Linking customer-based brand equity

with brand market performance:

a managerial approach
Ahmed H. Tolba
The American University in Cairo, New Cairo, Egypt, and
Salah S. Hassan
The George Washington University, Washington, DC, USA

Abstract
Purpose – The purpose of this paper is to attempt to bridge a gap in literature by defining and operationalizing specific brand equity constructs, and
testing the relationships between customer-based brand equity and brand market performance. Current literature has focused on building and
conceptualizing brand equity, there is no consensus on how it should be measured and what constructs should be included in the measurement process.
Design/methodology/approach – This study was conducted in two phases: a consumer-level analysis; and a brand-level analysis. A total of 6,410
observations have been identified (sample size consists of 5,598 usable observations). The second phase involved analyzing the data at the aggregate
brand level. Analysis included testing hypotheses on the correlations between customer-based brand equity constructs and brand market performance.
finally, detailed market and country-of-origin analyses are presented for managerial considerations.
Findings – Results from the consumer dataset have been averaged by brand (a total of 17 brands covering 76 percent and 75 percent of market shares
in both economy and luxury markets). At the consumer-level, structural equation modeling was conducted to test research hypotheses. Results varied
according to consumer usage. Attitudinal loyalty and satisfaction were found the strongest predictors of brand preference and intention to purchase. At
the aggregate brand level, correlation analyses supported the hypothesis that customer-based brand equity constructs are correlated with brand market
performance.
Practical implications – Analyses at the consumer and brand levels revealed interesting results about the US automotive market and suggested
important managerial considerations. Specific recommendations are offered in order to help companies prioritize their resource utilization and improve
their performance in the market.
Originality/value – This study offers a new model that links customer-based brand equity with brand market performance. It advances both academic
and practical findings, and opens the door for new streams of research that link academic models with practical applications. It advances specific
practical recommendations to companies and at the same time offers a reliable and valid academic model that could be applied on other industries and
countries.

Keywords Brand equity, Customer loyalty, Brand management

Paper type Research paper

Introduction perspective. Keller (1993) posited that companies are

motivated to study brand equity for two reasons: one is
The area of brand equity has received significant research financially-based to estimate the value of a brand more
attention in recent years. As a result, current marketing precisely for financial reporting purposes; and the other is
research studies attempt to conceptualize, measure, and strategy-based to improve marketing productivity. He argued
manage brand equity in a way that drives brand market that evaluating the brand in the minds of consumers is
performance, and helps firms in strategic decision making. An prerequisite for brand market performance.
emerging debate started to address whether brand equity While current literature has focused on building and
should be thought of from a consumer-oriented or a market conceptualizing brand equity, there has been no consensus on
performance-oriented perspective. Motameni and Shahrokhi how to measure it or on what constructs to include in the
(1998) recognized the confusion and disagreement in measurement process (Mackay, 2001a). This study attempts
conceptualizing and defining brand equity, and they to bridge this gap in the literature by defining and
identified two opposing perspectives or schools of thought: operationalizing specific brand equity constructs, and
the marketing perspective and the financial accounting developing an integrative brand equity model that links
customer-based brand equity with brand market
performance.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1061-0421.htm
Brand equity measurement models
Several academic studies attempted to measure brand equity
Journal of Product & Brand Management and offered different approaches and constructs to
18/5 (2009) 356– 366
q Emerald Group Publishing Limited [ISSN 1061-0421] incorporate in the measurement process. Erdem and Swait
[DOI 10.1108/10610420910981837] (2004) classified brand equity measurement models into:

356
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

.
component-based models (Aaker, 1991, 1996; Keller, knowledge equity in this model, attitudes and attachments
1993, Lassar et al., 1995; Keller and Lehmann, 2003); represent attitudinal equity and relationship equity
.
holistic models (Swait et al., 1993; Park and Srinivasan, respectively. Similarly, Vakratsas and Ambler (1999) defined
1994; Kamakura and Russell, 1993). consumer-based effects in terms of cognition (knowledge
equity), affect (attitudinal equity), and experience
While component-based models measure individual elements
(relationship equity). Also, they supported using
of brand equity, holistic models seek an overall evaluation of
“Preference,” and “Intention to purchase,” as a result of the
the brand. This study utilizes a component-based approach
three CBBE constructs.
whereby brand equity is thought of as a multidimensional
Knowledge equity (KE) is defined as the component of
concept (Keller, 1993) because the main purpose is to test the CBBE that evaluates consumers’ awareness of the brand
effect of each customer-based brand equity construct on (recognition and recall), and their familiarity with brand
brand market performance. characteristics, meaning, and functions. KE incorporates the
Aaker (1991) defined brand equity as “a set of five cognitive dimension in the minds of consumers as per the
categories of brand assets (liabilities) linked to a brand’s name “Hierarchy of effects model”, and measures how effective the
or symbol that add to (subtract from) the value provided by a brand message reached the target consumers.
product or service.” He identified five brand equity Attitudinal equity (AE) refers to consumers’ attitudes
constructs: towards a particular brand. AE incorporates the “affective”
1 brand awareness; dimension in the minds of consumers as per the Hierarchy of
2 brand perceived quality; Effects Model (Lavidge, 1961), and measures the
3 brand associations; effectiveness of the different marketing mix elements in
4 brand loyalty; and influencing consumer perceptions. Lassar et al. (1995)
5 other proprietary brands assets, such as patents, recognized the following constructs:
trademarks, and channel relationships. .
perceived quality, capturing the performance of the brand;
This definition has been utilized in various brand equity .
perceived value, capturing the utility and affordability of
empirical studies (Yoo and Donthu, 1997; Yoo et al., 2000; the brand; and
Washburn and Plank, 2002; Baldauf et al., 2003; Kim and .
social image (what is referred to in this study as Prestige),
Kim, 2004). Further, Keller (1993) defined “Customer-based capturing the social dimension.
brand equity” (CBBE) as “the differential effect of brand Further, Percy and Rossiter (1992) posited that brand
knowledge on consumer response to the marketing of the attitude has both cognitive and emotional dimensions. This
brand”. According to Keller, the differentiation effect is study proposes to add “affect” to attitudinal equity constructs
determined by comparing consumers’ reactions to the (which is supported by the Hierarchy of Effects Model) to
marketing of a brand with their reactions to same marketing capture the emotional dimension and render the scale more
of an unnamed version of the product. Keller highlighted two comprehensive.
brand equity constructs: brand knowledge and brand Relationship equity (RE) includes both customers’
response. Brand knowledge has been defined in terms of satisfaction with as well as their attitudinal loyalty towards
brand awareness and image, while brand response to the brand. RE incorporates the attachment dimension
marketing has been conceptualized in terms of consumer between consumers and the brand as per the Hierarchy of
perceptions, preferences, and behavior arising from marketing Effects Model, and measures the effectiveness of marketing
mix activities. Aaker (1996) introduced a new model for activities in building a relationship between the brand and its
measuring brand equity: “the Brand Equity Ten”. This model target consumers.
included customer-based brand equity constructs, such as Attitudinal loyalty has been defined as “the level of
awareness, associations, perceived quality, perceived value, commitment of the average consumer toward the brand”,
loyalty, and satisfaction, as well as market behavior measures, while behavioral loyalty has been defined as “the willingness
such as market share, market price and distribution coverage. of the average consumer to repurchase the brand” (Morgan,
This study adds to above-mentioned research by identifying 2000; Chaudhuri and Holbrook, 2001). Table I highlights the
the constructs to be included in a customer-based brand studies in the literature supporting the recommended
equity model, linking them to brand market performance, and constructs.
empirically testing this relationship on the US automotive
industry.
An integrative brand equity model
Several studies attempted to link customer-based brand
Proposed brand equity constructs
equity constructs with the “Hierarchy of Effects Model”, an
The proposed model breaks down the CBBE into three advertising effectiveness tool introduced by Robert Lavidge in
dimensions: 1961. The Hierarchy of Effects Model consists of three
1 knowledge equity (KE); components:
2 attitudinal equity (AE); and 1 cognitive;
3 relationship equity (RE). 2 affective; and
3 co-native (behavioral).
This classification has been supported by several studies in the
literature. First, Keller and Lehmann (2003) defined “Brand It assumes that potential buyers are new users who move from
knowledge” (what is referred to as CBBE in this study), as awareness to knowledge to liking to preference to intention to
consisting of awareness and associations, attitudes, and buy to actual purchase. More recent studies concluded that
attachments. While awareness is a main component of the adoption continuum could be heterogeneous and that

357
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Table I Proposed brand equity constructs

Construct Literature support
Knowledge equity (KE) Awareness and familiarity Lavidge (1961); Aaker (1991, 1996); Keller (1993); Agarwal and Rao (1996); Yoo and Donthu
(1997); Yoo et al. (2000); Mackay (2001b); Washburn and Plank (2002); Keller and Lehmann
(2003); Baldauf et al. (2003); Kim and Kim (2004)
Attitudinal equity (AE) Affect Lavidge (1961); Percy and Rossitier (1992); Chaudhuri and Holbrook (2001)
Prestige Lassar et al. (1995)
Perceived quality Aaker (1991, 1996); Lassar et al. (1995); Agarwal and Rao (1996); Yoo and Donthu (1997);
Yoo et al. (2000); Mackay (2001b); Washburn and Plank (2002); Baldauf et al. (2003); Kim and
Kim (2004)
Perceived value Lassar et al. (1995); Aaker (1996); Agarwal and Rao (1996); Mackay (2001b)
Relationship equity (RE) Satisfaction Aaker (1996)
Attitudinal loyalty Aaker (1991, 1996); Atilgan et al. (2005) Lassar et al. (1995); Yoo and Donthu (1997);
Chaudhuri (1999); Yoo et al. (2000); Chaudhuri and Holbrook (2001); Washburn and Plank
(2002); Baldauf et al. (2003); Kim and Kim (2004)

consumers do not have to follow all the steps in the Hierarchy .

usage: whether the respondent has tried the brand before
of Effects Model (Copland, 1963; Achenbaum, 1968; Assael or not;
and Day, 1968; Poczter, 1987). This argument calls for .
market: luxury versus economy sedan; and
segmenting consumers (cluster analysis) to group them based .
Country-of-origin of the car.
on their movement within the model steps. Figure 1 demonstrates the proposed integrative model.
Agarwal and Rao (1996) developed a model that links
brand equity to the Hierarchy of Effects Model. They focused
on single composite measures of brand equity, assessing the Research hypotheses
impact of individual measures on market share (Mackay,
2001b). They utilized the following brand equity constructs: Several studies in the literature support the relationships
awareness, familiarity, weighted attributes, value for money, knowledge equity has with brand preference and intention to
and overall quality of brand name. Customer-based brand purchase (in case of non-users) or behavioral loyalty (in case
equity has been thought of as a prerequisite for brand of users) (Poczter, 1987; Cobb-Walgren et al., 1995; Ghosh
preference, which in turn affects consumers’ intention to et al., 1995; Agarwal and Rao, 1996; Vakratsas and Ambler,
1999; Mackay, 2001a, b). Most of these studies assume that
purchase. Other empirical studies in the literature supported
consumer decisions are only rational (cognitive models):
the positive relationship between CBBE constructs and brand
preference and purchase intention (Cobb-Walgren et al., H1. Knowledge equity positively affects brand preference.
1995; Vakratsas and Ambler, 1999; Myers, 2003). The H2. Knowledge equity positively affects intention to
proposed model builds on the relationships proposed in these purchase/behavioral loyalty.
studies and links them with brand market performance Similarly, many studies support the relationships attitudinal
(BMP). Aaker (1996) included BMP constructs in the Brand equity constructs have with brand preference and intention to
Equity Ten model. He proposed three BMP constructs: purchase/behavioral loyalty. While by the Hierarchy of Effects
1 market share; Model (Lavidge, 1961) is the first study that links affect and
2 price premium; and preference, the Pure Affect Models support the relationships
3 distribution coverage. between attitudinal equity constructs and brand preference
Similarly, Chaudhuri (1999) developed a model that supports (Poczter, 1987; Agarwal and Rao, 1996; Vakratsas and
the impact of brand attitudes and brand loyalty on brand Ambler, 1999; Mackay, 2001a, b):
equity outcomes, defined as market share, price, and shelf H3. Attitudinal equity constructs positively affect brand
spacing. Further, Chaudhuri and Holbrook (2001) reference.
operationalized BMP in terms of market share and price H4. Attitudinal equity constructs positively affect intention
premium. Further, Keller (2001) defined “Macro” brand to purchase/behavioral loyalty.
strength in terms of market leadership and market share. Also,
Further, relationship equity constructs (satisfaction and
Baldauf et al. (2003) assessed their impact on profitability, attitudinal loyalty) positively affect brand preference and
market performance, customer value, and purchase intention. intention to purchase/behavioral loyalty (Cobb-Walgren et al.,
Finally, Keller and Lehmann (2003) introduced the “Brand 1995; Vakratsas and Ambler, 1999):
Value Chain” model, and emphasized the relationship
between the customer mindset (awareness, associations, H5. Relationship equity constructs positively affects brand
attitudes, attachments, and activity) and brand performance preference.
(price elasticity and premiums, cost structure, market share, H6. Relationship equity constructs positively affects
profitability, and expansion success). This study utilizes intention to purchase/behavioral loyalty.
market share as the main BMP measure. Finally, there are The seventh hypothesis tests the relationships between brand
moderating variables in the model: preference and intention to purchase/behavioral loyalty. These

358
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Figure 1 An integrative brand equity model

relationships are supported by the Hierarchy of Effects Model survey was constructed using Zoomerang software, an
(Lavidge, 1961; Poczter, 1987): internet-based survey tool. This study focused on two car
H7. Brand preference positively affects intention to segments: “Economy” sedans and “Luxury” sedans. Table III
purchase/ behavioral loyalty. details the scales utilized in the survey and their sources in the
literature.
Finally, this study aims to test four hypotheses, correlating Brand market performance (BMP) data has been collected
CBBE constructs, brand preference, intention to purchase, through the Automotive News web site (www.autonews.com).
and behavioral loyalty with BMP at the brand level: The most recent data on all car units sold and car prices in the
H8. There is a positive correlation between CBBE US market has been purchased from the web site.
Constructs and BMP. Three different models (or paths) have been identified
H9. There is a positive correlation between brand based on usage:
preference and BMP. 1 a model that includes observations of consumers who
H10. There is a positive correlation between intention to never tried a particular brand (never tried);
purchase and BMP. 2 a model that includes observations of consumers who
H11. There is a positive correlation between behavioral have just tried a particular brand (tried);
loyalty and BMP. 3 a model that includes observations of consumers who own
or owned a particular brand (owned).
Table II lists the literature supporting model relationships.
A total of four random samples have been drawn (two samples
per market). The first and second samples have been drawn
Data collection and sampling from the “Economy” market, each asking questions related to
An online survey was administered, targeting consumers in five economy car brands. Similarly the third and fourth
the USA who are capable and willing to buy a sedan car. This samples have been drawn from the “Luxury” market, each

Table II Literature supporting model relationships

Hypotheses Literature support
H1 and H2 Poczter (1987); Cobb-Walgren et al. (1995); Ghosh et al. (1995); Agarwal and Rao (1996); Vakratsas and Ambler (1999); Mackay
(2001a, b)
H3 and H4 Lavidge (1961); Poczter (1987); Cobb-Walgren et al. (1995); Agarwal and Rao (1996); Vakratsas and Ambler (1999); Mackay (2001a, b)
H5 and H6 Cobb-Walgren et al. (1995); Vakratsas and Ambler (1999)
H7 Lavidge (1961); Poczter (1987)
H8 Lavidge (1961); Poczter (1987); Cobb-Walgren et al. (1995); Agarwal and Rao (1996); Vakratsas and Ambler (1999)
H9 Poczter (1987); Vakratsas and Ambler (1999)
H10 Lavidge (1961); Poczter (1987)
H11 Lavidge (1961); Chaudhuri and Holbrook (2001)

359
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Table III Study scales

Construct Scale description Source a
Knowledge equity Three seven-point semantic differentials intended to measure a person’s familiarity Simonin and Ruth (1998) 0.80-0.94
with a specified brand name
Perceived quality Three seven-point semantic differentials measuring a person’s attitude toward the Keller and Aaker (1992) .0.70
quality of a specific brand
Perceived value Four seven-point Likert-type statements that assess the utility derived from the Sweeney and Soutar (2001) 0.80-0.91
perceived economic value of a particular product
Prestige Three seven-point Likert-type statements that measure how much a person Kirmani et al. (1999) 0.96
considers some specific object to be high class and exclusive
Affect Three seven-point Likert-type statements measuring the degree of positive affect a Chaudhuri and Holbrook 0.96
consumer has toward a brand (2001)
Satisfaction Three seven-point Likert-type items measuring the level of satisfaction a consumer Tsiros and Mittal (2000) 0.95
experiences with a product’s performance
Attitudinal loyalty Three five-point Likert-type statements attempting to capture consumer’s general Yoo et al. (2000) 0.90
loyalty to a specified brand
Brand preference Three five-point Likert-type statements measuring the degree to which a person Sirgy et al. (1997) 0.72-0.98
views a focal brand as preferable to a referent brand
Intention to buy Three seven-point Likert-type scale measuring the degree to which a consumer Putrevu and Lord (1992) 0.91
means to buy (or at least try) a specified brand in the future
Behavioral loyalty Three nine-point Likert-type scale, measuring the likelihood that a person will use Cronin et al. (2000) 0.93
an object again

asking questions related to five luxury car brands. The brands Consumer-level analysis
utilized in this study represent 76.3 percent and 74.5 percent Data analysis follows three steps:
of the economy and luxury sedan markets respectively. The 1 exploratory factor analysis to determine the appropriate
samples have been drawn using zSample, Zoomerang’s survey number of factors to be utilized in the analysis;
respondent service, which includes 2.5 million Zoomerang 2 confirmatory factor analysis for each of the three models
zSample participants (the study’s sampling frame). (never tried, tried, and owned) to test the adequacy of the
The target of this study was to collect at least 200 responses measurement models; and
per sample (a total of 800 responses). As a typical response 3 structural equation modeling for each of the three models,
rates for online survey ¼ 20 percent, A total of 4,000 surveys based on the results of the confirmatory factor analysis, to
(1,000 per sample) have been randomly distributed. This analyze relationships between latent constructs and test
study generated an overall response rate of 32 percent, research hypotheses.
yielding a total of 1,282 responses. While the measurement model specifies relationships between
The four samples have been distributed almost evenly
latent variables and their indicator variables, the structural
between males and females (49 percent females and 51
model specifies relationships between the latent constructs
percent males). The distribution of age and income across all
(Hatcher, 1994).
samples is reasonably representative of the study population.
First, exploratory factor analysis revealed that three
Further, the distribution of people across States has been
attitudinal equity constructs (affect, prestige and perceived
representative, yielding a Pearson Correlation Coefficient of
quality) loaded on the same factor. As a result, a new
0.96 between the actual population distribution and the
construct: Image was introduced, combining these three
sample distribution across States (significant at p ¼ 0.00).
constructs. Second, various tests were utilized to check the
Further, the different scales utilized to measure model
adequacy of the measurement models (confirmatory factor
constructs were analyzed to verify their reliabilities. While all
analysis), such as Chi square analysis, CFI (. 0.90), NNFI
scales have been driven from the literature, they have all been
(. 0.90), t-values of manifest variables (. 1.96), normalized
found reliable with Cronbach Alpha (a) ranging from 0.78 to
residuals distribution, composite reliabilities (. 0.70), and
0.98. According to Nunnally (1994), a scale of a . 0.7 is
variance extracted (.0.50). The results reveal that all three
considered reliable.
measurement models are adequate and fitting the data.
Finally, this study utilizes various tests to check the adequacy
of the structural equation models, such as Chi square analysis,
Results CFI (. 0.90), NNFI (.0.90), t-values of manifest variables
The data analysis process was conduced in two phases. First, (. 1.96), normalized residuals distribution, PNFI (. 0.60),
at the consumer-level, model relationships have been tested and Chi square difference. The results reveal that all
using structural equation modeling. Then, results have been structural models are adequate and fitting the data. The
averaged out by brand to form a brand dataset (consisting of following section details the results of the structural equation
20 brands total) in order to test the correlations between models for the three “Usage” cases: never tried, tried, and
model constructs and brand market performance. owned.

360
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Model 1: never tried The structural equation model reveals the following
For consumers who never tried a brand, Table IV details the regression equations:
results of the structural equation modeling.
The structural equation model reveals the following Preference ¼ 0:08 ðValueÞ þ 0:23 ðSatisfactionÞ
regression equations: þ 0:67 ðAttitudinal LoyaltyÞ þ 0:49 ½R2 ¼ 0:76

Intention to buy ¼ 0:79 ðAttituduinal LoyaltyÞ
Preference ¼ 0:34 ðValueÞ þ 0:23 ðImageÞ 0:84 ½R2 ¼ 0:29

þ 0:62 ½R2 ¼ 0:79
:
Intention to buy ¼ 0:49 ðPreferenceÞ þ 0:72 ½R2 ¼ 0:48
:

The results above show that that attitudinal loyalty is the

The results above show that value and image are the primary primary driver of both brand preference and intention to buy.
drivers of brand preference, which in turn affects intention to Further, the predictability of both regression equations is high
purchase (consistent with the Hierarchy of Effects Model). compared to the previous model (0.76 and 0.79 respectively).
However, the predictability of both regression equations is not This high predictability means that relationship equity is
high (0.29 and 0.48 respectively). Figure 2 demonstrates the crucial in predicting brand preference and intention to buy.
results of this model. Figure 3 demonstrates the results of this model.

Model 2: tried: Model 3: owned

For consumers who tried a particular brand, Table V details For consumers who own a particular brand, Table VI details
the results of the structural equation modeling. the results of the structural equation modeling.

Table IV SEM-never tried

Condition Result Evaluation
Chi square and p-value 1,231 (,0.0001) Results affected by sample size
Chi square/DF 14.6 (. 2) Results affected by sample size
CFI 0.94 (.0.90) Model fits the data
NNFI 0.93 (.0.90) Model fits the data
Manifest variable t-values and SE All . 1.96 Model fits the data
Normalized residuals distribution Centered @ zero, Symmetric Model is acceptable
PNFI 0.75 . (0.60) Model is parsimonious
Chi square difference test Chi square difference ¼ 21 df difference ¼ 4 Model’s fit is worse than the measurement model fit
Low predictability: explains 29 per cent and 48 per cent of the
Coefficients of determination PF: 0.29 INT: 0.48 variance in the data

Figure 2 Model 1 (never tried): final structural model

Table V SEM-tried
Condition Result Evaluation
Chi square and p-value 2,840 (,0.0001) Results affected by sample size
Chi square/df 16.2 (.2) Results affected by sample size
CFI 0.95 (. 0.90) Model fits the data
NNFI 0.94 (. 0.90) Model fits the data
Manifest variable t-values and SE All . 1.96 Model fits the data
Normalized residuals distribution Centered @ zero, Symmetric Model is acceptable
PNFI 0.77(.0.60) Model is parsimonious and simple
Chi square difference test Chi square difference ¼ 40 df difference ¼ 7 Model’s fit is worse than the measurement model fit
High predictability: explains 76 per cent and 79 per cent of the
Coefficients of determination PF: 0.76 INT: 0.79 variance in the data

361
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Figure 3 Model 2 (tried): final structural model

Table VI SEM-owned
Condition Result Evaluation
Chi square and p-value 1,063 (,0.0001) Results affected by sample size
Chi square/df 6.1 (. 2) Results affected by sample size
CFI 0.98 (.0.90) Model fits the data
NNFI 0.97 (.0.90) Model fits the data
Manifest variable t-values and SE All . 1.96 Model fits the data
Normalized residuals distribution Centered @ zero, Symmetric Model is acceptable
PNFI 0.81(. 0.60) Model is parsimonious and simple
Chi square difference test Chi square difference ¼ 32 DF difference ¼ 7 Model does not fit the data, but results affected by sample size
High predictability: explains 78 per cent and 91 per cent of the
Coefficients of determination PF: 0.78 INT: 0.91 variance in the data

The structural equation model reveals the following Brand-level analysis

regression equations: A correlation analysis has been conducted at the brand-level
because of the small sample size (20 brands). All research
Brand preference ¼ 0:14 ðSatisfactionÞ hypotheses have been supported. Results indicated that BMP
þ 0:77 ðAttitudinal LoyaltyÞ þ 0:47 ½R2 ¼ 0:78
had strong and significant correlation with knowledge equity,
relationship equity (satisfaction and attitudinal loyalty), brand
Behavioral Loyalty ¼ 0:37 ðSatisfactionÞ
preference, intention to purchase and behavioral loyalty. Only
þ 0:6302 ðAttitudinal LoyaltyÞ þ 0:31 ½R2 ¼ 0:91
attitudinal equity constructs (value and image) were not
found significantly correlated with BMP. Table VII
Similar to the previous model, attitudinal loyalty is the summarizes the results of the correlation analysis.
primary driver of both brand preference and intention to buy;
along with customer satisfaction. Also, the predictability of Effect of moderating variables
both regression equations is high (0.78 and 0.91 respectively). While “Usage” effect is interpreted within the structural
Figure 4 demonstrates the results of this model. equation modeling analysis, the other two moderating

Figure 4 Model 3 (owned): final structural model

362
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Table VII Correlation analysis results

KE Value Image SAT AL PF INT BL
BMP 0.718** 0.290 0.394 0.531* 0.704** 0.741** 0.485* 0.508*
Notes: * Significant at 0.05 level; ** significant at 0.01 level

variables (market and country-of-origin) are analyzed in this American (Dodge, Buick, Ford, Chrysler, Chevrolet, and
section in order to draw managerial implications. Cadillac); Japanese (Toyota, Nissan, Honda, Mazda, Acura,
Lexus, and Infiniti); and European (BMW, Audi, Mercedes,
Market analysis
and Volvo).
One of the major objectives of this study is to compare
An analysis of variance (ANOVA) has been conducted for
economy and luxury markets and possibly identify different
all metric variables in the model, analyzing the differences in
strategies to be deployed per market.
means across their countries of origin. ANOVA’s f-test
Independent samples t-tests have been conducted for all
indicated that, for all metric variables in this study, the means
metric variables in the model. Table VIII includes the means
of at least two countries of origin differ (the null hypothesis
of all metric variables per market as well as the significance of
that the three means are the same has been rejected at a
the differences for each construct.
It could be concluded that luxury consumers tend to be significance level of p , ¼ 0.00). Further, in order to identify
more knowledgeable about their car brands than economy which means differ and which country of origin is superior,
consumers. This could be attributed to the intense image- the Scheffe A-Posteriori test has been utilized. Table IX
based advertising for luxury car brands. Also, luxury details the findings of this test.
consumers perceive their brands’ image (quality, prestige, From Table IX, a series of interesting conclusions could be
and affect) significantly better than economy consumers. On drawn. First, Japanese cars are significantly behind American
the other hand, economy consumers rate their brands as and European cars in terms of knowledge equity (awareness,
providing higher value for money. Therefore, it could be recognition, and familiarity). This means that Japanese car
inferred that luxury consumers emphasize image, while brands have the potential to increase familiarity, which in turn
economy consumers emphasize value. might affect business performance. A further analysis by
Also, while luxury consumers are significantly more brand could identify more specific results.
satisfied with their car brands than economy consumers, the Second, European car brands enjoy a significantly better
latter tend to be more loyal. An explanation for this image (quality, prestige, and affect) than Japanese and
interesting finding is that luxury consumers are satisfied American car brands. This might be partially attributed to
with the excellent quality and image of their cars; yet they the fact that all selected European brands are in the luxury
tend to switch brands because of lack of differentiation. On market. Nevertheless, consumers’ perceptions indicate that
the other hand, economy consumers recognize that their cars European cars have higher quality, prestige, and likeability
are not of the best quality, thus their satisfaction with their than other car brands. Also, overall attitudinal equity (AE)
cars is not very high; yet, they are not willing to switch brands has been found significantly superior for European car
to avoid taking risk as they are happy with the current value brands.
offered. Third, Japanese car brands are perceived to offer a higher
value for money than American and European cars. This is an
Country-of-origin analysis expected conclusion, especially for the Economy market.
An important objective of this study is to test whether
Fourth, consumers are significantly more satisfied with
country-of-origin has an impact on CBBE constructs. The car
Japanese car brands, and accordingly, their attitudinal
brands covered in this study have three countries of origin:
loyalty is significantly higher. As a result, consumers have
higher relationship equity (RE) for Japanese cars than both
Table VIII Mean comparisons: independent samples T-tests European and American cars. Also, consumers’ intentions to
purchase or repurchase Japanese cars are significantly higher
Variable Economy Luxury Difference (E-L) Sig. ( p-value)
than European and American cars.
KE 5.49 5.78 2 0.29 0.00 Finally, while American cars enjoy a high level of knowledge
Quality 4.56 4.98 2 0.42 0.00 equity (KE) that is superior to Japanese cars, they suffer from
Value 4.50 4.10 þ 0.40 0.00 significantly low attitudinal equity (AE), relationship equity
Prestige 3.76 4.69 2 0.93 0.00 (RE), and intentions to purchase or repurchase. It could be
Affect 4.37 4.62 2 0.25 0.00 concluded that American car brands need to capitalize on the
Image 4.23 4.76 2 0.53 0.00 high KE and build a strong AE that would drive higher RE
AE 4.30 4.59 2 0.30 0.00 and lead consumers to purchase their brands. While
Preference 2.66 2.62 þ 0.05 0.21 European car brands clearly focus on Image and enjoy
Satisfaction 4.74 4.92 2 0.18 0.00 superior AE, they focus on the Luxury market, which is
Attitudinal loyalty 2.72 2.59 þ 0.13 0.01 characterized by low levels of loyalty. Japanese car brands, on
RE 3.73 3.75 2 0.02 0.60 the other hand, have the potential to further grow in the
Behavioral loyalty 5.29 5.17 þ 0.12 0.40 future by increasing KE. They emphasize value, and they are
Intention to buy 2.28 2.08 þ 0.20 0.00 successful in building superior RE, retaining customers, and
attracting new customers.

363
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Table IX Scheffe test for COO analysis

American – Japanese American-European Japanese-European
Variable Mean diff. Sig. Mean diff. Sig. Mean diff. Sig.
KE þ0.201* 0.000 20.096 0.095 20.296* 0.00
Quality 20.822* 0.000 21.108* 0.000 20.286* 0.00
Value 20.284* 0.000 þ0.416* 0.000 þ0.700* 0.00
Prestige 20.779* 0.000 21.705* 0.000 20.926* 0.00
Affect 20.742* 0.000 20.791* 0.000 20.048 0.75
Image 20.781* 0.000 21.201* 0.000 20.420* 0.00
AE 20.656* 0.000 20.796* 0.000 20.140* 0.01
Preference 20.097 0.053 þ0.016 0.954 þ0.113 0.10
Satisfaction 20.889* 0.000 20.724* 0.000 þ0.165 0.06
Attitudinal loyalty 20.459* 0.000 20.116 0.277 þ0.343* 0.00
RE 20.675* 0.000 20.422* 0.000 þ0.253* 0.00
Behavioral loyalty 22.029* 0.000 21.097* 0.000 þ0.932* 0.00
Intention to buy 20.387* 0.000 20.137* 0.049 þ0.250* 0.00
Note: * The mean difference is significant at the 0.01 level

Conclusions Further, analysis by market supported the above

conclusion. Economy consumers were found to have
This study advances an integrative brand equity model that stronger knowledge about the brands than luxury
links customer-based brand equity (CBBE) with brand consumers. Economy consumers rely on heavy research
market performance (BMP). CBBE constructs are identified before making this high involvement decision. Also, economy
from the literature, integrated in the proposed model, and consumers are price-sensitive and emphasize value, while
linked with brand preference and intention to purchase at the Luxury consumers appreciate the image of the luxury brands
individual consumer level. Exploratory factor analysis in the market. Interestingly, economy consumers tend to be
revealed that CBBE consists of five constructs: knowledge more loyal as they want to minimize risk, while luxury
equity (KE), perceived value, image (an aggregate of consumers are brand switchers due to the lack of
perceived quality, prestige and affect), satisfaction, and differentiation among luxury brands in the American market.
attitudinal loyalty. The model was validated using The final analysis conducted involved assessing the
confirmatory factor analysis, and the hypotheses were tested differences in consumers’ perceptions of the different
using structural equation modeling. countries-of-origin (American, European and Japanese). It
Results indicated that attitudinal loyalty and satisfaction was concluded that American brands are only strong in terms
(relationship equity constructs) are the primary drivers of of knowledge equity, while they are significantly inferior to
brand preference and intention to re-purchase among brand both European and Japanese brands in all other constructs.
users. This is an important finding for companies that need to Also, European brands are superior in terms of Image, while
emphasize loyalty programs and after-sale services, especially Japanese brands are the best in terms of offering value and
in the luxury market where brand switching is high. Among retaining customers. This is an alarming finding for American
non-users of the brands, value and image are the primary brands about their future in the market. In fact, this problem
drivers of brand preference, which in turn affects intention to has been magnified by the recent financial crisis. It is also
purchase. However, the model’s predictability (R2) is very important to recognize that Japanese brands are improving
low. Therefore, car companies in the USA must focus on test their image in the American market, and at the same time,
drives and consumer experience with the brands besides they are emphasizing relationship equity elements (loyalty and
normal image building activities. Further, findings showed satisfaction) which have been found to be the primary drivers
that brand market performance is significantly correlated with to preference and intention to purchase.
customer-based brand equity constructs.
An interesting finding revealed that value is the key driver of
brand preference among luxury consumers, while image is the Agenda for future research
main factor affecting brand preference among Economy This study opens the door for a series of research streams that
consumers. While this conclusion seems counter-intuitive, it could be useful in advancing knowledge in the area of brand
is attributed to the fact that luxury consumers are brand equity measurements. First, this study concludes that
switchers who perceive most luxury brands as somehow of satisfaction and attitudinal loyalty are the primary drivers of
similar image. Therefore, image is a “Point-of-parity”, while brand preference and intention to purchase. Future research
value is the “Point-of-difference” that would differentiate one should identify the antecedents of satisfaction and attitudinal
brand from another. Similarly, economy consumers recognize loyalty. It would be interesting to determine which marketing
the value offered by the different brands “Point-of-parity”, and branding activities would help increase consumer
but they appreciate the image “Point-of-difference” of some satisfaction and drive attitudinal loyalty in the US
of the brands and accordingly stick to them. automotive industry. Also, future research should attempt to

364
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

analyze brand image, and determine what aspects of image Erdem, T. and Swait, J. (2004), “Brand credibility, brand
should be emphasized, particularly in the luxury market. consideration, and choice”, Journal of Consumer Research,
Further, the models offered by this study should be Vol. 31 No. 1, pp. 191-8.
replicated on other industries in order to identify the primary Ghosh, A.K., Chakraborty, G. and Ghosh, D.B. (1995),
factors affecting preference, intention to purchase, and “Improving brand performance by altering consumers’
ultimately brand market performance. It would also be brand uncertainty”, Journal of Product & Brand
interesting to test the model on the automotive industry Management, Vol. 4 No. 5, p. 14.
internationally in order to assess cultural differences. This Hatcher, L. (1994), A Step-by-Step Approach to Using SAS
would help generalize the proposed model globally and drive
System for Factor Analysis and Structural Equation Modelling,
international marketing strategies.
SAS Institute, Cary, NC.
Finally, a longitudinal research should follow the proposed
Kamakura, W.A. and Russell, G.J. (1993), “Measuring brand
study to assess causal relationships between CBBE and BMP.
One of the limitations of this study is that BMP data precedes value with scanner data”, International Journal of Research in
CBBE data. This means that causality cannot be assessed at Marketing, Vol. 10, pp. 9-22.
the aggregate brand level. A longitudinal study might help Keller, K.L. (1993), “Conceptualizing, measuring, and
determine the time lag between consumers’ perceptions of a managing customer-based brand equity”, Journal of
certain brand, and the translation of these perceptions into Marketing, Vol. 57, January, pp. 1-22.
BMP results. Further, a time series analysis could be useful in Keller, K.L. (2001), “Branding and brand equity”, in Weitz, B.
tracking trends in each market, and offering a comprehensive and Wensley, R. (Eds), Handbook of Marketing, Sage
model that predicts BMP based on changes in consumer Publications, London.
perceptions. Keller, K.L. and Aaker, D.A. (1992), “The effects of
sequential introduction of brand extensions”, Journal of
References Marketing Research, Vol. 29 No. 1, p. 35.
Keller, K.L. and Lehmann, D.R. (2003), “How do brands
Aaker, D. (1991), Managing Brand Equity: Capitalizing on the create value?”, Marketing Management, Vol. 12 No. 3, p. 26.
Value of a Brand Name, Free Press, New York, NY. Kim, W.G. and Kim, H.-B. (2004), “Measuring customer-
Aaker, D. (1996), “Measuring brand equity across products
based restaurant brand equity: investigating the relationship
and markets”, California Management Review, Vol. 38 No. 3,
between brand equity and firms’ performance”, Cornell
pp. 102-20.
Achenbaum, A. (1968), “Relevant measures of consumer Hotel and Restaurant Administration Quarterly, Vol. 45 No. 2,
attitudes”, Proceedings of the ESOMAR Congress, Opatija. pp. 115-31.
Agarwal, M.K. and Rao, V.R. (1996), “An empirical Kirmani, A., Sood, S. and Bridges, S. (1999), “The
comparison of consumer-based measures of brand ownership effect in consumer responses to brand line
equity”, Marketing Letters, Vol. 7 No. 3, pp. 237-47. stretches”, Journal of Marketing, Vol. 63, January,
Assael, H. and Day, R. (1968), “Attitudes and awareness as pp. 88-101.
predictors of market share”, Journal of Advertising Research, Lassar, W., Mittal, B. and Sharma, A. (1995), “Measuring
Vol. 8 No. 4. customer-based brand equity”, Journal of Consumer
Atilgan, E., Aksoy, S. and Akinci, S. (2005), “Determinants Marketing, Vol. 12 No. 4, pp. 11-19.
of the brand equity: a verification approach in the beverage Lavidge, R.J. (1961), “A model for predictive measurements
industry in Turkey”, Marketing Intelligence & Planning, of advertising effectiveness”, Journal of Marketing, Vol. 25
Vol. 23 No. 3, pp. 237-48. No. 6, pp. 59-62.
Baldauf, A., Cravens, K.S. and Binder, G. (2003), Mackay, M.M. (2001a), “Evaluation of brand equity
“Performance consequences of brand equity management: measures: further empirical results”, Journal of Product &
evidence from organizations in the value chain”, Journal of Brand Management, Vol. 10 No. 1, p. 38.
Product & Brand Management, Vol. 12 Nos 4/5, pp. 220-34. Mackay, M.M. (2001b), “Application of brand equity
Chaudhuri, A. (1999), “Does brand loyalty mediate brand measures in service markets”, Journal of Services
equity outcomes?”, Journal of Marketing Theory and Practice, Marketing, Vol. 15 No. 3, pp. 210-21.
Vol. 7 No. 2, pp. 136-46. Morgan, R.P. (2000), “A consumer-oriented framework of
Chaudhuri, A. and Holbrook, M.B. (2001), “The chain of
brand equity and loyalty”, Journal of Market Research
effects from brand trust and brand affect to brand
Society, Vol. 42 No. 1, pp. 65-78.
performance: the role of brand loyalty”, Journal of
Motameni, R. and Shahrokhi, M. (1998), “Brand equity
Marketing, Vol. 65, April, pp. 81-93.
valuation: a global perspective”, Journal of Product & Brand
Cobb-Walgren, C.J., Ruble, C.A. and Donthu, N. (1995),
“Brand equity, brand preference, and purchase intent”, Management, Vol. 7 No. 4, p. 275.
Journal of Advertising, Vol. 24 No. 3, pp. 25-40. Myers, C.A. (2003), “Managing brand equity: a look at the
Copland, B. (1963), “An evaluation of conceptual impact of attributes”, Journal of Product & Brand
frameworks for measuring advertising results”, paper Management, Vol. 12 No. 1, pp. 39-51.
presented at the Advertising Research Foundation, 9th Nunnally, J.C. (1994), Psychometric Theory, 3rd ed., McGraw-
Annual Conference. Hill, New York, NY.
Cronin, J. Jr, Brady, M.K. and Hult, G.T. (2000), “Assessing Park, C.S. and Srinivasan, V. (1994), “A survey-based
the effects of quality, value, and customer satisfaction on method for measuring and understanding brand equity and
consumer behavioral intentions in service environments”, its extendibility”, Journal of Marketing Research, Vol. 31
Journal of Retailing, Vol. 79 No. 2, pp. 193-218. No. 5, pp. 271-88.

365
Linking customer-based brand equity with brand market performance Journal of Product & Brand Management
Ahmed H. Tolba and Salah S. Hassan Volume 18 · Number 5 · 2009 · 356 –366

Percy, L. and Rossiter, J.R. (1992), “A model of brand Tsiros, M. and Mittal, V. (2000), “Regret: a model of its
awareness and brand attitude advertising strategies”, antecedents and consequences in consumer decision
Psychology & Marketing, Vol. 9 No. 4, pp. 263-74. making”, Journal of Consumer Research, Vol. 26, March,
Poczter, A. (1987), “Attitude development hierarchy and pp. 401-17.
segmentation”, Review of Business, Vol. 9 No. 1, pp. 17-20. Vakratsas, D. and Ambler, T. (1999), “How advertising
Putrevu, S. and Lord, K.R. (1992), “Comparative and non- works: what do we really know?”, Journal of Marketing,
comparative advertising: attitudinal effects under cognitive Vol. 63 No. 1, pp. 26-43.
and affective involvement conditions”, Journal of Washburn, J.H. and Plank, R.E. (2002), “Measuring brand
Advertising, Vol. 23, June, pp. 77-90. equity: an evaluation of a consumer-based brand equity
Simonin, B.L. and Ruth, J.A. (1998), “Is a company known scale”, Journal of Marketing Theory and Practice, Vol. 10
by the company it keeps? Assessing the spillover effects of No. 1, pp. 46-62.
brand alliances on consumer brand attitudes”, Journal of Yoo, B. and Donthu, N. (1997), “Developing and validating a
Marketing Research, Vol. 35, February, pp. 30-42. consumer-based overall brand equity scale for Americans
Sirgy, J.M., Grewal, D., Mangleburg, T.F., Park, J.-o., Chon, and Koreans: an extension of Aaker’s and Keller’s
K.-S., Claiborne, C.B., Johar, J.S. and Berkman, H. conceptualizations”, Proceedings, AMA Summer Educator’s
(1997), “Assessing the predictive validity of two methods of Conference, Chicago.
measuring self-image congruence”, Journal of the Academy Yoo, B., Donthu, N. and Lee, S. (2000), “An examination of
of Marketing Science, Vol. 25 No. 3, pp. 229-41. selected marketing mix elements and brand equity”, Journal
Swait, J., Erdem, T., Louviere, J. and Dubelaar, C. (1993), of the Academy of Marketing Science, Vol. 28 No. 2,
“The equalization price: a measure of consumer-perceived pp. 195-211.
brand equity”, International Journal of Research in
Marketing, Vol. 10 No. 1, pp. 23-45.
Corresponding author
Sweeney, J.C. and Soutar, G.N. (2001), “Consumer
perceived value: the development of a multiple item Ahmed H. Tolba can be contacted at: ahmedtolba@
scale”, Journal of Retailing, Vol. 77, Summer, pp. 203-20. aucegypt.com

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints

366