• No results found

How does brand loyalty influence the evaluation of extension products and attitude toward the parent brand?

N/A
N/A
Protected

Academic year: 2021

Share "How does brand loyalty influence the evaluation of extension products and attitude toward the parent brand?"

Copied!
92
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Faculty of Economics and Business

How Does Brand Loyalty Influence the

Evaluation of Extension Products and Attitude

toward the Parent Brand?

Master Thesis

Version: Final

Chencan Liu

Student number: 10826963

June 29, 2015

Supervisor: drs. Jorge Labadie MBM

Sencond Supervisor: dhr. drs. R.E.W. (Roger) Pruppers

MSc. Business Administration

(2)

This document is written by Student Chencan Liu 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)

With the complement of this thesis, my study period for master degree then comes to an end. During the 10 months, I have countered countless challenges in study and in life, however it made this experience special and precious. These four months for conducting the research for this these equipped me with comprehensive academic knowledge, business insights and practical skills.

First of all, I would like to thank all the professors, classmates and friends who gave me help during my study. Especially to my supervisor, Jorge Labadie, who supported and enlightened me throughout the entire thesis process. I would like to thank my second supervisor, Roger Pruppers, who provided with a lot of valuable advises during the plenary meetings and whose lecture Consumer Behavior inspired me and made me have the initial interest of the research in the topic of branding.

I would like to thank all the participants who joined the research whose efforts made the basis for this thesis.

Last but not the least, I would like to thank all my friends and family for their constant supports and encouragement.

Chencan Liu June 29th201

(4)

1. Introduction... 1

1.1 General introduction about brand extension and brand dilution... 1

1.2 The gap: the relationship between brand loyalty and brand dilution... 1

1.3 Problem definition...2

1.3.1 Problem statement and research question...2

1.3.2 Sub questions... 4

1.3.3 Delimitations... 5

1.4 Contribution... 6

1.4.1 Theoretical contributions... 6

1.4.2 Managerial contributions... 6

1.5 Structure of the thesis...7

2. Understanding on brand extension strategy...7

2.1 Benefits from brand extension strategy... 7

2.2 Dilution effect from brand extensions... 13

2.3 Brand loyalty and brand extension... 18

3. Hypotheses... 23

4. Research design...29

4.1 Pretest... 30

4.1.1 Pretest 1 Choice of Brands...30

4.1.2 Pretest 2 Choice of extension product... 31

4.1.3 Pretest 3 Finding loyal customers...34

4.2 Variable Description...35

4.2.1 Brand loyalty...35

4.2.2 Fit between original and extension product...35

4.2.3 Attitude toward extension product...36

4.2.4 Change in attitude toward the parent brand...36

4.2.5 Category dominance... 36

4.2.6 Gender... 37

5. Methodology... 39

6. Results... 42

6.1 Frequency test and descriptive analysis...42

6.2 Reliability and correlation test... 43

6.3 Hypothesis testing - Stage 1 ANOVA analysis... 44

6.4 Hypothesis testing - Stage 2 Mediation effect... 47

7. Discussion... 56

7.1 General discussion... 56

7.2 Managerial implications...61

8. Limitations and future research...63

9. Conclusion...64

(5)

Appendix 2 Survey in pretest 3 finding loyal customers...77 Appendix 3 SPSS output...79

List of figures and tables

Figures

Figure 1 Conceptual model, relationships among variables based on

hypotheses 29

Figure 2 Total effect of brand loyalty on attitude toward extension product 49 Figure 3 Result in research, multiple mediation effects of Perceived fit

between original and extension products and Attitude toward the extension product on the relationship between brand loyalty on Change in attitude toward parent brand

53

Figure 4 Direct effect among Brand loyalty, Perceived fit between original and extension product, Attitude toward the extension product, and Change in attitude toward the parent brand

54

Tables

Table 1 Type of all the variables in the research 30

Table 2 Result for brand choice in pretest 1 31

Table 3 Product similarity rank in pretest 2 33

Table 4 Concept consistency rank in pretest 2 33

Table 5 Difference in brand loyalty test between two groups of respondents 35

Table 6 Constructs in research and their items 38

Table 7 Overall roles of variables in all hypotheses 42 Table 8 Means, standard deviations, correlations and reliability 44 Table 9 Role of variables in hypothesis 1,2, and 3 44 Table 10 ANOVA test for test the effect of brand loyalty on the perceived fit

between original and extension products H1 45 Table 11 ANOVA test for the effect of brand loyalty on the attitude toward

(6)

Table 13 Regression results for the perceived fit between original and extension product as mediator of the relationship between brand loyalty and attitude toward the extension product

47

Table 14 Direct effects among Brand Loyalty, Perceived fit between original and extension products, Attitude toward the extension product and control variables

48

Table 15 Regression result of multiple mediation effects on the relationship between BL and PBACi)Perceived fit between original and extension product as mediator, ii) Attitude toward the extension product as mediator, and iii) both Perceived fit between original and extension product and Attitude toward the extension product as mediators

50

Table 16 Direct effects among Brand loyalty, Perceived fit between original and extension products, Attitude toward the extension product, Change in attitude toward parent brand and control variables

51

List of abbreviations

BL --- brand loyalty

FIT --- perceived fit between original and extension products BEA --- attitude toward the extension product

PBAC --- change in attitude toward the parent brand BD --- brand dominance

(7)

The aim of this study is to discover how brand loyalty influences on the result of brand extension strategy. The main experiment was conducted through a between-subject experimental design, including 133 loyal customers as one group and 103 not loyal customers as the another group. A brand in auto industry and a fictitious extension product of the brand were used in the research.

The result showed that, compared to not loyal customers, loyal customer perceived higher fit between extension and original products, generated higher attitude toward the extension product and would change attitude toward parent brand more positively. Multiple mediation effects of the perceived fit between original and extension product and attitude toward extension product on the relationship between brand loyalty and the change in attitude toward parent brand were discovered. Contrary to previous studies in brand extension, the higher attitude toward extension product leaded to brand dilution instead of strengthening to the parent brand in this study.

The research therefore contributes to current academy with a deeper understanding of the brand extension and the reciprocal effect on parent brand, in the perspective of brand loyalty.

Keywords: Brand extension, Brand dilution, Brand attitude, Brand loyalty, Consumer

(8)

1. Introduction

1.1 General introduction about brand extension and brand dilution

Brand extension strategy is largely used to introduce new products and create more sales (Keller 1998). On the one hand, brand extensions can transfer perceptions and attitudes from the parent brand to extension products since brand extensions generate synergies between experience and communication (Erdem and Sun 2002). On the other hand, brand extensions can benefit parent brand by enhancing the brand meaning, strengthening brand equity and stimulating purchases of other products in the firm, especially for nonusers of the parent brand (Swaminathan 2001).

However, using brand extension strategies can cause brand dilution and bring risks to a company. Conducting a unsuccessful extension strategy can lead to damages in brand associations, brand attributes and brand beliefs, therefore causes huge damage to the brand assets (Holger 2008). If an extension strategy fails, the company needs to put a lot of effort to fix the damage in brand equity (Aaker and Keller 1990). Additionally, Sattler (2000) inferred an overview of potential risks of brand extension strategies which included negative effects on the existing brand equity, restriction to position the new product, increased need and administrative cost in coordination and retail distribution and uncertainty of success.

1.2 The gap: the relationship between brand loyalty and brand dilution

Sanjay and Kevin (2012) indicated that individual differences might relate to brand dilution effect. The interpretation of experiences is different for different people and

(9)

may therefore influence the evaluation of the brand extension. This view implied that brand loyalty might influence on the effects of a brand extension since loyal customers surely have different experience towards the brand than not loyal customers. The same authors indicated that loyal customers have richer, more developed knowledge structures of the brand, therefore they may have deeper convictions to the brand meaning than nonusers (Kevin and Sanjay 2003). From this point of view, it should be harder for loyal customers to try and accept a brand extension. However, the two authors (Sanjay and Kevin 2012) mentioned another opposite effect that loyal users may find attributional ways to deflect unfavorable extension evidence away from the parent brand. In another word, individual differences may also lead to a different way of processing as well.

The view that elaborated from same authors’ work interestingly includes two opposite effects from brand extension due to individual difference. Even though a great deal of academic work has been directed at the effect of brand equity dilution, no research has shown that brand loyalty may cause this effect as well. However, no research has specifically indicated that brand loyalty is related to the effect of brand extension and brand dilution yet. This research aims to fill this gap in the current literature.

1.3 Problem definition

1.3.1 Problem statement and research question

If consumers are committed to a brand attitude, dissonance theory predicts that consumers would tend to reject, distort, or rationalize information about a brand’s

(10)

association with a dislikable extension (Vicki Lane 1997). Cognitive dissonance is defined as psychological discomfort when an individual holds contradictory beliefs, ideas, values, or behaviors at the same time (Festinger 1957). The strong stereotype in the mind of loyal consumers makes it difficult for them to accept a change of the brand. Therefore, loyal consumers may be more likely to experience dissonance between the original brand’s attributes and the extended brand’s attributes.

People strive for general internal consistency (Festinger 1957). To reduce the discomfort, loyal customers will have to change. Loyal consumers can change the image of the brand in their mind or change the affective connections to the brand. Furthermore, people strive for consistency between the self-concept and behavior (Aronson 1992), therefore loyal customers may also change their purchasing behavior to reduce the. However, there is no research to discover which one of the pattern above is more likely to happen to loyal customers yet. Inspired by this, my research question is:

How does brand loyalty influence the evaluation of the extension product and attitude toward parent brand?

(11)

1.3.2 Sub questions

In order to answer the main research question, sub questions are hereby indicated to make a more comprehensive theoretical background. The evaluation of the extension product needs to be specified and the attitude toward parent brand needs to be explained. Brand attitudes were defined as the overall evaluations from customers of a brand (Keller 1993). In previous literature in the field of brand extension (Aaker and Keller 1990; Park and McCarthy 1997; Swaminathan 2001; Salinas and Perez 2009; Canli and Maheswaran 1998; Dwivedi 2009), the perceived fit between original and extension product and attitude toward extension product are generally used to research on the evaluation of the extension product, and the change in attitude toward parent brand is used to determine the reciprocal impact and dilution effect.

Therefore, the sub questions are as followed:

- How do loyal customers perceive the fit between original and extension product? Is it different compared to not loyal customers?

- How is the attitude toward extension brand different for the loyal customer compared to the not loyal customer?

- How will loyal customers change their attitudes toward the parent brand after receiving information about the extension product? Is the effect same for not loyal customers?

(12)

1.3.3 Delimitations

Due to the limited time frame, the research may have inevitable bias through data collection. It will be efficient to find experiments through social media and online forums since the experiment is searching for fans for certain brand. This brings bias to the result since the participants may be generally young and more interactive with digital information than others. Furthermore, students in University of Amsterdam are recruited as respondents in this research as well, including both international and local students. The different personal and cultural background may also have influence on the final result which will not include in the research.

The research is conducted by using a certain functional brand in auto industry. This will limit the application of the result, that it may not apply in luxury brands and brands in different industries. Furthermore, this research only uses perceived fit between original and extension brand and attitude toward the extension product as evaluation of extension product, and only uses the attitudinal change in parent brand as indicator for reciprocal effect. Other factors of evaluation toward a brand and indicate reciprocal effect will be not discovered and discussed. Therefore, the findings in the research will not provide a general framework on how brand loyalty influences the evaluation toward extension products and the reciprocal effect.

(13)

1.4 Contribution

1.4.1 Theoretical contributions

A theoretical model is indicated in this research to show the relationship among four main factors in the research, brand loyalty, perceived fit between original and extension product, attitude toward extension product and change in attitude toward parent brand. The model shows a complex relationship among all variables, including multiple mediation effects from the perceived fit and attitude toward extension product on the relationship between brand loyalty and attitude toward parent brand. The research discovers how brand loyalty influences the evaluation of a brand extension. Specifically, the research indicates how brand loyalty has impact on the perceived fit between original and extension product, and attitude toward the extension product. Additionally, the research discovers how brand loyalty influences on the change in attitude toward parent brand that impacted by the brand extensions.

1.4.2 Managerial contributions

The research will help company to understand how loyal customer may evaluate the extension product and how a brand extension strategy may influence its loyal customers. It may help managers make a better decision on whether conducting a brand extension strategy or not. The research will also provide insights for managers to generate marketing strategies after they conduct the brand extension strategy. It will help the managers to have a better way to promote the new products, design marketing campaigns and predict the market impacts for the new brand and the

(14)

mother brand. Managers may also have more insights of how to prevent brand dilution happening during the conduct of the brand extension. Furthermore, the research provides another perspective other than financial performance for managers to view the performance of a brand extension strategy.

1.5 Structure of the thesis

The thesis will begin with an introduction to existing literature about brand extension, brand dilution, brand loyalty and brand attitude. In chapter 3, seven hypotheses will be presented with related literature and explained with a conceptual framework among brand loyalty, perceived fit between original and extension product, attitude toward extension product and change in attitude toward parent brand. Then, the research design and methodology will be explained in next two chapters. In chapter 6, the analysis of data and results will be provided. The findings will be discussed and the managerial implications will be implied in Chapter 7. In the last part, limitations, possible direction for further research and conclusion with most important findings will be indicated. The thesis will be end up with references and appendix.

2. Understanding on brand extension strategy

2.1 Benefits from brand extension strategy

Brand extension strategy is widely used in companies since it can benefit companies in several perspectives. Keller (1998) indicated that almost 80 percent of the new products were generated by brand extension strategy to introduce new brands and to create sales in the 1990s.

(15)

First of all, using extension strategy can save companies time and costs. Launching a new product is not only time-consuming but also needs a big budget to create awareness and to promote the extension products (Tauber, E.M. 1981). The new products which produced with a known brand name already has a certain level of awareness in the early stage of its launch since the brand is existing in market already. Additionally, extension products can generate synergies between experience and the new products since it transfers consumers’ perceptions and attitudes from parent brand to extension products (Erdem and Sun 2002). It means that the image of parent brand will appear to the consumers when they reach a extension product. Therefore, a company can leverage its existing brand equity to guide consumers evaluate the extension product under the image of the mother brand. A conceptual model about how consumers use their knowledge to evaluate a brand extension was built by Keller and Aaker (1992). They explain it with the salience of the core brand associations in the extension context, the degree of relevance that consumers perceive this information to be to their extension evaluations, and the degree of how favorable that consumers perceive of the inferred associations in the extension context. According to their opinion, the extension evaluation depends on what information will be perceived by consumers about the core brand in the extension context, whether the information is suggestive for the extension product, and whether the information is viewed as good in comparison with competitors.

Other than saving costs, brand extension strategy was proved to have positive impact on the parent brand by adding brand equity and stimulating purchases of other

(16)

products in the company, especially for nonusers (Swaminathan 2001). Keller (1993) created a conceptual model of brand equity from the view of the individual consumer. He defined brand equity as overall evaluations of a brand from customers and classified brand associations into three major parts: attributes, benefits and attitudes. The first part of brand equity, brand attributes, is the descriptive features that a brand is characterized. Extension product is in different categories from the original brand, therefore the descriptive features can get more comprehensive through the extension products. The second part of the brand equity, brand benefits, refers to the personal value which consumers attach to the brand attributes. Extension products can add benefit to consumers since the brand then can do more for consumers by providing different kinds of products. With those views, it can be seen that brand extension strategy can benefit the mother brand equity.

However, the impact of extension product on attitudinal change of the brand is rather complicated. Brand attitudes were defined as the overall evaluations from customers of a brand (Keller 1993), and consumer brand choice is affected by those attitudes (Haugtvedt et al., 1993). A consumer’s attitude can also be valued as the total value of the brand’s attributes (Shocker and Srinivasan 1979). According to attitude component model, attitudes are evaluations of an object that have cognitive, affective, and behavioral components (Bem 1967). The cognitive component of attitudes represents the beliefs, thoughts, and attributes which are relevant with an object. Many times a person's attitude is probably based on the negative and positive attributes they associate with an object. The affective component of attitudes

(17)

represents consumer feelings or emotions connected to an attitude object. Affective responses influence attitudes in a number of ways. This negative affective response may make consumers hold a negative attitude towards spiders. The behavioral component of attitudes represents past behaviors or experiences regarding an attitude object and the idea that people may consider their attitudes from their previous actions. Consumers combine external stimuli to help them establish these thoughts and attitudes which are modified as they are exposed to and evaluate stimulus activity, then integrated with existing attitudes and thoughts to form and shape them to new levels (Fazio and Williams, 1986; Houston and Fazio, 1989), and in brand extension this process is complicated since the integration originates from two different kind of products. The complexity of attitudes and variant causes to the change of attitudes make the impact of extension product on consumers’ attitudes hard to be generalized.

The impact on the attitude part can lead to a significant result to a brand and influence the performance of an extension product or even the whole companies, therefore a large amount of literature in brand extension area are researching on the attitudinal impact for consumers. Brand attitude help the brand to overcome or offer a counter-attacking option if faced with a counter review or an offensive move from a aggressive competitive brand (Haugtvedt et al., 1993). Therefore, a brand with consumer’s a strong attitude is supposed to perform better and be strengthened than a brand with a weak attitude. A consumer’s attitude can also be valued as the total value of the brand’s attributes (Shocker and Srinivasan 1979). When faced with various kinds of brands in a product category, consumers tend to choose the brand to which

(18)

they hold the most favorable attitude (Keller, 1993). A brand with consumer’s a strong attitude is supposed to perform better since brand attitude helps a brand to overcome or offer a counter-attacking option if the brand is faced with a counter review or an offensive move from a competitive brand (Haugtvedt et al., 1993). Due to the important impact from brand attitude and complexity of it, a lot of scholars have been doing researches on brand extension focusing the attitudinal impact to consumers (Aaker and Keller 1990; Park and McCarthy 1997; Swaminathan 2001; Salinas and Perez 2009; Canli and Maheswaran 1998; Dwivedi 2009). A comprehensive findings on the factors which may impact on the result of extension strategy have been discovered.

Aaker and Keller (1990) researched on consumer behavior and provided a conceptual framework with thoughts for brand extension strategy. They used three dimensions to explain the fit of extension, “complement”, “substitute” and “transfer”. The complement indicates that to satisfy their specific needs, consumers take two product classes, the extension product and the parent brand product, as complementary. The substitute refers to products with the same user situation and satisfying the same needs so that the products can replace each other. The third dimension is transfer, indicating the relationship between extension product and manufacturer which “reflects the perceived ability of any firm operating in the first product class to make a product in the second class”. The first two dimensions are more focusing on the consumer’s needs and demands, and the last one is more focusing on the perceived ability of the firm. The perceived fit between original and extension product is

(19)

generally indicated as an important factor to have effect on the evaluation of the extension product (Aaker and Keller 1990; Park and McCarthy 1997 ; Farquhar and Herr 1992).

Researches have also found that brand name strategy can influence on consumer perspective of the extensions. Extension evaluations are influenced by category similarity once the extension is family branded (Sood and Keller 2012). The authors indicated that subbranding strategies could achieve better result through strengthening extension evaluations and protecting the parent brand from negative feedback. Compare with family branded extensions which trigger a faster and category-based processing strategy, sub-branded extensions can trigger a slower, more thoughtful sub-typing processing strategy. Additionally, it has been argued that the degree of perceived fit is a function of both similarity in product features and consistency in brand concept (Park, Milberg and Laweson 1991). Product feature similarity perceptions depend on identifying the relationships between product extensions and the brand's existing products, whether concrete or abstract, whereas the concept consistency depends on the extension the abilities of products to accommodate the brand concept. Besides those findings, brand dominance (Leong 1997), degree of a consumer’s involvement (Canli and Maheswaran 1998), online product reviews (Sachse and Mangold 2010), brand name strategy (Sood and Keller 2012), and individual differences (Sood and Keller 2012) has been discovered to influence on a success of a brand extension strategy as well.

(20)

2.2 Dilution effect from brand extensions

Even though brand extensions can benefit companies in several perspectives as indicated above, using brand extension strategies may can bring huge risks to a company as well. Sattler (2000) inferred an overview of potential risks of brand extension strategies which included negative effects on the existing brand equity, restriction to position the new product, increased need and administrative cost in coordination and retail distribution and uncertainty of success. The dilution effect on brand equity was indicated to happen if consumers’ evaluation of the brand’s association with the original product category were lower after than before the extension was made (Dwivedi 2009).

The large deal of researches on the failed extensions indicated that the category similarity or fit to the parent brand image could lead to brand dilution after the introduction of dissimilar extensions because the customers probably thought that the company was trying to take advantage of its brand name (Keller and Aaker 1990). Once the image of parent brand is separated, the brand dilution effect starts. It will be more likely to appear when a high level of similarity of “fit” is adopted by the company. Malaviya and Sternthal (2009) proved that two factors result in the feeling of nonfit and the brand dilution effect. The one is a stimulus which makes a comparison between a target and an alternative brand’s features at the outset of the message, the other one is a message format which brings a locomotion way of goal pursuit. Park, Milberg and Laweson (1991) proved that extension attitude affects brand associations. They also noticed the fit between the new product and either the

(21)

remaining products (category fit) or the brand image (image fit) can affect consumer attitude. Besides that, the author also indicated that consumer innovativeness plays a role as a moderating factor, then put forward that the characteristics of consumer personality can be far more crucial.

Some empirical studies illustrated the issue of reciprocal effects. Keller and Aaker (1992) discovered that a successful intervening extension improved evaluation of an average quality parent brand but did not influence a high quality parent brand. However, an unsuccessful intervening extension did not degrade evaluation of either a high quality or an average quality parent brand. Keller and Aaker also noticed that similarity of an extension did not influence parent brand evaluation. Meanwhile, Romeo (1991) noticed that negative extension information did not influence on parent brand evaluation. Moreover, research has illustrated the dilution effect on beliefs of parent brand when an extension was introduced (John, Loken and Joiner 1998; Loken and John 1993). Further, Kim, Han and Park (2001) discovered that parent brand evaluation was negatively affected, irrespective of whether line extensions, that is, extensions in the same product category as the parent brand, were positioned at a higher or a lower price and quality level than the parent brand.

Researchers has been researching on the feedback effects on parent brand equity by examining the influences of consumer experience with extensions. Even though the consumers had indirect product experience, dilution effects would still appear with poor performance of a dissimilar extension (Milberg, Park, and McCarthy 1997). However, other Sood and Keller (2012) argued about that, since such dilution pattern

(22)

did not occur in their study. According to their research, if consumers directly experienced a poorly performing and similar brand extension, dilution effects would happen. After consumers have a bad experience with dissimilar extensions, parent brands were not damaged. Furthermore, Sachse and Mangold (2010) recently launched a research and discovered that negative online product reviews have considerable disastrous influences on consumer-based brand equity, which result in a obvious brand equity dilution. Their study could demonstrate the fact that dilution effects can still happen even when consumers have indirect product experience as well.

Leong (1997) noticed that the effects of extending brands varing in their dominance of a particular product category seemed to be more complicated than previously conceived. The dilution of the strength of brand-product class association did not happen simply when a master brand was extended relative to a less dominant brand. In contrast, it appeared as a function of category dominance interacting with the success of an extension to a new product category rather than the similarity of the new product in the original category. Dilution occurred when a master brand was extended no matter what the outcome of the extension is, success of an extension of a less dominant brand led to an enhancement of the association between the brand and its original product class. For master brands, it was found that category dominance helped the recall of a brand in its original category and affected the brands to extend to a new product in a related category (Herr et al. 1993). However, master brands have more limitation to do so than less dominant brands since the transfer of this

(23)

effect is limited to products closely related to their original categories (Farquhar and Herr 1992). Sood and Keller (2003) supported the idea with a model of factors in brand equity dilution. The model included three main factors, strength, diagnosticity and inconsistency. Strength represents only an extension experience that is strong enough has the potential to cause brand dilution. Diagnosticity is that an extension experience is diagnostic of the parent brand only to the extent that consumers believe the extension is related to the parent. Inconsistency is that an extension experience consistent with the consumer’s image of the parent brand is less prone to affect consumer’s impression.

While literature on branding has always concentrated on the need for creating strong, positive attitudes toward the brand and the importance of consistency of the brand image, Buchanan, Simmons and Bickart (1999) proved that it is very important to comprehend the situational and contextual factors to guarantee the equity of the brand since it is established. The authors have discovered that context can establish conditions where consumers are prone to rely less on the previous attitudes and more on external cues (e.g. Retail display). In most cases, consumers rely on these previously formed attitudes in order to form the judgments about the brand. Then consumers comprehend displays and also separate displays as unique from other brands; this in itself may help strengthen the manufacturer's efforts to build and maintain brand image. What’s more, it also protects the brand from context effects through reducing the extent to which other brands are regarded as diagnostic or relevant. The high-equity brand valuations can be affected by presentation context and

(24)

those valuations are more prone to be affected when display factors result in constructive processing. Canli (2000) has indicated another external factor that timing also plays an important role in measuring brand equity dilution.

The researched also revealed that brand extension dilution was determined by the degree of a consumer’s involvement with the experience. Dilution effects are happened only with similar extensions when consumers are not actively involved. Less similar extensions were considered to be exceptions and their influence on the reputation of the parent brand is diminished (Canli and Maheswaran 1998). However, the authors found that if consumers actively process negative information about a brand extension, this would lead to parent brand dilution even with dissimilar extensions. Additionally, Kirmani, Sood and Bridges (1999) have discovered that greater love for owners for the parent brand would be expressed by more favorable responses to brand line stretches. This ownership effect was accompanied with upward and downward stretches of a functional brand and for upward stretches of a luxury brand, but not downward stretches of luxury brands. They also indicated that sub-branding strategies could be used to in order to avoid diminishing the brand's equity with current owners. What’s more, the degree of customer’s familiarity with the parent brand also affects the possibility of dilution by extensions. Vicki Lane and Robert Jacobson (1995) discovered evidence of greater brand dilution for consumers with a high “need for cognition” as compared to those with a low need for cognition. Those customers with a low need for cognition information more deeply.

(25)

2.3 Brand loyalty and brand extension

Based on the findings above on brand dilution effect which related to consumer’s involvement, owner effect and different customer’s groups , it is expected that brand loyalty may relate to the evaluation of extension product since brand meaning and experience to the brand is clearly different between loyal customers and not loyal customers. Sood and Keller (2012) have concluded that the individual differences would influence the interpretation of experiences and therefore influence brand dilution.

Brand loyalty was defined as a highly loyalty to rebuy or repatronize a preferred product or service consistently in the future by Oliver(1999), therefore, leading to repetitive same-brand or same brand-set purchasing, despite situational influences and marketing efforts may result in switching behavior. This definition emphasizes on two characteristics of brand loyalty. They are concept-behavioral and attitudinal loyalty(Aaker 1991; Assael 1998; Oliver 1999)which have been already mentioned in previous research. Behavioral loyalty is made up of repeated purchases of the brand. However, attitudinal brand loyalty contains a degree of dispositional commitment to certain unique value related to the brand (Chaudhuri and Morris 2001). Therefore, loyal users probably have deeper convictions concerning what is crucial to brand meaning than nonusers which may also lead to brand dilution according to the findings that indicated above. To retain customers, frequency programs has been largely used among companies, traditionally by loyalty cards in the real space to prevent brand switching at a product or store level (Dowling and Uncles 1997). It is

(26)

easier to implement frequency based loyalty program in e-markets since the presence of database technologies are an essential component for e-commerce sites (Deitel et al., 2001). Those concepts are easy to copy, however, hard to derive a sustainable and competitive advantage from them (Smith, 2000) since they are mainly a defensive tactic to prevent brand switching.

The traditional definition of attitudinal brand loyalty consists of cognitive, affective, and behavioral intent dimensions. In general, loyalty means satisfaction, but satisfaction may not bring loyalty. A satisfied customer is prone to be more loyal to a brand or store for a comparatively longer time than a customer who purchases for other reasons such as time limits and lack of information. Waddell(1995) and Oliver(1999) concluded that there is an asymmetric relationship between loyalty and satisfaction. Besides, Baldinger and Rubinson (1996) verified that highly loyal customers are more likely to stay loyal. In Schultz’s research, he illustrated important to satisfy customers (Schultz 2000). What’s more, it is more easily to turn a switching customer into a loyal one if the customer holds a positive attitude toward the brand.

Current literature have indicated one of the brand affects as a brand’s potential to cause an affective connections among customers. Brand trust brings brand loyalty because trust establishes exchange relationships(Morgan and Hunt 1994). Cannon (1997) emphasizes that the notion of trust is only relevant under the circumstance of uncertainty. Specifically, trust makes customers feel less uncertain during purchasing. Doney and Cannon (1997) put forward that the establishment of trust is a "calculative process" determined by the ability of an object or party (e.g., a brand) to continuously

(27)

meet its obligations and on an estimation of the costs versus rewards of staying in the relationship. In a word, brand trust is established after the long-term and careful consideration of customers. But the development of brand affect is more instant and less carefully reasoned.

Behavioral loyalty has been defined in terms of repeat buying behavior. In 1991, Aaker clearly made a difference between brand loyalty and brand associations. Behavioral intent connects attitude and behavior (Mittal and Kamakura, 2001). It will drive customers to make buying decision. What’s more, behavioral intent showed up in different forms such as an intent to buy a brand for the first time or a drive to rebuy a current brand. Previous brand loyalty research concentrated on maintaining and augmenting this rebuy intent (Oliva and Oliver 1992) and shifting behavioral intent to a real purchase (Kuhl and Beckmann 1985). From the perspective of traditional consumer marketing, a brand with strong customer loyalty possesses various advantages including ability to keep premium pricing, greater bargaining power with channels of distribution, diminished selling costs, bigger barrier to the new competitive products, and synergistic advantages of brand extensions to related product/service categories (Reichfeld, 1996). High brand loyalty can be expressed by a lack of multiloyal buying behavior and an intent to purchase only a single brand in one product category. In other words, behavioral loyalty in brings to a greater market share, and attitudinal loyalty brings to a higher relevant price for the brand.

However, other points of views of brand loyalty have also been pointed out. Jacoby & Kyner (1973) indicated that brand loyalty is different from either attitudes or habits,

(28)

but showed through them. Brand loyalty is different than purchase habit, because habitual purchases results from price loyalty (Day, 1969), however, high brand loyalty may result in a low level of price loyalty and certain extent of commitment towards a particular brand (Beatty & Kahle, 1988; Aaker, 1991). Besides, brand loyalty is distinguished from brand attitudes, because brand attitudes is supposed to be high for more than one brand, thus, lead to multiloyalty or loyalty to more than one brand (Jacoby, 1978).

In the study of brand extension, Sood and Keller (2013) pointed out that loyal consumers possess richer and more developed knowledge structures of the brand, and have deeper understanding of what is central to brand meaning than nonusers. They indicated that individual differences may affect brand evaluation and brand attitude. Since the knowledge structures of brand that loyal users hold are richer, more developed, loyal customer may interpret experience in a different way than non users, which may have impact on evaluation of the extension products. In other words, the interpretation of experiences is various, thus may affect the attitude of the brand extension. However, in another study from the same authors (Sanjay and Kevin 2012), they put forward another opposite effect that loyal users could probably find attributional ways to shift negative extension evidence away from the parent brand. In other words, individual differences may also result in differential processing too.

Additionally, cognitive dissonance supported the view that users with deep commitment to a brand may not easily accept extension products. The most classical stereotype in the mind of loyal consumers makes them hard to accept a change of the

(29)

brand. Therefore, loyal consumers more likely experience dissonance between the original brand’s attributes and the extended brand’s attributes. It is expected that consumers with commitment would deny, distort, or rationalize information about a brand’s association with a dislikable extension (Vicki Lane 1997). Dissonance theory is defined as the discomfort in psychological level when consumers have contradictory thoughts, ideas, values, or behaviors at the same time (Festinger 1957). When incompatibilty occurs, consumers are prone to be psychologically uncomfortable and they are eager to try to reduce this dissonance, and actively avoid situations and information which are likely to increase it. Because of this, people commit to dissonance reduction to keep their cognition and actions in line with one another. This creation of uniformity allows for a lessening of psychological tension and distress. In Festinger point of view, dissonance reduction can be achieved in four ways: change behavior or cognition, justify behavior or cognition by changing the conflicting cognition, justify behavior or cognition by adding new cognitions, and ignore or deny any information that conflicts with existing beliefs. Therefore, in order to diminish the discomfort, consumers can change the brand image and the attitude to the brand in their mind. In addition, people are eager for consistency between the self-concept and behavior (Aronson 1992), thus customers may also shift their buying behavior to clear off the discomfort.

Besides, other theories also make contributions to explain the psychological discomfort. In the field of behavioral economics, the endowment effect (also known as divestiture aversion) proposed that people attribute more value to things only

(30)

because they own them, according to the fact that people tend to pay more to keep something they own rather than to get something owned by others, even when there is no cause for attachment, or even if the item was only got minutes ago. Loss aversion illustrated the endowment effect that people put a higher value on their good products than on an identical good that they do not own (Kahneman, Knetsch, and Thaler 1990). According to these findings, loyal consumers may prone to value the brand image that they hold as if they own much more than they value what they don’t own but could get from the extension products.

3. Hypotheses

Kevin and Sanjay (2003) have indicated that the users of a brand have richer and more comprehensive knowledge structures of the brand, therefore they may have deeper convictions to the brand meaning than nonusers. Brand loyalty contains a degree of dis-positional commitment to certain unique value related to the brand (Arjun and Morris 2001). In this point of view, loyal customers would have deeper convictions to the brand than not loyal customers because loyal customers are not only using the brand, but also committed to the brand. Furthermore, cognitive dissonance theory (Festinger 1957) predicted that consumers who committed to a brand would reject the change of the brand image to keep internal consistency. Therefore, loyal customers may feel the extension brand is more different than other consumers do due to the deep convictions and stereotype of the brand meaning in their mind. In other word, loyal customer may perceive low fit between the extension and original product, since they do not like any change of the brand, even the

(31)

perceived fit is high to other consumers. Thus, following hypothesis is indicated:

Hypothesis 1: The perceived fit between original and extension product is higher when brand loyalty is absent than when the brand loyalty is present

Sood and Keller (2013) indicated that individual differences may affect brand attitude

because they affect the interpretation of experiences. Brand loyalty may influence the evaluation toward extension products since loyal customers certainly have different experience than non users. Loyal customers are more likely stick to a certain brand and generate affective connections to the brand (Aaker 1991; Oliver 1999). Additionally, Consumers who committed to a brand would tend to reject or distort the information of a dislikable extension product according to the dissonance theory (Vicki Lane 1997). However, no matter the extension product is likable or dislikable to other consumers, loyal customers could still experience the discomfort since the brand image is not consistent with the one in loyal customers’ mind. In this point of view, loyal customers would generate negative attitude toward extension products. This leads to the following hypothesis:

Hypothesis 2: The attitude toward brand extension product is higher when brand loyalty is absent than when brand loyalty is present

Research has shown that satisfied customer would be more loyal to the brand over time than the customer who purchased products because of other reasons such as time restrictions and information deficits (Waddell 1995). Additionally, there is an asymmetric relationship between loyalty and satisfaction (Oliver, 1999). Baldinger

(32)

and Rubinson (1996) indicated that highly loyal buyers tended to stay loyal if their attitude towards a brand was positive. In other word, loyal customers will be less likely to generate negative attitude toward the brand compared to not loyal customers. Canli and Maheswaran (1998) showed that some consumers would consider less similar extensions to be exceptions and thus their impact on the reputation of the parent brand was reduced. This effect may highly likely happen to loyal customers since they could therefore keep cognitive consistency in their minds. The following hypothesis is therefore proposed:

Hypothesis 3: the change of attitude toward parent brand is more positive when brand loyalty is present than when brand loyalty is absent

The fit between original and extension product is an important effect factor in brand extension literature. Research has examined the failed extensions that varied in similarity to the parent brand (Keller and Aaker 1990). When consumers see an extension product from a known brand name, they will evaluate it with different basis of perceived fit between the original and extension product. Keller (2003) stated that any association held in consumer memory about the parent brand may serve as a potential basis of fit. It is well accepted in current literature that greater similarity between parent brand and extension brand will lead to a greater transfer of attitude from the parent to the extension (Keller and Aaker 1990; Czellar 2003; Dwivedi 2009). As assumed in previous hypotheses, loyal customers would have lower attitude toward extension product (H2), it therefore might partially because they would perceive lower fit between original and extension product (H1). Therefore, there may

(33)

be a mediation effect of the perceived fit on the relationship between brand loyalty and the attitude toward extension product since the brand loyalty leads to a decrease in brand extension fit which in turn has a negative relationship with brand extension attitude. The following hypothesis is thus formulated:

Hypothesis 4: Brand extension fit mediates the relationship between brand loyalty and brand extension attitude, so that the perceived fit between original and extension product would make the effect of brand loyalty on the attitude toward extension product stronger.

It is well accepted in current literature that the perceived fit between original and extension product will have effects on the change in attitude toward parent brand. Dwivedi (2009) showed that the higher perceived fit, the more positive change would occur in attitude toward parent brand. Nevertheless, the brand dilution effect that the image of parent brand is hurt will be more likely to happen when there is a high degree of perceived fit involved (Park, Milberg and Lawson 1991). As assumed in H1, loyal customers may perceive lower fit than not loyal customers, this effect then could lead to a negative change in attitude toward parent brand for loyal customers. However, loyal customers is predicted to have more positive change in parent brand attitude than not loyal customer (H3). The perceived fit then predicted to be a mediator between brand loyalty and change in attitude toward parent brand. Therefore, following hypothesis is indicated:

(34)

loyalty and parent brand attitude change, so that the perceived fit between original and extension product would make the effect between brand loyalty and change in attitude toward parent brand weaker.

Literature has shown that attitude towards the extension product has positive influence on the parent brand. Zimmer and Bhat (2004) discovered that the evaluations toward parent brand would be enhanced after the introduction of an extension product. Dilution effects occurred with poor performance of a dissimilar extension even though participants had no direct product experience (Milberg, Park, and McCarthy 1997). Sachse and Mangold (2010) found that negative online feedback on the extension product would lead to a significant brand equity dilution. Sanjay and Keller (2012) found that dilution effects would happen when consumers directly experienced a poorly performing, similar brand extension, but consumer did not downgrade parent brands after a negative experience with dissimilar extensions. Those literature clearly showed that lower attitude toward extension product may cause a negative change in attitude toward parent brand. Loyal customers was predicted to hold a lower attitude toward the extension product (H2), this may then cause a negative influence on the attitude toward parent brand, even though loyal customers was predicted to have positive change in attitude toward parent brand (H3). In other words, the attitude toward parent brand may play a mediator role in the relationship between brand loyalty and the change in attitude toward parent brand. Therefore, following hypothesis is proposed:

(35)

between brand loyalty and change in attitude toward parent brand, so that attitude toward extension product would make the relationship between brand loyalty and change in attitude toward parent brand weaker.

As explained and assumed above, the perceived fit may mediate the relationship between brand loyalty and attitude toward parent brand (H5), and the attitude toward extension product may mediate the relationship between those two factors as well (H6). Additionally, the perceived fit is predicted to make the effect between brand loyalty and attitude toward extension product stronger (H4). Therefore, there might be a multiple mediation effect in the whole framework, which is the perceived fit and attitude toward extension product would mediate the relationship between brand loyalty and the change in attitude toward parent brand. In other words, loyal customers would perceive lower fit, therefore would hold lower extension attitude, and then cause a negative influence on the change in attitude toward parent brand. Additionally, loyal customers were predicted to have a positive change toward parent brand (H3). Therefore, it is expected that perceived fit and attitude toward extension product would influence the relationship between brand loyalty and the change in attitude toward parent brand. Thus, following hypothesis is proposed:

Hypothesis 7: Brand extension fit and brand extension attitude together mediate the relationship between brand loyalty and parent brand attitude change, so that the perceived fit and attitude toward extension product make the relationship between brand loyalty and change in attitude toward parent brand weaker.

(36)

Figure 1 Conceptual model, relationships among variables based on hypotheses

4. Research design

The overall design was a between-subjects research design, including two independent variables, two dependent variables and tow control variables. The role of variables are different in different hypothesis (table 1). The first factor brand loyalty, loyal and not loyal. The survey will be sent through online social medias in order to find loyal customers for the two research brands. The two dependent variables are brand extension attitude and the change of the attitude towards parent brand. Another two factors are brand dominance and gender which may influence the result of the research. All items (shown in table 4) used in the questionnaire were derived from literature in English in order to have a high validity.

(37)

Table 1 Type of all the variables in the research

Variables Type

Brand loyalty (BL) Ordinal Categorical

Fit between original and extended product (FIT)

Interval continuous Numerical Brand extension attitude(BEA) Interval continuous Numerical Change in attitude towards the parent

brand (PBAC)

Interval continuous Numerical Category dominance (CD) Interval continuous Numerical

Gender Nominal Categorical

4.1 Pretest

4.1.1 Pretest 1 Choice of Brands

The purpose of this pretest was to identify a brand name with a specific brand concept. The auto industry was chosen for the study. According to summary for brand loyalty measures (Bennett 2011), durable goods are with no brand switching, sole brand loyalty type and high involvement. Additionally, there is no conclusion in previous literature of what proportion of the possible factors relates to brand loyalty. Therefore brand loyalty for durable goods is less complicated to measure and more likely to be accurate compared to consumable and service market. Then, the functional brand image was chosen to be researched since it is purchased by larger group of customers than a luxury brand and people with positive attitude are more likely to purchase it instead of just being fans. Therefore the result of the research will be more reliable to predict the buying behavior. Finally, Volkswagen who met the criteria were chosen. Another consideration in stimulus selection was that only well-known brand names be examined for their potential to be extended. The reason is that fictitious brands carry neither brand loyalty which is the main factor in this research, nor attitudes that are

(38)

assumed to be changed by brand extensions.

Specifically, to find well-known brand within auto industry, a pretest was conducted with 20 master students who attended school in Netherlands. The participants were asked to mention 3 functional cars. The most mentioned functional car brands were Volkswagen, Toyota, Ford Motor, Volvo, and Chevrolet. Then another pretest with 40 students who were studying in Europe were asked to indicate the car brands above as functional or luxury brands (with a choice of “not sure”). The results showed 95 percent (38 out of 40) participants agreed Volkswagen as a functional brand. The result of pretest 1 is shown as table 3.

Table 2 Result for brand choice in pretest 1

Choice

Brand Functional Premium not sure

Chevrolet 29 5 6

Volvo 33 3 4

Toyota 35 2 3

Ford Motor 38 0 2

Volkswagen 39 0 1

4.1.2 Pretest 2 Choice of extension product

To find possible extension product that is perceived as middle fit so that it is easier to observe the difference of the perceived fit, higher or lower, during the final test. Since the fit between original and extension product has two perspectives, one is product fit and another one is brand concept fit (Park, Milberg and Laweson 1991). Both of the perspectives will be considered in this pretest.

(39)

During this pretest, 10 master students who are studying in the Netherlands were asked to mention 3 similar extension products and 2 dissimilar one. This pretest testified both of the perspectives. The most mentioned 15 similar extension products and 5 dissimilar products were gathered. Then they were asked to write down 3 important characteristics of Volkswagen. The most mentioned 2 characters were “reliable” and “durable” for Volkswagen. Another test among 80 students who were studying in Europe was conducted, using 7-point scales to generate a set of extension products that would vary independently on the dimensions of similarity to a car, prestige concept, and functional concept. Specifically, 40 subjects rated each of the 20 products on a 7-point scale according to their similarity to a car. The result of product similarity is shown as table 2 below. Then 20 different subjects rated each product according to how important the characteristics "durability" and "reliability" are when people buy these products. These characteristics correspond to the functional concept that is associated with Volkswagen and found in pretest 2 stage 1. Then, a third group of 20 subjects rated each product on how important the characteristics of "prestige" and "luxury" would be in deciding to buy these products. These characteristics correspond to the prestige concept that was also found in pretest 2 stage 1.

On the basis of the results (table 4 & table 5), I selected watch as the extension product which has middle feature similarity and high concept consistency.

(40)

Table 3 Product similarity rank in pretest 2

Rank Extension Product Mean(7-point scale)

1 Motorcycle 6.10 2 GPS device 5.45 3 Plane/Jet 4.70 4 sound equipment 4.25 5 Sunglasses 4.00 6 Watch 3.87 7 Leather jacket 3.45 8 Wallet 3.30 9 Gloves 3.28 10 Suit 3.25 11 Perfume 3.18 12 Phone 3.08 13 Sport shoes 3.02 14 solar panel 2.87 15 Television 2.63 16 Ring 2.58 17 Energy drink 2.53 18 Knifes 2.42 19 Furniture 2.35 20 Chewing gum 1.90

Table 4 Concept consistency rank in pretest 2 Rank Concept consistency--Volkswagen

(Mean in 7-point scale)

Concept consistency--Lamborghini (Mean in 7-point scale)

1 Motorcycle 5.50 Watch 6.0

2 Plane/jet 5.18 Sunglasses 5.6

3 GPS devices 5.09 Motorcycle 5.4

4 Watch 4.68 Plane/Jet 5.3

5 Sound equipment 4.64 Perfume 5.1

6 Phone 4.41 Phone 5.0

7 Solar panel 4.32 Sound

Equipment 4.9

(41)

4.1.3 Pretest 3 Finding loyal customers

This test is to test if people who are following fan pages and joining online forum of Volkswagen are relatively high in brand loyalty than people who were reached in other places and claimed not loyal for the brand. 50 people who were found through the online platforms and another 50 master students who claimed they were not loyal customers joined this pretest as two different groups. The items (table 6, page 38) used to test brand loyalty were in 5-Likert scale, would be explained in the “variable description” part. The value of Crobach’s alpha was 0.918 which showed that the scale has high reliability to indicate the result. Additionally, an one-way ANOVA test was used to test if there was significant difference in brand loyalty between the respondents from online fan pages of Volkswagen and respondents from university who claimed not loyal to Volkswagen. The result (Table 5) shows that there was significant difference between two groups, F(2, 100) = 91.56, p < .001. The average scale mean for those respondents who were found through online platform was 3.78 which is much higher than the average scale mean among the university students (average scale mean = 2.35) who claimed not loyal to the brand.

Therefore, the respondents who were found and reached though online social media and forums of Volkswagen would be indicated as loyal customer, and people who were reached in school and claimed not loyal for Volkswagen would be indicated as not loyal customer in this study.

(42)

Table 5 Difference in brand loyalty test between two groups of respondents SS DF MS F Sig. Group difference 144.480 1 144.480 669.473 .000 Error 21.150 98 .216 Total 165.630 99 4.2 Variable Description 4.2.1 Brand loyalty

It is commonly accepted in the literature (Jacoby and Chestnut, 1978; Mellens et al., 1996; Day, 1969) that the majority of loyalty measurements can be categorized as two kinds, behavioral or attitudinal. According to summary for brand loyalty measures (Bennett 2011), durable goods are with no brand switching, low purchase frequency, sole brand loyalty type and high involvement. Therefore, the measurement for brand loyalty consists of 4 attitudinal measures and 2 behavioral measures in this research, including loyalty items adapted from Odin and Florence (1999) and involvement scale (Kapferer and Laurent 1983), which translated and testified by Rodgers and Schneider (1993).

4.2.2 Fit between original and extension product

The degree of fit of Volkswagen watch supposed to be medium fit among all proposed extension products and ranked relatively high in perceived brand concept consistency. It was chosen by the result of pretest 2. However, in the final test, the fit is expected to be influenced by brand loyalty and therefore will be perceived differently. The fit of the brand extension was assessed using four items that were constructed based on Tauber’s definition of fit.

(43)

4.2.3 Attitude toward extension product

Brand attitudes were defined as the overall evaluations from customers of a brand (Keller 1993). According to attitude component model, attitudes are evaluations of an object that have cognitive, affective, and behavioral components (Bem 1967). Therefore, the cognitive, affective and behavioral attitude will be tested. The measure structures was adapted partly from the study by dwivedi (2009) whose survey is based on Bousch and Loken (1991) and, Hem and Iversen (2002).

4.2.4 Change in attitude toward the parent brand

A measure of the change in attitude in the parent brand is used to measure feedback effect. What’s more, brand equity is indicated to be diluted if subjects’ evaluation of the brand’s association with the original product category were lower after than before the extension was made (Dwivedi 2009). Hence, the summed difference between the before and after extension scores on the parent brand attitude was used to determine dilution. Higher positive scores indicated greater dilution. The items measuring PBAC were based on the same items that measured brand extension attitude.

4.2.5 Category dominance

Leong (1997) indicated that brand dominance has complex influence on extending brands. In his research, category dominance was measured by the strength of the brand’s association with the original product category. Therefore, category dominance is concluded as a control variable in the research. The assessment for category dominance was based on Leong’s research to measure the strength of brand’s

(44)

association, with three seven-point items: not dominant/dominant; poor/excellent example of the product category; and an agree/disagree statement that it represents the standard of the product category.

4.2.6 Gender

Literature has shown that men and women are different in buying patterns, motivations and social interaction (Candler 1991; Campbell 1997; Cross and Madson 1997) which may also influence the final conclusion for this research. Therefore, I conclude gender as one of the control variables.

(45)

Table 6 Constructs in research and their items

Construct Items

Brand Loyalty (BL) (Used in pretest 3) (5-point Likert scale)

1 I say positive things about X* to other people 2 When I buy X, it is like giving a gift to myself 3 The X I buy shows what type of man/woman I am 4 When I buy a car, X is always the first thing I’m looking at

5 Even when I hear negative information about the brand of X, I still stick to that brand

6 Even the products from X become more expensive, I will not switch to another car brand

Perceived fit between original and extension product (FIT) (5-point Likert scale)

1 The watch made by X makes sense

2 The decision from X to launch watch is very surprising

3 The decision from X to launch watch seems logical 4 The launch of watch by X in the market was expected

Attitude toward extension product (BEA)

(5-point Likert scale)

1 My attitude towards X watch is very positive 2 I am very negatively disposed towards X watch 3 According to me X watch is in great quality 4 I admire X watch a lot

5 I’m highly interested to buy a X watch Attitude toward parent brand

(5-point Likert scale)

1 My attitude towards X is very positive

2 I am very favorably disposed towards products from X

3 I think X is a brand with a bad reputation in auto industry

4 I admire brand X a lot

5 I’m not interested to buy a X car Parent brand attitude change

(PBAC)

The summed difference between the before and after extension scores on the parent brand attitude

Brand dominance (BD) (5-point Likert scale)

1 I think X it is not dominant/ dominant in auto category

2 I think X is a poor/excellent example of the product category

3 I think X represents the standard of the auto category

Gender Male / Female

(46)

5. Methodology

Data were collected by means of both online survey and face-to-face distribution. Survey administration started on May 23th2015 and was closed 10 days later on June

2nd 2015. The surveys were answered by 102 master students who claimed were not

loyal to Volkswagen in University of Amsterdam, and by 133 random fans of Volkswagen in online social media platforms. 150 of the respondents were female and 85 were males. The research was conducted between groups, one group of loyal customers and one group of not loyal customer. As stated in the pretest part, the 133 fans indicated the group of loyal customers, and 102 master students referred to the group of not loyal customers.

To perform the statistical analyses among the six variables in total (Table 7), brand loyalty (BL), the perceived fit between original and extension product (FIT), attitude toward extension brand (BEA), change in attitude toward parent brand (PBAC), brand dominance (BD) and gender, the Statistical software Package for Social Sciences (SPSS) was used. A frequency test firstly conducted after all data gets collected to identify errors in dataset. Then, counter-indicative items were recoded. The value of PBAC was counted by the difference between scale mean of parent brand attitude before extension(PBA1) and parent brand attitude after extension(PBA2). The scale mean was countered and new items were created into FITtot, BEAtot, PBA1tot, PBA2tot, PBACtot and BDtot. Therefore, the scale mean of item PBAC, the PBACtot, was computed with the difference of scale mean of PBA1tot and scale mean of PBA2tot. Moreover, to separate loyal customers from not loyal

(47)

customers, the loyal customers were coded as “1” and not loyal customers were coded as “0”. Gender as one control variable was coded as “1” for males and “2” for females.

After preparation of the dataset, a reliability analysis was tested to indicate whether the scales used in the survey were reliable (Cronbach’s Alpha>.70). This was aimed to show whether the most measures have an acceptable internal consistency. Then, a normality check was conducted to indicate whether the collected data was distributed symmetrically. A correlation analysis (2-tailed) was conducted among all variables to identify the correlations among all variables with a figure that presented with the value of Cronbach’s Alpha for each variable.

Finally, hypothesis testing will be conducted through different tests. For hypothesis 1, 2 and 3, the research aimed to test the difference between the group of loyal customers and the group of not loyal customers. In all of these three hypotheses, the independent variable is brand loyalty (BL), which is a categorical variable. Whereas, all dependent variables in these three hypotheses were continuous variables, the perceived fit between original and extension product (FIT) in hypothesis 1, attitude toward extension brand (BEA) in hypothesis 2 and change in attitude toward parent brand (PBAC) in hypothesis 3. Therefore, to test hypothesis 1, 2 and 3, one-way ANOVA test was used. More specifically, the univariate analysis of variance was used to indicate the result for controlling of gender and bran dominance.

(48)

effects in these hypotheses since the outcome variables and mediator variables are all continuous variables. Specifically, brand loyalty (BL) was the independent variable in all these 4 hypotheses. In hypothesis 4, the mediator was the perceived fit between original and extension product (FIT) and dependent variable was attitude toward extension brand (BEA). In hypothesis 5, the mediator was FIT and dependent variable was change in attitude toward parent brand (PBAC). In hypothesis 6, the mediator was BEA and dependent variable was PBAC. In hypothesis 7, there were two mediators, FIT and BEA, and the dependent variable was PBAC. Since the model proposed in this research can be computed by a SPSS macro, Process, made by Preacher & Hayes (2008), the regression test was therefore conducted through this program to analyze the effect with efficiency and therefore minimize errors in operation. The macro computed confidence intervals for the indirect effect of brand loyalty on parent brand attitude change as well as direct effect between variables. Additionally, the results of all the analyses were shown in table 7 (Page 42).

Referenties

GERELATEERDE DOCUMENTEN

This study showed that graduates are significantly more loyal to brands than students, so when brands can ensure that students have a preference for their brand (and

Furthermore, EU researchers who want to return after a mobility experience outside Europe experience difficulties related to the following job aspects: finding a suitable

I will argue throughout this thesis that according to the social relations between gender and space, women are restricted in their access to public space and, as a result, occupy

The Dutch government fell when the Freedom Party withdrew their support, unable to agree with the government on pounds 15 billion of government spending cuts.. Populists like

More precisely, this paper studies the relation between environmental policy and environmental patenting activity in the area of four renewable energy technologies (i.e. wind,

Moreover, the interaction effect of stereotypical thinking and extension type on brand extension acceptance also has a weak significant impact even though the dependent variable

H1: Low bonuses of executives have a higher effect on the brand loyalty of the customer than high bonuses H2: A positive financial period of three years has a better effect on

Hence, the main question examined in this study is: how does brand repositioning influence customer loyalty and how do exposure and self-congruity affect