STUDY OF THE RELATIONSHIP BETWEEN BRANDAWARENESS, BRAND ASSOCIATION, PERCEIVED QUALITY, AND BRAND LOYALTY

Shu -Hsien Liao Dept. of Management Sciences and Decision Making, Tamkang University, NO. 151, Yingjuan Rd, Dansui Jen, Taipei 251, Taiwan, R.O.C michael@mail.tku.edu.tw Retno Widowati PA Department of Management Science Tamkang University, Taiwan. NO. 151, Yingjuan Rd, Dansui Jen, Taipei 251, Taiwan, R.O.C wati704@yahoo.com Da-Chian Hu Department of Management Science, Tamkang University Taiwan. NO. 151, Yingjuan Rd, Dansui Jen, Taipei 251, Taiwan, R.O.C nowhere2@mail2000.com.tw ABSTRACT This research investigated direct and indirect relationship between brand equity constructs which includes brand awareness, brand association, perceived quality, and brand loyalty. We implement Structural Equation Modelling with LISREL to examine the hypothesis. The finding showed that there is a significant and positive direct effect between brand awareness toward brand associations, and brand association toward perceived quality and brand loyalty. We argue that brand association plays as a suppressor in our model that leads to inverse relation between brand awareness and perceived quality. With respects to the mediating effect of perceived quality to the relationship between brand awareness towards brand loyalty and brand association towards brand loyalty, our finding showed that perceived quality does not play as a mediator role in this study. On the other hand, we find that brand association is a very important variable which mediate the relationship between brand awareness toward brand loyalty. Keywords: Brand Equity, Brand awareness, Brand Association, Perceived Quality, Brand Loyalty.

INTRODUCTION The issue of brand equity emerge as one of the most crucial topics for marketing management in 1990s and its concept and measurement has interested academicians and practitioners for more than one decade. There have been three different perspectives for considering brand equity; The customer- based perspectives, the financial perspectives and combined perspectives. While this study focus on customer based brand equity. In recent years, customer-based brand equity has garnered considerable attention. Operationalization of customer based-brand equity usually fall into two groups: consumer perception and consumer behaviour, e.g. Mahajan, Rao, and Srivastava (1991) claimed that a customer-based brand equity could be measured by the level of consumer perception. While Farquhar (1990) claimed that brand equity is reflected by the change of consumer attitude while purchasing a product. Later researchers, beside using two approach: consumer perception and consumer behaviour, some combined the two approach and some were related brand equity to other variable as antecedents and consequences of brand equity, e.g. Keller (1993, 2001) mentioned about brand knowledge as combination between brand awareness and brand image, Lassar et al. (1995) evaluate only perceptual dimensions, Blackston (1995) study about the concept of brand meaning which include objective brand (personality characteristic, brand image) and subjective brand (brand attitude), Dyson et al (1996) using brand loyalty and brand attitude, Motameni and Shakroki (1998) proposed concept of global brand equity using brand strength and Prasad and Dev (2000) using brand performance and brand awareness to develop brand equity index. While Aakers (1991, 1996) incorporated the measurement, suggested measure four dimensions of brand equity: brand awareness, brand association perceived quality, and brand loyalty. Considering its comprehensiveness, this study based on the concept of brand equity by Aaker (1991, 1996) who established the four dimensions model of consumer based brand equity which the dimensionality has been tested by some researchers: (Cobb Walgren et al, 1995; Yoo & Donthu, 2001; Washburn and Plank, 2002; Pappu et al, 1998, 2005; Atilgan, 2005; and Kim and Kim, 2005). Consistent with Aakers conceptualization, Cobb-Walgren et al. (1995) and Pappu et al (2005) found four dimensions of brand equity, while contrasted with findings of Yoo and Donthu (2001,2002) and Washburn and Plank (2002) that also based on Aakers conceptualization, but observed only three brand equity dimensions. While previous studies tested and found associative relationship among the dimensions of brand equity, beside testing the dimensionality, this present research tested the direct and indirect causal relationship among comprehensive dimensions of brand equity which included brand awareness, brand association, perceived quality and brand loyalty. By doing so, we also could find which dimensions has influenced brand loyalty. In addition, although numerous local or global different product categories have been employed to measure the brand equity, literature on brand equity within the hospitality industry still relatively limited (e.g. Kim and Kim (2005) using luxury hotel and chain restaurant; Atilgan (2005) using beverage industry; even Pappu et al. (2005) using car brands and television brands This research would provide manager and researcher a conceptual framework to describe the relationships between the brand equity constructs and a more thorough understanding of consumer behaviour that implies for marketing strategist for making better strategic

decision about target market definition and product as well as better tactical decisions about specific marketing-mix actions. LITERATURE REVIEW Brand Equity The issue of brand equity has emerged as one of the most crucial topics for marketing management since 1990s( Leuthesser,1988; Keller,1993; Cobb-Walgren, Ruble, & Dontu, 1995: Lassar, Mittal & Sharma, 1995; Aaker, 1996; Dyson, Farr, and Hollis, 1996; Faircloth et al. 2001, Esch et al, 2006; Ramos & Franco et al. 2005. Brand equity has been considered in many context: the value added to the product (Jones, 1986; Lethesser 1988; Farquhar, 1990; Aaker,1991,1996,1999; Keller, 1993,1998, 1999; Kapferer, 1997); value of the firm (Aaker, 1991; Kim & Kim, 2005); value of the customer(Aaker1991; Martensen & Gronholt, 2003); brand preference, purchase intention (Lattin, 1987; Zeithaml 1988; Hardie et al 1993; Cobb-Wagren 1995); brand loyalty, brand awareness perceived quality, brand associations (Aaker, 1991; Keller,1993; Gralpois 1998, Pappu et al, 2005; Atilgan et al, 2005); differential effect of brand knowledge of consumer response to the marketing of brand (Keller,1993); incremental utility (Simon & Sullivan, 1993); consequence of marketing efforts (Ramos & Franco, 2005). There have been three different perspectives for considering brand equity; The customerbased perspectives, the financial perspectives and combined perspectives. While this study focus on the customer based perspectives. Customer- Based Brand Equity. The advantage of conceptualizing brand equity from the Customer- based perspective is that it enables managers to consider specifically how their marketing programs improves the value of their brands in the minds of consumers. Within the marketing literature, operationalization of customer based-brand equity usually fall into two groups (Cobb-Walgren et al. (1995); Yoo & Donthu (2001)): consumer perception (brand awareness, brand associations, perceived quality ) and consumer behaviour (brand loyalty, willingness to pay a high price). Mahajan, Rao, and Srivastava (1991) claimed that a customer-based brand equity could be measured by the level of customers perception. Also operationalized by Lassar et al. (1995) as an enhancement of the perceived utility and desirability that a brand name confers on a product. According to them, costumer- based brand equity indicates only perceptual dimensions, not including behavioral or attitudinal such as loyalty or usage intention, which differs from Aakers (1991) who suggested to measure brand equity including behavioural and attitudinal dimensions. Farquhar (1990) maintained that brand equity is reflected by the change of consumer attitude while purchasing a product. Aaker (1991) incorporated definitions, the four dimensions of brand equity namely brand awareness, brand association perceived quality, and brand loyalty. Customer- Based Brand Equity is defined as a set of Brand assets and liabilities linked to a brand, its name and symbol that add to or subtract from the value provided by a product or service to a firms consumers ( Aaker,1991). 3

On the other hand, some researcher related the customer based brand equity with other construct, e.g. Farquhar and Ijiri (1991) proposed a model by judging the corporations marketing efforts on its brand directly. While Lassar et al (1995) focused on relationship between customer based and financial/ market based brand equity measurement. Customer-based brand equity in this respect, is the driving force for incremental financial gains to the firm.
Table 1. Main different concepts:

Main contributor Mahajan Rao (1991) Farquhar (1990) Aaker ( 1991)

Concept Measure customer based brand equity by the level of customers perception -Brand equity is reflected by the change of consumer attitude while purchasing a product.

Measuring the four dimensions of brand equity: brand awareness, brand association perceived quality, and brand loyalty. Keller (1993) Adopted two basic approaches ( direct and indirect ) to measure customer- based brand equity emphasizing two constructs: brand awareness and brand image. The indirect approach to identify potential sources of costumer- based brand equities The direct approach focuses on consumer response to different elements of firms marketing program. Farquhar & Judging the corporations marketing efforts on its brand directly. Ijiri (1991) Lassar et al, Relationship between customer based and financial/ market based 1995 brand equity measurement. Customer-based brand equity in this respect, is the driving force for incremental financial gains to the firm. Hypothesis Relationships between Brand Awareness and Brand Associations. Aaker (1991,1996) argued that brand equity is a multidimensional construct, which consists of brand loyalty, Customer based brand awareness, perceived quality, brand loyalty and brand awareness and brand associations. brand equity occurs when the consumer has a high level of awareness and familiarity with the brand and holds some strong, favourable, and unique brand association memory. Fiske and Taylor (1995) contended that brand associations could be recalled in customers mind as emotional impressions. Brand awareness influences consumer decisions making by affecting the strength of the brand associations in their mind.(Keller,1993,1997). Pitta and Katsanis ( 1995) also pointed out that there are several dimensions of brand awareness with brand associations . They further indicated

that brand associations of the product can be stored in consumers minds after brand awareness of the product are already in their memory. Brand awareness and brand associations were found to be correlated (Atilgan et al 2005; and Pappu et al 2005). Moreover, high levels of brand awareness positively affect the formation of the products brand image ( association) (Ramos and Franco,2005). Esch et al ( 2006) also found that brand awareness affects brand image ( association). These literature review leads to hypotheses H1 H1: Brand Awareness has a positive direct effect to Brand Associations. The Relationship between Brand Awareness, Brand Association and Brand Loyalty Yoo, Donthu and Lee, (2000) researched about brand awareness with brand associations and brand equity. These researchers indicated that brand awareness with brand association has a significant positive effect on brand equity. While brand loyalty can be defined as combination of elements including the degree of customer satisfaction and the positive of brand associations. Thus it can be seen that if customers had higher brand associations and brand awareness, brand loyalty would increase. Similarly in Atilgans (2005) study, the more favourable associations consumers have towards a brand, the more their loyalty and vice versa. These literature review leads to hypotheses H2, H3 H2: Brand Awareness has a positive direct effect to Brand loyalty H3 : Brand Association has a positive direct effect to Brand Loyalty The Relationship between brand awareness, brand association and perceived quality. As studied by Aaker (1991), Keller (1993) and Pappu et al (2005), consumer who hold favourable associations towards a brand are also likely to develop favourable perceptions of quality and vice versa. Brand awareness in their study has been defined as consumers ability to recall that a brand is a member of product category. Consumers brand awareness is likely to be high when they have strong associations for the brand and when they perceived the quality of the brand to be high and vice a versa. These literature review leads to hypotheses H4 and H5: H4: Brand Awareness has a positive direct effect to Perceived Quality. H5: Brand Associations has a positive direct effect to Perceived Quality. The Relationship between Perceived Quality and Brand Loyalty. For several studies found that brand loyalty is related perceived quality ( Mc Connel,1968, Shapiro,1970 Szybiloo and Jacoby,1974) as reviewed by Lau and Lee (1999). Atilgan et al. (2005) studied the relationship between the dimensions of brand equity and brand equity itself. They concluded that brand loyalty is the most influence dimension of brand equity. Even their study did not give enough support to the existence of a direct causal relationship between the three dimensions brand awareness, brand association, and perceived quality and brand equity However, observed pair-wise

comparison suggested that there is a correlation between brand loyalty, brand awareness, and perceived quality. As a result they suggested that concentrating brand loyalty, should not undervalue the effect of brand awareness and perceived quality to brand loyalty. As Pappu et als study (2005) is envisaged that consumers perception of quality will be associated with their brand loyalty. The more brand loyal a consumer is, the more he or she is likely to perceived the brand as offering superior quality and vice a versa. These literature review leads to hypotheses H6 H6 : Perceived Quality has a positive direct effect to Brand Loyalty The Mediating Effect of Perceived Quality Brand awareness has been define as consumers ability to recall that a brand is a member of product category (Aaker, 1991). Consumers brand awareness is likely to be high when they have strong associations for the brand and when they perceived the quality of the brand to be high and vice a versa. Similarly, consumers perception of quality of a brand is likely to be high when they have strong association with the brand and vice versa Pappu et al. (2005). According to Aaker (1991) while brand awareness builds the familiarity liking sight and is a signal of substance/commitment, perceived quality acts as a differentiation tools. These literature review leads to hypotheses H7, H8 H7 : Perceive Quality is mediating the relationship between Brand Awareness and Brand Equity. H8: Perceived Quality is mediating the relationship between Brand Association and Brand loyalty. The Mediating Effect of Brand Association. Brand awareness influences consumer decisions making by affecting the strength of the brand associations in their mind.(Keller,1993,1997). Pitta and Katsanis (1995) also indicated that brand associations of the product can be stored in consumers minds after brand awareness of the product are already in their memory. Later Atilgan et al; and Pappu et al (2005) found that brand awareness and brand associations were correlated. These literature review leads to hypotheses H9 H9: Brand Association is mediating the relationship between Brand Awareness and Brand Loyalty.

Research Map To illustrate the relationship between brand equity dimensions and the relationships with another construct based on literature review, we present a research map shown in Figure 2.

FIGURE 1. Research map

METHOD. This research examined the relationship among the four most important dimensional constructs of brand equity which is includes brand awareness, brand associations, perceived quality and brand loyalty Research Framework Based on literature review and research hypotheses as illustrated in the last chapter, this study develop the conceptual framework of this research as shown in Figure 3

FIGURE 2: The Research Framework The proposed research framework present the relationship between four construct of customer based brand equity namely Brand awareness, brand association, perceived quality and brand loyalty. Measures Measures of brand equity consist of the four construct of customer based brand equity. This study employed a five point Likert scale anchored from 1 (strongly disagreed) to 5 (strongly agreed) which were adopted from Atilgan (2005) and Kim and Kim (2005). Brand awareness. Sample responded to four items designed to assess their ability to recognize and recall the brand as a member of a certain project category. The items from Atilgans questionnaires ( Atilgan et al, 2005) which including I am aware of this restaurant (BAW1), I can recognize this restaurant among other restaurants(BAW2), I know what this restaurant looks like( BAW3), Some characteristic of Mc Donald come to my mind quickly(BAW4). Brand Associations Fourteen items by Kim and Kim ( 2005) were adopted to measure anything linked in memory to the brand: It is crowded (BA1), It is noisy (BA2), The price is reasonable (BA3), Service is prompt (BA4). It is conveniently located (BA5), It has a differentiated image from other restaurant brands( BA6). It tastes good compare with price (BA7), Employees are very kind( BA8), It has a very clean image( BA9), It has cheerful and enchanting atmosphere (BA10), There are many event (BA11) I feel comfortable to visit alone (BA12), It has a long history( BA13), Its brand is familiar to me( BA14).

Perceived Quality Ten items by Kim and Kim ( 2005) were adopted to measure customer perception of overall quality or superiority of a product or service with respect to its intended purpose relative to alternatives including, The physical facilities (e.g building, sign, room dcor, illumination) are visually appealing(PQ1), The restaurant staff gives customers individual attention(PQ2), The appearance of staff members are clean, neat, appropriately dressed (PQ3), The restaurant has operating hours convenient to all their customers ( PQ4), The staff provides prompt service at promised times (PQ5). The staff handles customers complaints effectively(PQ6). The staff is always willing to customers(PQ7), The knowledge and confidence of the staff are good(PQ8), The food quality of the restaurant is good(PQ9), and The restaurant insists on error-free service( PQ10).

Brand Loyalty Five items by Kim and Kim (2005) were adopted to measure the attachment that customer has to a brand including : I regularly visit this restaurant( BL1), I intend to visit this restaurant again( BL2), I usually use this restaurant as my first choice compared to other restaurants(BL3), I am satisfied with the visit to this restaurant(BL4), I would recommend this restaurant to others (BL5), and I would not switch to another restaurant for the next time( BL6). Pilot study and pre-test of questionnaires. We prepared questionnaires adopt from Kim and Kim ( 2005) and Atilgan et al (2005) which also based on Aakers model and translated from English into Chinese, then reviewed and revised by a Professor and 4 PhD candidates. We complete a questionnaires contained 34 items/ statement. The data in this pre-test study were collected through self-administered questionnaires in Taipei, Taiwan, distributed to 82 respondents of teenagers customers in the spot location of Mc Donalds restaurant. From the result of pre-test for the statistic evidence reasons we excluded 6 statements/ items, 5 items from brand associations questionnaires and 1 statement/ item from brand loyalty because their factor loading is too low (<0.4) and item to total correlation of 5 brand association items is too low (<.05) and item to total correlation of brand loyalty item is too high (>1). Finally we get 28 statements/ items contain four dimensions measurement. Sample and Procedures After pre-test and modification of questionnaires, questionnaires were distributed in the spot location of Mc Donalds restaurant. We chose Mc Donalds fast food restaurant ( quick service restaurant/ QSR) as our object of research because it is one of the good example of company brand, well-known and the largest global brand of

chain restaurant but implement local adaptation. Compare to the closest competitors, got sales 2.6 $ billion or occupied 45% of market share in 2006. We chose Taiwan sample, because Taiwan as other place in Asia become more important market that there are many growth opportunities internationally. The chain had posted an average of 20 percent growth in single-store sales over the past three years, making Taiwan one of the top 20 markets internationally for McDonald's expansion. Even Mc Donalds targeting a diverse market ranging from children to elderly people, but we chose segment of teenagers as our research target, because Mc Donalds is a global phenomenon so are the teenagers. The teenager identity became inextricably linked to leisure and hedonic consumption. In this study we selected teenagers consumers in the age between 13 -19 years old. We targeted the frequent consumer of Mc Donalds restaurant who is at least come to the Mc Donalds restaurant 4 times a month or once a week (Kara et al, 1997) other than less frequent consumers To proceed the formal data collection, we first explain the purpose of this study and show respondents how to fill in the questionnaires. If respondents have problems in filling in the questionnaires, they can directly ask us . It takes ten to fifteen minutes to complete the formal questionnaires. Survey questionnaires were hand distributed to 427 respondents, and retrieved with the aid of undergraduate and graduate university students. A total of 418 usable questionnaires or 98.82 percent response rate. From the retrieved questionnaires we separate into the frequent and less frequent consumers data. We conducted our analysis on 213 valid data of frequent consumers. Before the process of quantitative data analysis began, we employed human efforts in dealing with retrieved questionnaires, using SPSS 13.00 for Windows to analyse the data of questionnaires. Incomplete questionnaires were regarded as invalid. We enlisted the code for the retrieved questionnaires and entered them into the table From the demographic questions we got the data as follows: The sample of 213 consumers consisted of 50.2 % male, and 28.2 % were junior high school and 50.7% were high school students. Their age range from 13 to 19 years old: 13 years old : 7.5%, 14 years old : 8.5%, 15 years old; 12.2%, 16 years old: 18.3 %, 17 years old: 20.7%, 18 years old: 15%, 19 years old: 17.4%. Their frequency to visit the restaurant; 4-5 times a month: 66.7%, 6-7 times a month: 13.1%, more than 8 times a month: 19.3 %.Of the samples, 50.7% have income (pocket money) less than NT 3000 dollars monthly, and 30% have NT 3000 - NT 5000 dollars monthly. The purpose to visit the restaurant of the sample 32.4% are just to eat ,21.6% to study , 31.0%, to chat. The samples reason to visit Mc Donalds restaurant because of: Clean: 28.2%, The climate: 20.7%,Quality of food: 19.2%, Quality of service: 7.5%, Fashion: 8%, Others: 16.4%. Their preference of taste are : no prefer : 40.4%, salty: 27.7%, sweet: 14.1 % sour: 13%, spicy: 11.7%. They stated that advertising not influenced them: 64.8%, influenced them: 35.2%. They are not influenced by new product 76.5%, influenced by new product 23.5%. The respondent also visit other fast food restaurant: KFC: 70.9%, Pizza Hut :

10

12.2%, Burger King : 6.1%, Other : 10.8%. While they visit location of Mc Donalds around: shopping center: 43.2%, school: 25.8%, tourist places: 17.4%, office: 13.6% ANALYSIS AND RESULT Testing of Measurement model We performed confirmatory factor analysis on the four variables: Brand awareness, Brand association, Perceived Quality, and Brand Loyalty. We tried to asses overall fit of the model. First we find that the overall fit is not very good. Then we exclude 11 items ( BL3, BL5, PQ1, PQ2, PQ4, PQ9, BA1, BA2 BA8, BA9, BAW4) which has the modification index is too high (>0.5) and standard solution is too low (< 0.4), then we find a better overall fit. The measurement model provided an acceptable fit to the data when considering fit statistics. GFI 0.91; SRMR 0.063; RMSEA 0.053; NNFI 0.97; CFI 0.98; 2 179.90 DF; 113; 2/ DF 1.592 Reliability Analysis Cronbachs coefficient and item to total correlations are both used to measure the internal consistency of each identified construct. The reliability of the construct is acceptable if Cronbachs exceeds 0.70 and item-to-total correlations have greater than 0.50 ( Hair, Anderson, Tatham & Black, 1998) Chronbachs coefficient, item-to-total correlations, mean and standard deviation are listed in Table 2 which shows that this research has achieved the high reliability (Cronbachs coefficient all above 0. 74 and item-to total correlations most around 0.50 ). In term of the quality measurement model for the full sample, the constructs display satisfactory levels of reliability, as indicated by composite reliability ranging from 0.66 to 0.83. The value of skewness shows that less than 2 and kurtosis less than 7 it means that the data is normal, meeting the normality assumption ( Curran et al, 1996)

11

TABLE 2. Reliability Variab/ Cron items bach alpha Mean BL .745 3.770 BL1 3.73 BL2 4.03 BL3 3.66 BL4 3.68 BL5 3.75 PQ PQ1 PQ2 PQ3 PQ4 PQ5 PQ6 PQ7 PQ8 PQ9 PQ10 .861 3.578 3.67 3.29 3.85 4.00 3.72 3.34 3.62 3.39 3.65 3.25 3.7522 3.74 3.85 3.49 3.71 3.82 3.85 3.68 3.77 3.87

Std. Deviati on .56655 .847 .729 .901 .741 .794 .56775 .893 .900 .796 .906 .855 .906 .802 .809 .790 .842 .51258 .888 .837 .833 .765 .758 .726 .953 .813 .834

Skewness .011 .404 1.058 -.497 198 -.086 .093 -.226 -.127 -.694 -.884 -.747 -.003 -.076 -.224 -.255 .065 .087 -.762 -.792 .092 -.346 -.533 -.127 -.663 -.287 -.250

Kurto sis -.246 -.668 -.640 -.245 -1.84 -2.07 1.7 -.297-.157 1.237 -.694 1.075 .171 .112 .024 -.016 -.224 .31 .648 1.045 -.301 .585 .960 -.352 .454 -.089 -.627

Items to Composite totalcorrel reliability ation 0.6649 .440 .573 .462 .572 .525 0.8260 .502 .568 .570 .453 .662 .625 .654 .633 .583 .482 0.7351 .519 .432 .535 .620 .487 .514 .496 .398 .447

BA BA1 BA2 BA3 BA4 BA5 BA6 BA7 BA8 BA9

.801

BAW .801 4.0516 .61691 -.288 -.404 0.7694 BAW1 4.37 .751 -1.190 1.709 .572 BAW2 4.37 .745 -.861 .182 .638 BAW3 3.69 .858 .066 .591 .507 BAW4 3.78 .874 -.163 -.366 .533 Note : BL: Brand Loyalty, PQ : Perceived Quality, BA: Brand Association, BAW: Brand Awareness.

12

Validity Analysis (1). Content Validity Content Validity means whether the content of questionnaire reflects the subjects of the study or not. It also checks whether the measurement selected by the researchers can measure the topic correctly. Because of being constructed based upon previous research, the questionnaire in the formal study shows content validity. (2). Convergent Validity. As shown at table 3, all of the estimated parameters were statistically significant (p < .05) the T- value of all questions are between 6.13- 12.00 indicate excellent validity. It means that all the measurement model in our study has convergent validity. TABLE 3. Convergent Validity Variable Brand loyalty Parameter 1 2 3 4 5 6 7 8 9 Factor loading 0.45 0.51 0.49 0.45 0.60 0.66 0.62 0.56 0.44 Standard error 0.06 0.05 0.05 0.05 0.05 0.06 0.05 0.05 0.06 T-value 7.02*** 9.62*** 9.01*** 8.36*** 11.01*** 11.57*** 12.51*** 10.91*** 7.64*** Standardized solution 0.53 .0.70 0.66 0.56 0.70 0.73 0.77 0.69 0.52

Perceived quality

Brand Associations

10 11 12 13 14 15 16 17

0.46 0.56 0.44 0.43 0.51 0.59 0.68 0.37 T 2.58

0.06 0.05 0.05 0.05 0.07 0.05 0.05 0.06

8.14*** 11.53*** 8.69*** 8.78*** 7.79*** 12.00*** 14.00*** 6.13***

0.55 0.73 0.58 0.59 0.53 0.79 0.91 0.43 T 3.29 at p

Brand Awareness T 1.96

at p0.05 level*

at p 0.01 level** and

0.001 level *** (3). Discriminant validity To assess discriminate validity, we tested a series of Chi- square ( 2) difference tests on the factor correlations among all the constructs (Anderson and Gerbing, 1998). 13

This was done for one pair of variables at a time by constraining the estimated correlation parameter between them to 1.0 and then performing a Chi- square 2 difference test on the values obtained for the constrained and unconstrained models (Anderson and 2 Gerbing, 1998). The resulting significant difference in indicates that the two constructs are not perfectly correlated and that discriminate validity is achieved (Bagozzi and Phillips, 1982). Based on Table. 4 all of the 2 difference in this study is greater than 3.84, which this is a good evidence for the dimensions discriminate validity. TABLE 4.Discriminant Validity Variable Model Unconstrained model BL- PQ BL-BA BL-BAW PQ- BA PQ-BAW BA-BAW . Correlation analysis Means, standard deviations, and inter-correlations for the study variables are reported in Table 5. Correlations reflecting the relationship between research variables predicted by the hypotheses were positive significant. TABLE 5. Descriptive statistics and correlations matrix of construct Variable Mean SD BL PQ BA BAW BL PQ BA BAW 3.7700 3.5789 3.7522 4.0516 .56655 .56775 .51258 .61691 1 .474(**) .546(**) .388(**) 2 179.90 239.21 203.23 244.01 195.52 333.75 320.93 DF 113 114 114 114 114 114 114 59.31 23.33 64.11 15.62 153.85 141.03 1 1 1 1 1 1 2 df

1 .725(**) .344(**)

1 .533(**)

** Correlation is significant at the 0.01 level (2-tailed). Correlation can only reveal the degree of relationship between construct. To analyse the direct and indirect effect, as well as mediating effect among the construct, we applied structural equation modelling. The Structural Model After testing the measurement model, we proceeded to examined the proposed structured model and the hypotheses.

14

The main purpose of this study is to analyze the causal relationships, including the mediating effect between brand awareness, brand associations, perceived quality and brand loyalty. From the structural model we got the result as follows: TABLE 6. Direct Effect.
Path BAW->BL BAW-> PQ BAW->BA BA->BL PQ-> BL BA -> PQ

K 11 K 21 K 31 F 12 F 13 F 23

Parameter estimate 0.14 -0.19 0.51 1.02 -0.45 0.98

Standard error 0.14 0.08 0.10 0.47 0.39 0.17

Tvalue 0.97 -2.25 5.25 2.17 -1.15 5.73

Standard solution 0.14 -0.19 0.51 1.02 -0.45 0.98

Hypotheses Positive Positive Positive Positive Positive Positive

Result Not supported Not supported Supported Supported Not supported Supported

The Direct Relationship Table 6 shows the path coefficient from brand awareness ( BAW) to brand loyalty (BL) is positive but not significant: ( K 11 = 0.14, T-value 0.97 p> 0.05) this not support the hypotheses ( H2) that brand awareness has a positive direct effect to brand loyalty. The path coefficient from Brand awareness ( BAW) to Perceived Quality (PQ) is significant even inverse: ( K 21= -0.19, T-value -2.25 p< 0.05) this not support the hypotheses (H4) that brand awareness has a positive direct effect to perceived quality. We argue that brand association plays as a suppressor in our model that leads to inverse relation between brand awareness and perceived quality. The path coefficient from Brand awareness (BAW) and Brand Association (BA) is significant( K 31= 0.51 T- value 5.25 p < 0.05), thus the hypotheses that brand awareness has a positive direct effect to brand association (H1) is supported. The path coefficient from Brand Associations ( BA) to Brand Loyalty ( BL) is significant( F 12= 1.02, Tvalue 2.17 p< 0.05) thus the hypothesis that Brand association has a positive direct effect to brand loyalty (H3) is supported. The path coefficient from Perceived Quality (PQ) to Brand loyalty (BL) is not significant even inverse ( F 13= -0.45, T-value -1.15 p>0.05) this not supported the hypothesis (H6) that Perceived quality has a positive direct effect to Brand loyalty. The path coefficient from Brand Association (BA) to Perceived Quality ( PQ) is significant ( F 23=0.98, T- value 5.73 p< 0.05) thus the hypothesis that Brand Associations has a positive direct effect to Perceived Quality (H5) is supported.

15

TABLE 7. The Total and Indirect Effect Endogenous Brand Associations Effect T value B. Awareness Direct .51*** 5.25*** Indirect --Total .51*** 5.25*** B. Associations Direct Indirect Total P. Quality Direct Indirect Total *p 0.05 level T 2.58 Perceived Quality Effect T value -.19 .50*** .31*** .98*** -.98*** -2.25* 4.63*** 3.65*** 5.73*** -5.73*** Brand Loyalty Effect T value .14 .38*** .52*** .58 *** -.58 *** -.45 --.45 T 3.29 0.97 2.76** 4.84*** 4.15*** -4.15*** -1.15 --1.15 at ***p

Exogenous

Endogenous

Endogenous

1.96

at **p 0.01 level and

0.001 level Indirect relationships ( Mediating Effect) Table 7 shows the existence of mediating effect on the structural model. Fully mediation ( only indirect effect) is found in the relationship between Brand awareness and Brand Loyalty via mediator, Brand Association. Fully mediation is also found in the relationship between Brand awareness and Perceived Quality because there is an inverse direct relationship between Brand awareness and Perceived quality but there is an indirect effect between Brand Awareness and Perceived Quality via Brand Association. We find that perceived quality is not mediate the relationship between Brand Awareness( BAW) toward brand loyalty (BL) and between Brand Associations (BA) toward Brand Loyalty (BL) this not support the hypothesis H7 and H8. In this case we find that Brand Association( BA) mediate the relationship between Brand Awareness(BAW) and Perceived Quality (PQ) and between Brand awareness (BAW) toward Brand Loyalty, thus the hypothesis H9 is supported .

16

H1 ! . 7

PX 11 ! .79 PX 21 ! .91 P X 31 ! .43 \1 PY 62 ! .73 P Y 82 ! .69 PY 42 ! .56 PY 72 ! .77 PY 52 ! .70 PY 92 ! .52

I 1 ! .7

H ! .17 H !.

11

I 11 ! . 7

PY 10.3 ! .55

L


I 1 ! .7

Note: BAW : Brand awareness, BL: Brand Loyalty, PQ : Perceived Quality, BA : Brand Association Figure 3. Result of proposed model in Lisrel Common method variance According to the technique of Harmans one factor test, if a single factor emerge from the factor analysis or one factor accounts for more than 50 % of the variance in the variable, common method variance is present (Matilla and Enz, 2002). Our analysis revealed 3 factor structure with no general factor present ( the 1 st factor account for 34 % variance). It does provide support for the absence of such general bias in the finding ( Matilla and Enz, 2002). CONCLUSION From this research we could present some contribution and managerial implication, as well as its limitation and suggestion for future research. Contribution and Implications One of the contribution of this research is that the result established that four dimensions model of consumer based brand equity is the distinct dimension/construct consistent with the conceptualization of Aakers (1991, 1996) that was similar to CobbWalgren et al. (1995) and Pappu et al (2005), while contrasted with findings of Yoo and Donthu (2001,2002) and Washburn and Plank (2002) that also based on Aakers conceptualization, but observed only three brand equity dimensions. Another contribution



I1 ! .

PY 13.3 ! .53 PY 1 .3 ! .59



I1 ! .

PY 1 .3 ! .58

L3



PY

.3 ! .73

1 ! 1.

17

PY 31 ! .66

I !. 7

L1

P Y 21 ! .70

I !. 1

I 1 ! .7

1!. 1



1 ! .1

PY 11 ! .



I !.

I !. 1 I !. 7

I7 ! . 1

I !.

I 9 ! .73

is that this present research enriched the consumer based brand equity measurement by tested and found that there are direct and indirect causal relationship among dimensions of brand equity, while previous studies tested and found associative relationship among dimensions of brand equity. In this study, we found that brand association is the most important variable which affect brand loyalty. We also found that brand awareness affect brand loyalty via brand associations. This has been indicated by Pitta and Katsanis (1995) that brand associations of the product can be stored in consumers minds after brand awareness of the product are already in their memory. Thus brand association is mediate the relationship between brand awareness toward brand loyalty support our hypothesis. We also found that brand association plays as a suppressor in our model that leads to inverse relation between brand awareness and perceived quality, and perceived quality toward brand loyalty that against our hypothesis. Finding shows that brand association become a very important dimension to affect brand loyalty. It has implication that manager have to maintain or strengthen their effort upon the Brand association as: to keep their clean-lines, cheerful atmosphere, good taste and price, and friendly staffs. Even Mc Donalds targeting a diverse market ranging from children to elderly people, manager have to be aware about customers characteristic relevant with their age and individual habit toward the restaurant. The descriptive statistic of demographic characteristic and information about individual habit of the consumer in this research may could be considered, or become an example for the manager to get information from their customer. While earlier studies were conducted using American (e.g. Yoo and Donthu, 2001,2002, Yoo et al 2000; Washburn and Plank, 2002), Korean samples ( Yoo and Donthu,2001), Australia sample ( Pappu, 2005), and Turkey sample (Atilgan, 2005) this present studies used Taiwan sample. Thus, this is one of the few studies testing Aakers (1991) framework of brand equity and measuring brand equity in Australia and Asia. Finally, the present study measured brand equity in a given product category for a given brand that not used in previous studies. Limitation and future research direction This research describe the causal relationship between brand equity constructs without relate them to their antecedents like marketing efforts and with their consequences like value of the firm, that might be more useful for the marketing strategies. Even this research using universal questionnaires, but using different setting might be affect the result. So, for future research direction it also would interesting to do the research in different setting as cross-national, regional and cultural research, which are necessity that imply to international marketing strategies. The method we chose to test the common method variance might be not the best method. For future study could employ other method.
.

Acknowledgment This research was funded by the National Science Council, Taiwan, under contract No. NSC 96-96-2416-H-032-003-MY2 18

REFERENCES Aaker DA, 1991.Managing Brand Equity,The Free Press, New York. Aaker DA and Keller, 1992. The Effect of sequential introduction and brand Extensions,Journal of Marketing Research. Aaker, A, David, 1992. The value of Brand Equity, Journal of Business Strategy. Vol 13. Aaker, A. David, 1996. Measuring Brand Equity Across Products and Markets, California Management Review, Spring Aaker, A.D and Jaubun, 2001. Value Relevance of Brand Attitude in High-Tech, Journal of Marketing Research 38. A Ailawadi L, Kusum and Keller, Lane Kevin, 2004. Understanding retail branding conceptual insights and research priorities: Journal of retailing, 80. Atilgan, Eda et al, 2005. Determinants of brand equity, Marketing intelligence and planning. 23. Bentler, PM, 1990. Comparative fit indexes in structural models, Psychological Buletin Caldwell , Niall and Joao R. Freire, 2004.The differences between branding a country, a region and a city: Applying the Brand Box model. Journal of Brand Management. Curran P.J.et al 1996. The robustness of test statistics to noonormality and specification error in confirmatory factor analysis. Psychological Bulletin, 109: 512-519. Datta, PR, 2003. The Determinant of Brand Loyalty, Journal of American Academy of Business, 3, pp-138-144 Diamantopoulus,A and Sigouw, JA,2000. Introducing LISREL: a guide for the uninitiated. Sage Publication. Doyle, Peter,2001. Building Value-based branding strategist, Journal of strategic Marketing. Esch, Franz-Rudolf et al, 2006. Are brands forever? How brand knowledge and relationships affect current and future purchases, Journal of Product & Brand Management ,13/2. Gounares, S and Vlasis Stathakopoulos, 2004. Antecedent and consequences of brand equity management. Journal of Brand Management, April. Ghozali, Imam and Fuad, 2005. Structural Equation Modelling, BP-Undip, Semarang. Johansson, Johny and Ilenka A. Reihanen, 2005. The esteem of Global Brands, Journal of Brand management, June. Jones, Peter and David Hilker, 2002. Customer Perception of Service Brand: A case study of J.D. Wetherspeons, British Food Journal vol 104 no 10. Jones, Richard, 2005. Finding sources of Brand value: Developing a stakeholder model of brand equity, Journal of brand management, October. Keller Lane, Kevin, 1993. Conceptualizing, Measuring, and Managing Customer-based Brand equity, Journal of Marketing, Jan: 57 Keller Lane, Kevin, 2001. Building Customer-Based Brand Equity. Marketing Management. Kim, Bumm Hong, 2003. The Effect on consumer-based brand equity on firms financial performance, The Journal of consumer marketing: 20, 4/5. Kim Gon Woo and Hong Bumm Kim, 2004. Measuring Customer based restaurant Brand Equity. Cornell Hotel and restaurant administration quarterly, May.

19

Kim Bumm Hong and Woo Gon Kim,2004. The relationship between brand equity and firmsperformance in luxury hotels and chain restaurants, Tourism management. Lassar, et al, 1995. Measuring customer Based Brand Equity. Journal of Consumer Marketing. Vol 12. Mc Donald Ek and Sharp RM 2000. Brand awareness effect on consumer decision making for a common repeat purchase product, a replication. Journal of Business Research, 48. Martin M, Ingrid and David W Stewart, 2001. The Differential impact of Goal Congruency on attitudes, intentions, and transfer of brand equity, Nov,38. Mattiesen, Insa, Ian Phau, 2005. The Hugo Boss connection : Achieving Global Brand Consistency across country. Journal of Brand Management Mattila, A. S. and Enz, C.A, 2002. The role of emotions in service encounters, Journal of Services Research, 4(4), 268-277. Nelson, Susan, 2005. Beyond Branding, Journal of Brand Management. October. Netmeyer et al, 2004. Developing and validating measures of facets of customer-based brand equity, Journal of Business Research. Odin, Yorick et al, 2001. Conceptual and operational aspects of brand loyalty, An Empirical investigation, Journal of Business Research,53. Pappu, Ravi,2005. Consumer-based Brand equity: improving the measurement, Journal of Product and Brand management. Park Su Chan, and V, Srinivasan, 1994.A survey- based method for measuring and understanding Brand Equity and its extendibility, Journal of Marketing Research. Podsakoff, M. Philip et al, 2003. Common Method Biases in Behavioural Research: A critical Review of the Literature and Recommended Remedies. Journal of Applied Psychology, vol 88 Punj, Girish and Junyean Moon, 2002. Positioning Options for achieving brand association, A psychological categorization framework, Journal of Business research,55. Ramos, Angel F, Villuejo and Manuel J. Sanchez-Franco, 2005.The impact of Marketing Communication And Price Promotion on Brand Equity, Journal of Brand Management, August. Rio,A Ballen et al, 2001. The effect of brand associations on consumer response, Journal of Consumer marketing:18: 4/5. Van Riel et al, 2005. Marketing antecedents of industrial brand equity: An empirical investigation in specialty chemical. Industrial Marketing Management, 24. Washburn, JH. And Plank, RE, 2002. Measuring brand Equity: an evaluation of a consumer-based brand equity scale. Journal of Marketing Theory and Practice, Vol.10 no 1. pp,46-61. Yoo,B. and Donthu,N, 2001. Developing and validating a multidimensional consumerbased brand equity scale. Journal of Business Research, 52(1),1-14. Zeithaml, VA, 1988. Consumer perceptions of price, quality and value: a means-end model and synthesis of evidence. Journal of Marketing, Vol 52 No 3, pp. 2-22

20