this strategy can have not only on the extension itself, but also on the evaluation of the mother brand depending on how strong or weak is the gender identity of it.

(11)

11

1.5 Thesis outline

The study will start by describing the existing literature concerning product and brand gendering, brand extensions more in general, evaluations of cross-gender brand extensions, and the importance of brand equity in a brand extension context. Therefore, Chapter 2 will explain what is already known about this topic and where some gaps are present in the literature. In Chapter 3 a conceptual framework will be presented and the hypotheses will be developed and discussed. Chapter 4 will describe the methodology of the study. In detail the pre-test, the research design, the sample and measurement will be discussed. The analysis and the results will be presented in Chapter 5. After analyzing the data, the discussion and conclusions will be drawn upon these results, including the limitations of the study, managerial and academic implications and recommendations for future research will be given. The pre-test, the main questionnaire and the SPSS output will be presented in the appendix section.

Chapter 2: Literature review

2.1 Gendered brands and products

“Does a product, like a person, have gender?” is an interesting question that researchers have studied (Iyer and Debevec 1986; Milner and Fodness 1996). Previous findings demonstrate that most products have gender and sex-typed identities as masculine or feminine. Assigning a gender to a product or brand means imbuing it with a masculine or a feminine image and identity (Alreck 1994). Gendered products' characteristics are usually adapted in order to include symbols in design features, advertising, promotion and distribution processes which permit the identification with them by one sex (Alreck 1994). Thus, gendering a product or brand is realized by operating in two ways: first the product or brand is designed to attract the stereotypical man or woman; second, to give it a gendered image, it has to be highly associated with the masculine or feminine sex role through marketing communication tools (Alreck 1994). In order to support gender image integrity,

(12)

12 consumers used to purchase products that have gender identities congruent with their own, therefore, consumers' behavior is considered to be consistent with a person's gender identity (Fugate and Phillips 2010). Indeed, according to Martin and Bellizzi (1982) the product with the image that is most similar to the consumer's self-image will be the one purchased or desired, whereas products with different images will be rejected.

The desire for self-congruence with products has been studied at both a brand and a product category level (Cowart et al. 2008). For example, according to Morris and Cundiff (1971) there is an interaction of sex role identification and anxiety in response to males' attitudes towards hair spray that is considered to be a feminine product category. The desire for self-congruence at a brand level can be related to the fact that brands possess different personalities and consumers may express their own personality through brand use (Aaker 1997). Brand personality is defined as "the set of human characteristics associated with the typical user of a brand" (Aaker 1997, p. 347). Brand gender, defined as “the set of human personality traits associated with masculinity and femininity applicable and relevant to brands” (Grohmann 2009, p. 106) is a dimension of brand personality. Aaker (1997) uses as an example Virginia Slims which tends to be thought of as feminine and Marlboro which tends to be perceived as masculine, according to distinct user imagery.

Masculinity and femininity are relevant traits of human personality; therefore, it is likely that consumers relate them to brands (Grohmann 2009). Previous research has developed scales measuring masculinity and femininity in human personality traits context, however, the measurement of human characteristics might be not suitable in a brand personality context. As a consequence, Grohmann (2009) developed a scale that measures the gender aspects associated with brands. The dimensions are captured by a two-dimensional, 12-item descriptive adjective scale (masculine brand personality – MBP and feminine brand personality - FBP) that is applicable to utilitarian, symbolic and mixed brands. The adjectives related to MBP are: adventurous, aggressive, brave, daring, dominant and sturdy; whereas the ones related to FBP are: expresses tender feelings,

(13)

13 fragile, graceful, sensitive, sweet and tender. These dimensions can be used in conjunction with Aaker‟s (1997) “Big Five” dimensions of brand personality for the categorization of brands: sincerity, excitement, competence, sophistication, and ruggedness (Grohmann 2009).

Defining brands in terms of their gender can be a meaningful tool to examine consumer perceptions of competing brands and to find different opportunities for positioning (Grohmann 2009). From previous studies, it resulted that brand personality contributes to create brand loyalty through the development of brands‟ symbolic meanings, which increase consumers‟ emotional connection to the brands (Fournier 1998). This process has an important role in symbolic, emotional or experiential consumption experiences (Holbrook and Hirschman 1982). As a result, brand personality is a salient part of consumer-based brand equity (Keller 1993). Indeed, gender dimensions of brand personality influence affective, attitudinal, and behavioral consumer responses in a positive way when they are congruent with consumers‟ sex role identity enabling consumers to manifest a fundamental aspect of their self-concept (Grohmann 2009). In fact, according to Grohmann (2009), compared to brand personality-self-concept incongruence, brand personality-self-concept congruence drives more positive consumer responses in terms of favorable brand attitude and purchase intentions, brand affect and trust, attitudinal and behavioral loyalty, and positive word-of-mouth. Finally, brand personality plays an important role in a brand extension context, especially regarding fit issues, that are analyzed in the following section.

2.2 Brand extensions

Brand extension is the „„use of established brand names to enter new product categories or classes‟‟ (Aaker and Keller 1992, p. 35). Brand extensions are attractive to companies because they allow them to take advantage of brand name recognition and image to enter new markets (Aaker and Keller 1990). When the new extension is introduced, consumers judge it on the basis of their attitude towards the parent brand and the extension category. If the consumer does not know the parent brand and its products, he or she will evaluate the new extension by looking at the experience

(14)

14 with the extension category that he or she had (Czellar 2003). On the contrary, if the consumer does not know the extension product category, the attitude towards the extension will be established on the basis of his or her attitude towards the parent brand (Czellar 2003). A clear consumer's knowledge of both the parent brand and the extension category will bring to a perception of fit (or perceived similarity) between the two elements. The perception of fit can have two effects on consumers' attitudes: it can mediate the transfer of attitudes from the parent brand and extension category to the new extension; or it can work as a moderator among these dimensions (Czellar 2003). As a result, consumers' attitudes towards brand extensions impact their behavior in terms of purchase intention, choice and repeat purchase (Czellar 2003).

Perceived fit is determined by the number of shared associations between the extension product category and the brand (Czellar 2003). Furthermore, it is a function of both product feature similarity perceptions – product category fit – and brand concept consistency perceptions – brand image fit – (Park et al. 1991; Czellar 2003). While product category fit relies on the perceived similarity between the extension category and the existing product categories of the parent brand, brand image fit refers to the match between the specific image of the brand and the extension product category. Researchers have widely investigated about the role of perceived similarity in brand extensions. For example, according to Barone et al. (2000) positive consumer mood increases fit perceptions for moderately far extensions, and advertising can be used in order to improve consumers‟ fit perceptions by manipulating the content of ads (Kim et al. 2001). Park et al. (1991) have found that for both function-oriented and prestige-oriented brands, positive responses occur when brand extensions are made with high brand concept consistency and strong product feature similarity.

According to previous studies the higher the perceived fit, the more favorable is the consumer's attitude towards the extension. Therefore, fit is an important element in brand extension evaluations and consumers should accept the new product as logical in order to evaluate it in a positive way.

(15)

15 However, beyond similarity perceptions there can be other factors affecting consumers' evaluations of the extension. These factors can be related to the characteristics of the parent brand, the brand extension or the consumer. For example, perceived high quality of the parent brand (Aaker and Keller 1990) and the extent to which the extension is launched as part of a sequence of launches (Aaker and Keller 1992) affect the evaluation of the extension. Furthermore, looking at consumers' characteristics, consumer expertise is shown to moderate the impact of product-related brand associations, brand affect and fit on brand extension attitude (Broniarczyk and Alba 1994). In conclusion, consumers can form either positive or negative attitudes towards the extension that may translate into favorable/unfavorable associations or an affective form of liking/disliking the extension. As a result, the attitude towards the extension may change the consumer‟s initial attitude towards the parent brand or the extension category (Czellar 2003).

However, while previous research has widely studied parent brand characteristics such as quality and associations, and consumers' characteristics such as expertise, knowledge, and mood, it has been paid little attention to the possible impact of characteristics such as the gender of brands and consumers in a brand extension context. Since the focus of this study is on cross-gender brand extensions, in this case product feature similarity is not significantly relevant because the extension is in the same product category and has the same features. Therefore, the most crucial characteristic in this context is a brand‟s perceived masculinity or femininity (Jung and Lee 2006). As a consequence, we talk more precisely about cross-gender line extensions. Thus, in this context only the brand related image fit dimension is examined. For instance, when Triumph extends into swimwear to target males, the important aspect to take into account is whether men will accept the feminine image of this brand on themselves (Jung and Lee 2006). According to Martinez Salinas and Pina Pérez (2008), fit is an important factor to be considered in a brand extension context because the most fundamental aspect for the success of an extension is coherence with the image of the extended brand. Indeed, their study shows that even if the new product does not belong to the

(16)

16 same category, the company has to succeed in the transfer of the brand essence from the parent brand to the extension in order to achieve a positive consumers' evaluation.

2.3 Gender of consumers and evaluation of cross-gender brand extensions

Traditionally our consumption has been gendered, and consumers have used gendered products and brands in order to enhance their identities (Avery 2012). Product gendering is still playing a central role and even shifting to the direction of more strongly identifying products as gendered rather than undifferentiated or asexual (Fugate and Phillips 2010). As a consequence, companies may find it hard to attract the other sex to brands associated with one gender only (Avery 2012). For instance, the attempts of Pepsi and Coke to sell Diet Pepsi and Diet Coke to men failed, thus, they decided to introduce new brands to attract males to diet soda. Consequently, the companies launched Pepsi Max and Coke Zero trying to dissociate the new brands from feminine diet sodas by claiming that they are the “diet cola for men” and “it's not for women” (Avery 2012).

Gendered brands restrict what we buy due to the fact that what is feminine is seen to be the antithesis of what is masculine (Bem 1993) and through consumption we create, enhance and accomplish our gender identities. Despite the fact that gender roles are becoming more blurred (Stokburger-Sauer and Teichmann 2013) research shows that in consumer behavior gender differences are still present, especially regarding the importance of congruence between a product or brand and one's self. While in some previous studies both men and women resulted to be psychologically uncomfortable using products or brands which do not seem to be made for them, according to Fugate and Phillips (2010) women exhibit a lower need for gender congruity in consumption decisions compared to men. Indeed, men are more likely to look for gender congruence with product categories than females, and they search for more cues in the marketing mix such as packaging, colors, product shape, logos and brand names (Fugate and Phillips 2010). Morris and Cundliff (1971) found that men, compared to women, are more sensitive towards the acceptance of female products or brands, whereas women are more likely to buy masculine brands

(17)

17 than men do for the feminine ones (Fry 1971; Vitz and Johnson 1965; Worth, Smith and Mackie 1992). These findings are in line with Avery‟s (2012) research, in which the Porsche Cayenne SUV launch was examined to investigate how men consuming a gendered brand respond to perceived brand gender contamination (Avery 2012). First of all, the results show that gender still plays a meaningful role in consumers' choice and usage of brands (Avery 2012). Moreover, men who have engaged with a brand will fight for its meaning, trying to maintain a dominant position in the social hierarchy when it is contaminated by women. In other words, the Porsche man uses defensive practices inside his brand community in order to defend his masculine identity sign, avoiding associations with women and fighting against the feminization of his brand (Avery 2012).

In the last years, many gendered brands have tried to expand in order to target the opposite sex with the potential of extending their business and, at the same time, limiting launch costs. This strategy has been defined as cross-gender brand extension, that is, the extension of the same brand name to target the opposite sex. Cross-gender brand extensions by masculine brands for women are not so new (e.g. Gillette razors, Adidas sportswear), whereas, extending feminine brands to target men is more recent (e.g. Triumph swimsuits, Nivea skincare). In fact, there are still many feminine brands which have not targeted men yet (e.g. Victoria‟s Secret underwear, Chanel high-end fashion, Pandora jewelry) whereas few masculine brands have maintained their unique masculinity (e.g. Ermenegildo Zegna high-end fashion, Old Spice grooming products). Despite that, research has poorly investigated on cross-gender brand extensions, therefore, this strategy requires a greater understanding of the conditions for consumer acceptance of the extensions and the impact on the parent brand.

Jung and Lee (2006) conducted a study with the main goal of defining under what conditions a cross-gender brand extension can be successful. For this purpose, the cross-gender extension of each brand was manipulated introducing an extension to target the opposite gender group. Four dependent variables were measured: perception of overall fit, perception of brand image fit, attitude

(18)

18 towards the extension, and attitude towards the original brand (Jung and Lee 2006). Based on the results, the four dependent variables were found higher when the extension was made from the masculine brand to target female consumers than vice versa, in other words, women are more receptive towards and have a higher acceptance of cross-gender extensions than men. However, Jung and Lee's (2006) research has some limitations such as the use of only two product categories (high-end fashion and hair gel/spray) with low masculine/feminine image, resulting in a difficult generalization of the results. Furthermore, their study was conducted in Asian countries using university students. As a consequence, it is not possible to generalize the results among western cultures and also a wider sample (i.e. various ages and social classes) is necessary to well represent the population.

In this regard, Ulrich's (2013) research was designed to go beyond the biological sex effect shown by Jung and Lee (2006) and to overcome some of the above mentioned limitations. The main goal was to investigate the impact of multifactorial consumer gender on cross-gender brand extensions. An experimental study was conducted in France (representative of western cultures) using fictitious cross-gender extensions from existing brands. Gendered brands were chosen among three product categories: face skincare (feminine image, used for testing an extension to the masculine), shampoo (low masculine/feminine image due to usage by both sexes, used for testing an extension to the masculine) and soda drinks (low masculine/feminine image, used for testing an extension to the feminine). According to her findings and contrary to Jung and Lee (2006), biological sex has no significant impact on the acceptance of cross-gender brand extensions, indeed, men result as receptive as women (Ulrich 2013). As a consequence, biological sex has also no effect on the evaluation of the parent brand post-extension. However, gender role attitudes have a significant impact on the acceptance; indeed, consumers with traditional gender attitudes are less favorable to cross-gender extensions than egalitarian consumers, with considerably lower attitude and purchase intent towards the extensions, and lower perception of fit. Moreover, gender attitudes affect the

(19)

19 evaluation of parent brands post-extension, showing a lower attitude for traditional individuals comparing to the egalitarian ones (Ulrich 2013). Thus, the research shows that independently of consumer biological sex, differences exist between traditional and egalitarian consumers in evaluating the extensions (Ulrich 2013). Furthermore, evaluations of brand extensions are not affected by sexual orientation or consumer age, solving one of the possible limitation of Jung and Lee's (2006) study in which age was considered to be a possible element affecting the evaluation of extensions across gender (Ulrich 2013). In conclusion, while cross-gender brand extensions can be accepted likewise by men and women, consumers with traditional gender attitudes (independently of their age and biological sex) will exhibit more hesitation towards the extensions. However, Ulrich's (2013) research has some limitations such as the examination of three product categories with strong feminine image, and low femininity/masculinity, therefore, it is necessary to conduct a study using also a product category with a strong masculine image in order to generalize the results. Furthermore, the findings are restricted to cultures with a moderate degree of masculinity/femininity like France; so, other countries should be included for a generalization of the results. Finally, further research is necessary to understand if the reluctance in accepting cross-gender brand extensions of consumers with traditional cross-gender attitudes can increase with the strength of the gender of the parent brand.

2.4 Brand equity in a brand extension context

Consumers used to associate specific sets of attributes with a brand, creating a particular image of it. When a brand extension is introduced, it will have its own set of attributes, which can be either consistent or inconsistent with the mother brand. As a consequence, a brand extension can influence not only consumers‟ attitude towards the extension but also how they evaluate the parent brand (Loken and Roedder John 1993). After the introduction of a brand extension, a modification of consumers‟ current beliefs about the mother brand can occur, leading to a re-evaluation of it. The overall impact may go beyond consumers‟ attitudes towards the mother brand and it can also affect

(20)

20 its brand equity.

Keller (1993, p.1) defines brand equity in terms of “the marketing effects uniquely attributable to the brand, for example, when certain outcomes result from the marketing of a product or service because of its brand name that would not occur if the same product or service did not have that name”. The study of brand equity can be related to two main reasons: one is for financially purposes to assess the value of a brand; the other is for strategic motivations to improve marketing productivity (Keller 1993). Since the brand is one of the most important assets of a company for the improvement of marketing productivity, it is necessary to understand the knowledge that has been generated about the brand in consumers' minds from past marketing investments (Keller 1993). Therefore, the fundamental basis of brand equity is consumer memory (Pitta and Katsanis 1995).

Keller (1993) describes brand equity in terms of components of brand knowledge, introducing the concept of customer-based brand equity, that is, “the differential effect of brand knowledge on consumer response to the marketing of the brand” (Keller 1993, p.2). Customer-based brand equity takes place when the consumer is familiar with the brand and recalls some favorable, strong and unique brand associations in memory (Keller 1993). Focusing on brand knowledge, Keller (1993) defines two important components of it: brand awareness and brand image. Brand awareness is the consumers' ability to identify how well brand identities do their function under different conditions. Brand awareness relates to two other concepts: brand recognition and brand recall performance. Brand recognition is the consumers' ability to confirm prior exposure to the brand as seen or heard before, when given the brand as a cue. Brand recall is the consumers' ability to retrieve the brand from memory when given the brand category as a cue. Increasing brand awareness is important because it enhances the probability of the brand being considered by the consumer and can affect decisions about brands in consumers' buying contexts. Brand image is described as “perceptions about a brand as reflected by the brand associations held in consumer memory” (Keller 1993, p.3). Brand associations hold the meaning of the brand for consumers and are categorized as: attributes,

(21)

21 benefits and attitudes. The dimensions of positive brand associations are: favorability, strength and uniqueness. Associations that are unique to the brand, strongly and favorably held, are crucial for a brand's success (Pitta and Katsanis 1995).

The study of brand equity is important due to its ability of enhancing the likelihood of consumers' brand choice, leading to brand loyalty and differentiating the brand from the competition (Pitta and Katsanis 1995). Moreover, with higher levels of awareness, brand equity should also increase the effectiveness of marketing communications (Pitta and Katsanis 1995). Brand equity is considered to be an investment for the company and as such it has a limited life (Pitta and Katsanis 1995). Indeed, it can increase and get stronger, or decline and be harmed also by the actions of the company (Pitta and Katsanis 1995). One of the actions that can be critical for a company because of its possible negative effects on brand equity is the brand extension strategy.

Brand extensions are successful when they contribute to improving brand associations for the mother brand, strengthening the image and fostering greater recognition or new associations (Martinez and Pina 2003). However, one of the possible downsides of brand extensions is the cannibalization of sales and the deterioration of the brand image (Martinez and Pina 2003). As brand image is an asset that is built up over the long term, it is important to investigate the possible negative effects that a brand extension, through the creation of new associations in the mind of consumers or the alteration of current feelings and beliefs about a brand, can have on it. In a cross-gender brand extension context the brand equity of the mother brand can be affected due to the fact that the extensions are inconsistent with the brand gender of the mother brand (Lieven et al. 2014). Therefore, it is necessary to investigate to what extend the effect on brand equity occurs and if the overall impact depends on the strength of the gender image of the parent brand and on customers‟ biological sex.

(22)

22

Chapter 3: Conceptual framework and hypotheses development

In the last years usually sex-typed brands extend their products to target the opposite sex (e.g. Dior with Dior Homme, L‟Oréal with MenExpert, Unilever with Dove Men Care, etc.); (Azar 2013). As a consequence of the increasing popularity of cross-gender brand extensions, researchers have started to study the effect of brands‟ sexual attributions on brand extension (Jung and Lee 2006; Ulrich 2013). However, research on cross-gender brand extension is still scarce and results of past studies are inconsistent. Therefore, the aim of this study is to shed light on the inconsistent results of previous researches and to address the gaps identified in the literature.

3.1 Brand image fit and evaluation of the extension

Brand image fit (or brand concept consistency) is different from product category fit because it evaluates how a brand image has an impact on consumer perceptions of fit rather than how the latter is affected by the similarity of product features, attributes or benefits (Diamantopoulos, Grime and Smith 2002). According to the associative network theory, brand image is defined as a mental scheme formed by a network of concepts (nodes) interconnected by linkages or associations. When an extension coherent with the pre-existing associations is introduced, the brand scheme will absorb the new concept without meaningful changes (Martinez Salinas and Pina Pérez 2008). However, if the extension significantly differs from the scheme, the latter alters and adjusts to the new associations (Park et al. 1993). In any case, far extensions bring to new associations and will alter the brand schema (Park et al. 1993). Conversely, the introduction of extensions close to the mother brand can be a tool for reinforcing current associations (Martinez Salinas and Pina Pérez 2008). When a new extension is introduced men and women can respond to the brand concept consistency differently. In other words, men and women can have a different perception of brand image fit between the mother brand and the extension that can be explained by a dissimilar way of processing information. Indeed, previous studies demonstrate that there is a difference between men and women in processing information. According to the selectivity theory women possess a lower

(23)

23 threshold for elaborative processing than men, resulting in the use of different strategies for processing information (Lau and Phau 2009). Brand image fit considers the image of a brand as the main evaluation criterion and involves specific brand associations that are connected to the initial image of an extension with the image of the parent brand (Lau and Phau 2009). The concept that each gender engages in diverse strategies of information processing has an impact on the evaluation of fit between the mother brand and the extension. Lau and Phau (2009) conducted a research with the aim of investigating the impact of gender on perceptual fit evaluation of prestige brands. The results show that women can establish stronger brand image fit due to their greater sensitivity in the recognition of brand images that match their self-concept. Indeed, when women receive moderately incongruent information they employ an item-specific processing strategy, whereas men use a schema-based processing strategy in order to evaluate the information. Women link easily their existing knowledge to the new information due to the fact that their item-specific strategy increases their ability to identify similar existing information in memory. Conversely, the employment of the schema-based processing strategy by men requires a higher elaboration threshold in order to recall and access to the existing information. However, according to Lau and Phau‟s (2009) findings, when brand images are either congruent between the parent brand and the extension or extremely incongruent, men and women will not show any differences in the brand image fit evaluation.

As a consequence, applying the results of Lau and Phau (2009) to this research we can assume that moderately incongruent information occurs when the cross-gender brand extension is made from a weak initial gender image of the mother brand extending to target the opposite sex. On the other hand, extremely incongruent information arises when the cross-gender brand extension refers to a strong initial gender image of the mother brand extending to target the opposite sex. Therefore, the following hypothesis is developed:

(24)

24 H1: the evaluation of a cross-gender brand extension will differ depending on the perceived

strength of the initial gender image of the mother brand (strong vs. weak), regardless of consumers’ gender.

Furthermore, as demonstrated by past research a greater perception of fit leads to a more positive consumers‟ evaluation of a brand extension (Aaker and Keller 1990). Therefore, applying the findings of Lau and Phau (2009) to our study regarding the different perceptions of brand image fit between men and women, the following hypotheses are developed:

H2: when the initial gender image of the mother brand is perceived to be weak, compared to men,

women will exhibit a higher perception of brand image fit, resulting in a more positive evaluation of the cross-gender brand extension than men.

H3: when the initial gender image of the mother brand is perceived to be strong, men and women

will exhibit the same perception of brand image fit, resulting in a similar evaluation of the cross-gender brand extension.

3.2 Evaluation of the mother brand post-extension launch

When using the same brand name for an extension, new attributes and beliefs are introduced in the mind of consumers which can be perceived to be consistent or inconsistent with the image of the mother brand (Loken and Roedder John 1993). One critical issue in this context is the dilution effect that may arise when the mother brand beliefs are changed by new information expressed by the brand extension that is inconsistent with the parent brand beliefs. Indeed, according to the results of an experimental investigation conducted by Loken and Roedder John (1993) the dilution effect is more likely to arise when brand extension attributes are inconsistent with the mother brand beliefs. Another important factor affecting brand equity, which is strictly related to the focus of this research, is brand gender. The importance of brand gender is due to the fact that it is a dimension of

(25)

25 brand personality and one of the drivers of customer-based brand equity is, in fact, brand personality (Keller 1993). According to Freling et al. (2011) a favorable brand personality enhances brand attitudes, purchase intentions, consumer trust and loyalty, leading to greater levels of brand equity. Consequently, since consumers apply the judgment of human personality traits (including gender) to brands (Aaker 1997; Grohmann 2009), brand gender can also be a predictor of brand equity.

For this purpose, Lieven et al. (2014) investigated on the relationship between masculine and feminine brand personality and brand equity. The results show that brands associated with high levels of masculinity and femininity drive higher equity than do androgynous and undifferentiated brands (Lieven et al. 2014). As brand equity results from favorable, strong and unique brand associations (Keller 1993) it means that the higher brand equity found for masculine and feminine brands (i.e. brands with a strong positioning in terms of brand gender), is based on these associations. Indeed, consumers can easily recognize and categorized with a high degree of certainty masculine and feminine brands (Lieven et al. 2014). As a consequence, this ease of categorization of sex-typed brands, elicits more favorable responses to strongly masculine or strongly feminine brands (Lieven et al. 2014). On the other hand, undifferentiated and androgynous brands are more difficult to categorize because of their ambiguous gender positioning and as a result they cannot drive higher brand equity (Lieven et al. 2014). These outcomes suggest that a dilution of gender associations of the mother brand due to inconsistent brand extensions can potentially have a negative effect on brand equity (Lieven et al. 2014).

Applying the findings of Lieven et al. (2014) to our research, the dilution effect is expected to be higher when the initial gender image of the mother brand is perceived to be strong, compared to a weak one, due to a higher inconsistency between the image of the mother brand and the image of the extension. Past research shows that a brand extension introduction can influence not only consumers‟ attitudes towards the extension itself but also how consumers evaluate the mother brand

(26)

26 post-extension launch (Loken and Roedder John 1993). Therefore, considering the overall evaluation of the mother brand post-extension launch in terms of both attitudes towards the mother brand and brand equity and in line with hypotheses H1, H2 and H3, the following hypotheses are predicted:

H4: the evaluation of the mother brand post-extension launch will differ depending on the perceived

strength of the initial gender image of the mother brand (strong vs. weak), regardless of consumers’ gender.

H5: when the initial gender image of the mother brand is perceived to be weak, women will have a

more positive evaluation of the mother brand than men, resulting in a lower dilution effect.

H6: when the initial gender image of the mother brand is perceived to be strong, men and women

will have a similar evaluation of the mother brand, resulting in a higher dilution effect.

3.3 Conceptual framework

The conceptual framework is shown in Figure 3.1. In summary, the hypotheses that will be tested in the research are: H1 - the evaluation of a cross-gender brand extension will differ depending on the perceived strength of the initial gender image of the mother brand (strong vs. weak), regardless of consumers‟ gender. H2 - when the initial gender image of the mother brand is perceived to be weak, compared to men, women will exhibit a higher perception of brand image fit, resulting in a more positive evaluation of the cross-gender brand extension than men. H3 - when the initial gender image of the mother brand is perceived to be strong, men and women will exhibit the same perception of brand image fit resulting in a similar evaluation of the cross-gender brand extension. H4 – the evaluation of the mother brand post-extension launch will differ depending on the perceived strength of the initial gender image of the mother brand (strong vs. weak), regardless of consumers‟ gender. H5 - when the initial gender image of the mother brand is perceived to be weak,

(27)

27 women will have a more positive evaluation of the mother brand than men, resulting in a lower dilution effect. H6 - when the initial gender image of the mother brand is perceived to be strong, men and women will have a similar evaluation of the mother brand, resulting in a higher dilution effect. H1 H2, H3

H5, H6

H4

Figure 3.1: Conceptual framework.

Brand image fit

Strength of the initial gender image of the mother brand: Weak Strong Evaluation of the cross-gender brand extension. Evaluation of the mother brand post-extension launch. Gender of the

consumer: Male Female

(28)

28

Chapter 4: Methodology

4.1 Pre-test

Before starting the experiment, a quantitative pre-test was conducted with the aim of obtaining the right stimulus resources and manipulations. The purpose of the pre-test questionnaire was to select and identify which brands are perceived to be strongly and weakly feminine and strong and weak in masculinity, by consumers. For this purpose, a neutral product category was selected in order to avoid that the product category itself was already seen as masculine or feminine by consumers. In other words, an equally feminine and masculine product category can prevent that gender stereotyping of the product category can distort the results of the study. Furthermore, both male and female consumers must be willing to use the product in the category to avoid a negative influence in the main study results. It is also necessary the existence of well-known brands with a strong and weak masculine and feminine image within the product category. Moreover, these gendered brands must be not expanded yet with a cross-gender extension on the marketplace, to enable the experimental testing of fictitious extensions of real brands (Ulrich 2013).

First, Jung and Lee‟s (2006) list was consulted, as they claimed to include in their study all potential product categories where gender crossing of a brand is relevant (e.g., fragrance, shampoo, alcoholic beverages, tobacco, fashion, undergarments, swimwear, magazines, shoes, sportswear, spectacles, and hair gel/spray). Second, because a neutral product category is necessary for this study, the categories that do not meet this requirement were rejected. Therefore, because from past research (Lieven et al. 2014; Azar 2013; Jung and Lee 2006; Ulrich 2013) product categories such as cigarettes, alcoholic beverages and cars, were tested and perceived to be strongly masculine, they do not meet the requirement of this study. Furthermore, product categories such as high-end fashion, lingerie and cosmetics are perceived to be strongly feminine, thus, again they were rejected for this research. In Azar's (2013) study, respondents evaluated each product category separately for its levels of masculinity and femininity, resulting in the identification of four genders for product

(29)

29 categories with different levels of masculinity and femininity (i.e. masculine, feminine, androgynous and undifferentiated). From the study it resulted that Jeans is considered to be an androgynous product category within the clothing segment. Therefore, based on Azar‟s (2013) results, this product category was considered for the research, because of its neutral gendered image.

The list of brands within the product category was identified on the basis of these criteria: 1) The brands should be well known;

2) They should have a gendered image;

3) They should have no cross-gender extension yet;

4) The brands should be in the same category and positioning to prevent bias.

Five feminine brands and five masculine ones were listed within the product category. The purpose was to choose one strong and one weak feminine brand, and one strong and weak masculine brand to test the extension in two directions: from feminine to masculine and from masculine to feminine. The following feminine brands were identified: Miss Sixty, Juicy Couture, Met, Motivi and Max Mara. On the other hand, these masculine brands were selected: Jack Jones, Antony Morato, Ben Sherman, Marlboro Classic and Energie.

A number of 24 people were asked to rate each brand in terms of its gendered image, likability and familiarity. Therefore, the pre-test asked about the familiarity, likability and the perceived strength of masculinity and femininity images of the chosen brands by the focus group. The familiarity was measured by a seven-point Likert scale answering to the following statement: “I am familiar with (name of the brand)”, where 1= strongly disagree and 7= strongly agree. The likability, that is, the attitude towards the brand, was measured by a seven-point Likert scale answering to the following statement: “I like (name of the brand)”, where 1= strongly disagree and 7= strongly agree. The strength of the gender image of each brand was measured by a seven-point Likert scale: 1= strongly

(30)

30 feminine, 7= strongly masculine. Furthermore, based on Grohmann's (2009) research, respondents had to indicate on a seven-point Likert scale if the masculine brand personality (MBP) traits: adventurous, brave, and dominant would describe each brand. The same procedure was applied considering the feminine brand personality (FBP) traits: tender, graceful, and sensitive. These traits are based on the scale developed by Grohmann (2009) that measures the gender aspects associated with brands. Therefore, respondents answered to the following statements “(name of the brand) can be described as adventurous/brave/dominant/tender/graceful/sensitive” for each brand, where 1= strongly disagree and 7= strongly agree. See Appendix I for the pre-test structure.

In total 24 respondents (50% male, 50% female) filled in the questionnaire consisting of the ten different brands. First brands were measured in terms of familiarity and likability. Table 4.1 shows the results of these two variables:

Mean Familiarity Mean Likability

Miss Sixty M= 4.28 , SD= 1.73 M= 3.94 , SD= 1.39 Juicy Couture M= 2.54 , SD= 2.08 M= 3.86 , SD= 1.33 Met M= 2.96 , SD= 2.05 M= 4.07 , SD= 1.33 Motivi M= 4.50 , SD= 2.12 M= 4.07 , SD= 1.51 Max Mara M= 4.88 , SD= 1.97 M= 5.08 , SD= 1.35 Jack Jones M= 4.33 , SD= 1.99 M= 4.21 , SD= 1.32 Antony Morato M= 3.80 , SD= 1.94 M= 3.52 , SD= 1.73 Ben Sherman M= 3.78 , SD= 2.30 M= 4.61 , SD= 1.37 Marlboro Classic M= 4.92 , SD= 1.64 M= 4.63 , SD= 1.53 Energie M= 4.75 , SD= 1.78 M= 4.04 , SD= 1.55

Table 4.1: Mean familiarity and likability brands.

Because the familiarity was measured on a seven-point scale, results are sufficient if the average score is higher than four. Therefore, only the familiar brands were considered for further analysis. Thus, based on the results, Miss Sixty, Motivi, Max Mara, Jack Jones, Marlboro Classic and

(31)

31 Energie were considered. The brands were measured in terms of masculinity and femininity. Using Grohmann‟s (2009) scales the results in terms of MBP and FBP of the familiar brands are shown in Table 4.2 for each brand:

Adventurous Brave Dominant Tender Graceful Sensitive TOT MBP TOT FBP Miss Sixty M= 3.56 SD= 1.44 M= 3.63 SD= 1.34 M= 3.47 SD= 1.46 M= 3.81 SD= 1.40 M= 3.88 SD= 1.29 M= 3.75 SD= 0.95 M= 10.66 M= 11.44 Motivi M= 3.07 SD= 1.27 M= 3.00 SD= 1.33 M= 3.30 SD= 1.44 M= 4.52 SD= 0.75 M= 4.59 SD= 1.05 M= 4.54 SD= 0.71 M= 9.37 M=13.65 Max Mara M= 3.65 SD= 1.44 M= 3.62 SD= 1.36 M= 4.12 SD= 1.56 M= 4.27 SD= 1.25 M= 5.08 SD= 1.13 M= 4.46 SD= 0.86 M= 11.39 M= 13.81 Jack Jones M= 4.29 SD= 1.43 M= 4.21 SD= 1.53 M= 4.13 SD= 1.30 M= 3.17 SD= 1.40 M= 3.25 SD= 1.36 M= 3.17 SD= 1.37 M= 12.63 M= 9.59 Marlboro Classic M= 5.21 SD= 1.38 M= 4.96 SD= 1.23 M= 5.00 SD= 1.32 M= 3.33 SD= 1.49 M= 3.42 SD= 1.64 M= 3.38 SD= 1.21 M= 15.17 M= 10.13 Energie M= 4.13 SD= 1.39 M= 4.38 SD= 1.41 M= 4.33 SD= 1.24 M= 3.54 SD= 1.44 M= 3.38 SD= 1.44 M= 3.29 SD= 1.37 M= 12.84 M= 10.21

Table 4.2: Mean MBP and FBP familiar brands.

As expected the feminine brands scored on average less on the MBP and more on FBP compared to the masculine brands and vice versa.

Finally, in order to evaluate the strength of masculinity and femininity of each brand, respondents had to evaluate how strongly they perceived the overall masculinity and femininity of each brand. The results regarding the familiar brands are shown in Table 4.3:

(32)

32 Strength Femininity/Masculinity Miss Sixty M= 2.16 , SD= 1.22 Motivi M= 2.04 , SD= 1.06 Max Mara M= 2.31 , SD= 1.19 Jack Jones M= 5.33 , SD= 1.05 Marlboro Classic M= 5.63 , SD= 1.38 Energie M= 5.08 , SD= 1.18

Table 4.3: Mean of the strength of femininity/masculinity familiar brands.

Based on the results of the pre-test, four gendered brands were chosen, therefore, the main survey will have two directions: from feminine to masculine and vice versa. Familiarity was the first factor that was considered because respondents need to know the brand in order to fill in the main questionnaire correctly. Then, the four brands have to satisfy the requirements needed to test the hypotheses for the main survey. Therefore, one feminine brand must be perceived as weak in femininity and one as strong in femininity and vice versa for the masculine traits. Based on these requirements, among the above mentioned brands the choice of the four brands that will be used for the main survey is justified as follows. Because the brand Miss Sixty and Jack Jones scored lower in familiarity (Miss Sixty mean= 4.28, Jack Jones mean= 4.33) compared to the other brands examined in the pre-test (each of them scored a mean equal or above 4.50) they were not considered. Furthermore, both brands Miss Sixty and Jack Jones did not show a significant difference in terms of the overall perceived strength of masculinity/femininity (Miss Sixty mean= 2.16, Jack Jones mean= 5.33) compared to the other tested feminine and masculine brands. The other four brands satisfy the requirements for the main survey. Indeed, Motivi scored on average 4.50 in familiarity and fulfilled the requirement for being considered as a strong feminine brand, scoring 13.65 on average on the FBP and 2.04 on the femininity strength. Max Mara scored on average 4.88 in familiarity and fulfills the requirement for being considered as a weak feminine brand, scoring 13.81 on average on the FBP and 2.31 on femininity strength. The same reasoning

(33)

33 was applied for the masculine brands. Indeed, Energie scored on average 4.75 in familiarity and fulfilled the requirement for being considered as a weak masculine brand, scoring 12.84 on the MBP and 5.08 on masculinity strength. Finally, Marlboro Classic scored on average 4.92 in familiarity and fulfilled the requirement for being considered as a strong masculine brand, scoring 15.17 on the MBP and 5.63 on masculinity strength. In conclusion, the chosen brands for the cross-gender brand extension from masculine to feminine were: Marlboro Classic and Energie; whereas for the extension from feminine to masculine, Max Mara and Motivi were considered.

4.2 Experimental design

The brands identified in the pre-test were used in the main survey to answer the research question. Then, the hypotheses described in the conceptual framework (Figure 3.1) were tested using quantitative research method. An experiment was conducted using fictitious extensions of the chosen existing brands. The cross-gender extensions in the questionnaire were fictitious to avoid prior associations in minds of respondents and prevent bias. The gender crossing extensions of each brand were manipulated by providing a scenario that announced an extension to target the opposite gender group. The scenario was structured in the same way for both strong and weak gendered brands and it was developed using the product category Jeans for the extension because it is considered to be a neutral product category as explained in the pre-test paragraph (Azar‟s 2013). Based on Jung and Lee‟s (2006) research, an example of the scenario considering the two brands Motivi and Marlboro Classic was: “Marlboro Classic (Motivi) is now a leading men‟s (women‟s) apparel brand name in the clothing category. Due to the success of its men‟s (women‟s) apparel line, the top management of Marlboro Classic (Motivi) decides that it is time to take advantage of this success. The company decides to launch a new product line–Marlboro Classic Ladies (Motivi for men) to target female (male) consumers, maintaining the same brand name. The new product line includes jeans for women (jeans for men)”. In addition, a detailed description of the new product line with the main important features was provided.

(34)

34 In order to test hypotheses H1, H2, H3, H4, H5 and H6 the dependent variables were: evaluation of the cross-gender brand extension; brand image fit and evaluation of the mother brand post-extension launch. The independent variable was the strength of the brand's gender image while the gender of the consumer was considered as a moderator. Furthermore, brand image fit was analyzed as a potential mediator as indicated in the conceptual model structure. One main questionnaire was developed with four different scenarios in which two were designed only for men introducing a brand extension from a feminine brand to target males and two for women, introducing a brand extension from a masculine brand to target females. The scenarios intended for a specific target group had only respondents who matched the target group (male or female consumers). Therefore, to ensure internal validity, the respondents were linked to a scenario after they had filled in their gender and they were randomly linked to the strong or weak cross-gender brand extension depending on their biological sex. Hence, it was a mixed 2 (brand gender strength: weak vs. strong) X 2 (gender of the respondent: male vs. female) design. The reason for this design was to measure the main effect of the strength of the initial gender image of the mother brand on consumers‟ evaluations of a cross-gender brand extension and on the parent brand evaluation post-extension launch.

The dependent variable related to the evaluation of the extension was measured in the start of the questionnaire to prevent bias in respondents‟ judgments. The cross-gender brand extension evaluation was measured in terms of both attitudes towards the extension and purchase intent of the extension. Therefore, three statements were used to measure the attitude towards the extension based on the research of Klink and Smith (2001), and Park et al. (1991) by asking to rate the statement on a seven-point Likert scale: “I feel favorable towards the extension”, “I think the extension is a good idea”, and “I feel pleased towards the extension”. Based on the research of Johnson (1979) one item on the buying intention of the product was asked based on the statement: “If I need this product I would buy this brand”. Then, the evaluation of the mother brand

(35)

post-35 extension launch was measured both in terms of attitudes towards the parent brand and brand equity. Therefore, attitude towards the original brand was measured using two items on a seven-point Likert scale adopted from Holbrook and Batra (1987) by asking to rate the statements: “I think this is a good brand”; “I like this brand”. Furthermore, brand equity was measured on five items, on a seven-point Likert scale. Three items were derived from the Overall Brand Equity Scale (OBE; Yoo et al. 2000: “It makes sense to buy X instead of any other brand, even if they are the same”; “Even if another brand has same features as X, I would prefer to buy X”; “If there is another brand as good as X, I prefer to buy X”). Two additional items addressed the brands‟ ability to generate price premiums and brand satisfaction (Aaker 1996: “It makes sense to pay more for X than for a similar product of another brand”; “I would recommend X to my friends”). Moreover, brand image fit was measured using three items from Bhat and Reddy (1997; 2001) and Loken and Roedder John (1993) by asking to rate the statements on a seven-point Likert scale: “I think that the extension seems logical”; “I think that the extension is consistent with the image of the brand”; “I think that the extension would be typical of the brand”. Finally, to guarantee that the results can only be determined by the dependent variables, some control variables were also included. These variables asked about familiarity and involvement related to the brand and the product category.

The following table shows the scales and items used for the main survey:

Scales Perceived Fit

Perception of brand image fit

Bhat and Reddy (1997; 2001) Loken and Roedder John (1993)

Items

I think that the extension seems logical

I think that the extension is consistent with the image of the brand I think that the extension would be typical of the brand

Evaluation of the extension

Attitude towards the extension

Klink and Smith (2001)

I feel favorable towards the extension I think the extension is a good idea

Strength of the gender image of the mother brand : the effect in a cross-gender brand extension context

Master Thesis Marketing