Developing and validating a multidimensional consumer-based brand [605683]

Developing and validating a multidimensional consumer-based brand
equity scale
Boonghee Yooa,*, Naveen Donthub,1
aDepartment of Marketing and Business Law, G. R. Herberger College of Business, St. Cloud State University, St. Cloud, MN 56301, USA
bDepartment of Marketing, J. Mack Robinson College of Business, Georgia State University, Atlanta, GA 30303, USA
Received 1 January 1997; received in revised form 1 July 1999; accepted 6 August 1999
Abstract
Little systematic research has been done to develop a scale to measure consumer-based brand equity. The authors report the results of a
multistep study to develop and validate a multidimensional consumer-based brand equity scale (MBE) drawn from Aaker’s and Keller’sconceptualizations of brand equity. A total of 1530 American, Korean American, and Korean participants evaluated 12 brands from three
product categories (athletic shoes, film for cameras, and color television sets). Multistep psychometric tests demonstrate that the new brand
equity scale is reliable, valid, parsimonious, and generalizable across several cultures and product categories. The authors discuss theoreticaland practical implications of the study. D2001 Elsevier Science Inc. All rights reserved.
Keywords: Multidimensional brand equity scale; Multistep psychometric tests; Brand equity
In the past decade, researchers have focused a tremen-
dous amount of attention on the brand equity construct,
which refers to the incremental utility or value added to aproduct by its brand name. Using a consumer-based beha-vioral view of brand equity, we define brand equity as
consumers’ different response between a focal brand and an
unbranded product when both have the same level ofmarketing stimuli and product attributes. The difference inconsumer response may be attributed to the brand name anddemonstrates the effects of the long-term marketing investedinto the brand.
Researchers have found that a product’s brand equity
positively affects future profits and long-term cash flow
(Srivastava and Shocker, 1991), a consumer’s willingness
to pay premium prices (Keller, 1993), merger and acquisi-tion decision making (Mahajan et al., 1994), stock prices(Simon and Sullivan, 1993; Lane and Jacobson, 1995),sustainable competitive advantage (Bharadwaj et al., 1993),and marketing success (Ambler, 1997). Almost every
marketing activity works, successfully or unsuccessfully,to build, manage, and exploit brand equity (see Aaker,1991; Keller, 1993; Yoo, Donthu, and Lee, 2000). How-ever, despite this considerable amount of interest, research
that identifies and attempts to understand brand equity
phenomena has been hampered because there has been noagreement regarding what brand equity is and, more im-portant, how it should be measured.
Although several studies have examined brand equity,
their main focus was not on developing a brand equitymeasure. Currently, researchers use ad hoc measures such asprice premium (Aaker, 1991), conjoint analyzed value of the
brand name (Rangaswamy et al., 1993; Cobb-Walgren et al.,
1995), composite multiattribute weighted scores of thebrand name (Park and Srinivasan, 1994), a collection ofconsumer-based measures (Agarwal and Rao, 1996), and ascanner data-based measure (Kamakura and Russell, 1993).Other ad hoc measures include financial values of a brand,such as future earnings (Aaker, 1991), incremental cashflow (Simon and Sullivan, 1993), equalization price (Swait
et al., 1993), and momentum accounting-based value (Far-
quhar et al., 1991). However, these measures were devel-oped without rigorous psychometric tests, and they were notparsimonious enough to manage.
* Corresponding author. Tel.: +1-320-255-3161; fax: +1-320-255-
4061.
E-mail addresses : byoo@stcloudstate.edu (B. Yoo), ndonthu@gsu.edu
(N. Donthu).
1Tel.: +1-404-651-1043; fax: +1-404-651-4198.
0148-2963/01/$ – see front matter D2001 Elsevier Science Inc. All rights reserved.
PII: S 0148-2963(99)00098-3
Journal of Business Research 52 (2001) 1– 14

The purpose of this study is to develop a multidimen-
sional measure of consumer-based brand equity and assessits psychometric properties. Here, consumer-based means
measurement of cognitive and behavioral brand equity at
the individual consumer level through a consumer survey.In contrast, for firm-based measures, researchers collectfinancial market, accounting, and store-level scanner datawithout contacting consumers; these then identify dollar-metric and financial brand equity at the firm or brand level.Unlike most previous studies, this study develops a mea-sure of brand equity that is reliable, valid, and parsimo-
nious. In addition, the measure’s latent structure is assessed
for generalizability across multiple samples drawn fromseveral cultures, specifically, Koreans, Korean Americans,and Americans. In particular, our measure is developedusing the brand equity dimensions that Aaker (1991, alsosee 1996) and Keller (1993) suggest, which have beenpopularly accepted as valid and comprehensive. However,key questions about Aaker’s and Keller’s measurements
remain unanswered, specifically, the structural validity of
the measurement.
A consumer-based brand equity scale is beneficial in
several ways. First, it offers a means to test brand equitytheories. According to Aaker (1991), brand equity providesvalue to customers by enhancing their interpretation andprocessing of information, confidence in the purchase deci-sion, and satisfaction. Brand equity also provides value to
the firm by enhancing efficiency and effectiveness of
marketing programs, prices and profits, brand extensions,trade leverage, and competitive advantage. Similarly, Keller(1993) proposes that enhancing brand equity results in theability to command larger margins from consumers, elicitsincreased consumer information search, and improves mar-keting communication effectiveness, licensing opportu-nities, and consumers’ responsiveness to brand extensions.
A brand equity measure would allow investigation of the
role of brand equity in Aaker’s (1991) and Keller’s (1993)models. Specifically, it may be used to measure the brandequity of existing brands, then to examine the relationshipof brand equity to the resulting firm and consumer benefits.
Second, the measure would be more useful for testing
consumer-based brand equity theories than other previousmeasures have been. A consumer-based brand equity study
needs a measure that assesses an individual customer’s
brand equity. However, some of the previous measures aredesigned to measure brand equity of aggregate products atthe industry or firm level (e.g., Simon and Sullivan, 1993;Mahajan et al., 1994). Others measure an individual custo-mer’s brand equity (e.g., Rangaswamy et al., 1993; Swait etal., 1993; Park and Srinivasan, 1994; Cobb-Walgren et al.,1995), but the psychometric properties have not been
reported or fully analyzed. As a result, the previous mea-
sures may not be appropriate to studies that examineconsumer-based brand equity phenomena.
Many scale development studies take an emic approach,
in which a scale is first developed in one culture, thenvalidated or replicated in other cultures (see Netemeyer et
al., 1991; Durvasula et al., 1993). We develop a brandequity measure with an etic approach, in which a universal
measurement structure across cultures is sought using multi-
ple cultures simultaneously (Geertz, 1973). The outcomemeasure that an etic approach produces is functionally,conceptually, linguistically, and metrically equivalent acrosscultures, which provides the basis for generating valid cross-cultural comparisons (Berry, 1980; Leung and Bond, 1989;Meredith, 1993; Rosenzweig, 1994).
After surveying several cultures, we assess the univers-
ality of our brand equity measurement. Confirming the
universality enables cross-cultural benchmarking of brandequity, because the measure is not bound to a particularculture or country. Cross-cultural benchmarking is useful fortracking performance and developing strategies in the do-mestic and international markets, and in the cross-culturalstudy of brand equity.
We collect data from South Korea and the US because
these two countries show an adequate range of cultural
variation. According to Hofstede’s (1991) work, Koreansand Americans are different in every major cultural dimen-sion. In comparison with Americans, for example, Koreansrate high on Confucian dynamism (long-term orientation),low on individualism, and high on uncertainty avoidance.Therefore, South Korea is a good counterweight to the US,and Korean responses should reduce potential bias that
could result from developing and validating a measure using
samples drawn only from a Western culture.
In the following sections, we first examine the brand
equity construct and identify its relevant dimensions asproposed by Aaker (1991) and Keller (1993). Next, wediscuss the procedures used to generate and select scaleitems in the pilot and main studies. Then, we assess internalconsistency, validity, and cross-cultural metric equivalence.
Finally, we discuss the implications of the final scale to both
practitioners and researchers.
1. The brand equity construct
Brand equity has many definitions and forms, such as
favorable impressions, attitudinal dispositions, and beha-
vioral predilections (Rangaswamy et al., 1993); brand
loyalty, brand awareness, perceived quality, brand associa-tions, and other proprietary brand assets (Aaker, 1991);brand knowledge such as brand awareness and brandassociations (Keller, 1993); loyalty and image (Shockerand Weitz, 1988); the added value endowed by the brandname (Farquhar et al., 1991); incremental utility (Kama-kura and Russell, 1993); the difference between overall
brand preference and multiattributed preference based on
objectively measured attribute levels (Park and Srinivasan,1994); and overall quality and choice intention (Agarwaland Rao, 1996). One important consensus among thedefinitions is that brand equity is the incremental value
B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 2

of a product due to the brand name (Srivastava and
Shocker, 1991).
Collectively, brand equity consists of four dimensions:
brand loyalty, brand awareness, perceived quality of brand,
and brand associations, as proposed by Aaker (1991, 1996)and Keller (1993). These dimensions may be used toexplore the findings of marketing and consumer behaviorresearch in relation to brand equity (see Barwise, 1993);thus, we develop a brand equity measure that capitalizes onthese dimensions.
Aaker (1991, p. 39) defines brand loyalty as ‘‘the
attachment that a customer has to a brand.’’ In our study,
brand loyalty refers to the tendency to be loyal to a focal
brand, which is demonstrated by the intention to buy thebrand as a primary choice (Oliver, 1997). In contrast, someprevious research has focused on the behavioral aspects ofbrand loyalty (e.g., Guadagni and Little, 1983; Gupta,1988). Brand awareness is ‘‘the ability for a buyer to
recognize or recall that a brand is a member of a certain
product category’’ (Aaker, 1991, p. 61). Thus, brand aware-
ness consists of both brand recognition and recall (Rossiterand Percy, 1987; Keller, 1993). Perceived quality is ‘‘the
consumer’s judgment about a product’s overall excellenceor superiority’’ (Zeithaml, 1988, p. 3). It therefore is basedon consumers’ or users’ (i.e., not managers’ or experts’)subjective evaluations of product quality. Aaker (1991, p.109) defines brand associations as ‘‘anything linked in
memory to a brand’’ and brand image as ‘‘a set of [brand]
associations, usually in some meaningful way.’’ The asso-ciations have a level of strength (Aaker, 1991; Aaker andKeller, 1990; Keller, 1993), and a link to a brand will bestronger when it is based on many experiences or exposuresthan when it is based on few (Aaker, 1991).
2. Item generation
After careful consideration of the literature and on the
basis of the definitions previously established, we generateda pool of 48 candidate scale items to reflect the dimensionsof brand equity. To establish content validity, the items wereevaluated for conformity to the theoretical definitions andredundancy. After screening the items independently, then
jointly, we retained 22 items for initial psychometric assess-
ment: five on brand loyalty, four on brand awareness, sevenon perceived quality, and six on brand associations.
We designed five brand loyalty items to capture the
overall attitudinal loyalty to a specific brand rather thandirectly measuring actual brand-loyal behavior (see Gua-dagni and Little, 1983; Gupta, 1988). We adopted andmodified Beatty and Kahle’s (1988) brand loyalty items.
Brand awareness may be measured as brand recognition or
brand recall. We measured simple brand recognition ratherthan brand recall. Four items were based on previousresearch (Srull, 1984; Alba and Hutchinson, 1987; Rossiterand Percy, 1987).We designed perceived quality items to assess consu-
mers’ subjective judgment about a brand’s overall excel-lence (Zeithaml, 1988) because perceived quality represents
overall quality rather than individual elements of quality
(Petroshius and Monroe, 1987; Aaker and Keller, 1990;Boulding and Kirmani, 1993). We adopted seven of theitems used by Dodds et al. (1991). We designed brandassociations items to measure ‘‘the strength of connection toa brand node as a function of both the amount or quantity ofprocessing the information received at encoding and thenature or quality of the processing of the information
received at encoding’’ (Keller, 1993, p. 5). On the basis of
this definition, we developed six items for the overallperceptual strength of brand associations.
2.1. Stimuli selection
We selected three product categories as stimuli: film for
cameras (a low-cost, fast-replacement cycle, short-term
experience good), athletic shoes (a medium-cost, med-
ium-replacement cycle, medium-term experience good),and color television sets (a high-cost, slow-replacementcycle, longer-term experience good). An experience goodis a product that consumers must actually experience,through consumption or purchase, to judge its quality(Nelson, 1974). Three criteria guided the selection ofthese product categories. First, the categories were differ-
ent in price range, frequency of purchase, consumers’
product involvement, and consumption situation (e.g.,place, time, and interaction among users). This widevariety of products was selected to enhance the assess-ment of the cross-product applicability of the scale.Second, the participants in this study, college studentconsumers, were familiar with the categories. Their ex-perience with the products enabled them to provide
reliable and valid responses to a questionnaire. For ex-
ample, in both Korea and the US, more than 90% of theuniversity student consumers purchase athletic shoes.Annual market reports by research firms in Korea andthe US have shown that a significant number of bothKorean and American students experience these productcategories. Third, only consumer goods were selectedbecause of their high brand equity. Consumer goods’
brand value variance is explicitly attributable to the brand
equity variance (Simon and Sullivan, 1993).
We surveyed 12 brands: six athletic shoes (Adidas, Asics,
LA Gear, Nike, Puma, and Reebok), four films (Agfa, Fuji,Kodak, and Konica), and two color television sets (Samsungand Sony). We used two criteria in selecting the brands.First, the brands had to be available in both Korea and theUS. Second, the brands had to be markedly different.
According to the market reports mentioned previously, the
market share of the selected brands in the Korean andAmerican markets ranged from 0.5% to 12.4% in athleticshoes, 0.5% to 41.2% in film, and 0.1% to 37.2% intelevision sets. We selected brands that held the highest,B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 3

lowest, and mid-level market shares. In addition, we se-
lected brands with various countries of origin. The US is thecountry of origin for LA Gear, Nike, Reebok, and Kodak;
Germany for Adidas, Puma, and Agfa; Japan for Asics, Fuji,
Konica, and Sony; and South Korea for Samsung.
3. Item purification: the pilot study
3.1. Subjects
The pilot sample was composed of 460 undergraduate
university students, 230 from South Korea and 230 from theUS. Elimination of incomplete data resulted in 414 useablesurveys, 218 Koreans (117 men, 101 women) and 196Americans (111 men, 85 women).
Researchers have suggested that the use of student
subjects in measurement development research threatensthe external validity and generalizability of findings due to
the non-representativeness and unique characteristics of the
population (e.g., Burnett and Dunne, 1986; Wells, 1993).However, in a cross-cultural study, well-matched (i.e.,maximally homogeneous within and between cultures)samples are more useful than representative samples be-cause they allow more exact theoretical predictions andreduce the confounding effects of other factors (Hofstede,1991). More important, students are accepted for theory
testing research in which the multivariate relationships
among constructs, not the univariate differences (i.e., meanscore comparisons) between samples, are being investi-gated (Calder et al., 1981). Students have been effectivesurrogates for non-students or adults in various empiricalstudies that have examined, for example, warranty andconsumers’ risk perceptions (Shimp and Bearden, 1982);country of origin, attitude toward advertising, and attitude–
preference relationship (Yavas, 1994); product warnings
and safe behavior (Cox et al., 1997); price–quality percep-tions (Lichtenstein and Burton, 1989); dissonance reduc-tion behavior (Sheth, 1970); self-identity acculturation(Owenbey and Horridge, 1998); and hiring recommenda-tions (Olian et al., 1988).
3.2. Instrumentation
We developed the questionnaire in English. Three other
marketing researchers and an English-language professorexamined its wording and the face validity of the questions.Next, two bilingual experts fluent in both English andKorean translated the questionnaire into Korean. The verbalequivalence between the Korean and English versions waschecked through back-translation with the help of two other
bilinguists in Korea (Douglas and Craig, 1983). This process
was repeated until we agreed that the Korean questionnairewas adequately translated with compatible meaning.
In the pilot study, we used only four existing brands of
athletic shoes as brand stimuli: Adidas, Nike, and Reebokfor both Koreans and Americans, and one local brand,
Converse or Prospecs, for Americans or Koreans, respec-tively. We prepared a different version of the questionnaire
for each brand surveyed.
3.3. Procedures and measures
Different versions of the questionnaire were assigned
randomly to participants. The participants were told thepurpose of the study was to provide managers with newinsights about effective ways to manage brands successfully.
Approximately 60 participants were assigned to each of the
four different versions of the questionnaire. The participantswere asked to indicate any unclear or uncomfortable word-ing in the questions. The questionnaire consisted of items tomeasure the dimensions of brand equity, as well as demo-graphic questions. Brand equity items were evaluated withfive-point Likert scales anchored at 1 = ‘‘strongly disagree’’and 5 = ‘‘strongly agree.’’
3.4. Analysis and results
To select the items that would enter the main study, we
computed the reliability of the items of each construct. Wedropped any items of a construct with low correlation to thecomposite variable, retaining only those that had a 0.70 orhigher Cronbach’s acoefficient for both samples (see
Nunnally and Bernstein, 1994). Next, to obtain reasonable
parsimony, when two items contributed similar Cronbach’sacoefficients, we dropped the weaker item. Also, we
reworded several items to enhance clarity, based on theparticipants’ comments. With this process, we selected sixitems for perceived quality, three for brand loyalty, three forbrand awareness, and five for brand associations. Wereevaluated these items in the main study.
4. Data collection and measures: the main study
In the main study, we administered the questionnaire to
Korean, Korean American, and American samples. Follow-ing a process recommended by Anderson and Gerbing(1988), Nunnally and Bernstein (1994), and others, we
conducted a series of exploratory and confirmatory factor
analyses on the data. Our goal was to identify a final set ofitems with acceptable discriminant and convergent validity,internal consistency reliability, parsimony, and cross-cultur-al metric equivalence. We analyzed data at an individuallevel, a multigroup level, and a pooled level.
4.1. Subjects
Undergraduate students at major universities in South
Korea and the US participated. A total of 650 Koreanparticipants in South Korea and 350 Korean American and650 American participants in the US answered the question-B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 4

naire. After invalid responses, such as non-American citizens
in the American sample, were excluded, the self-adminis-tered questionnaires yielded 1530 eligible responses: 633
Koreans (435 men and 198 women), 320 Korean Americans
(173 men and 147 women), and 577 Americans (272 men and305 women).
We included Korean American participants to examine
the validity of the sample selection. In the process of theiracculturalization into the host (i.e., American) culture,Korean Americans are theorized to show some transitionfrom Korean to American cultural and demographic
characteristics (Wallendorf and Reilly, 1983; Deshpande
et al., 1986). In that sense, the culture of Korean Amer-icans is a hybrid of Korean and American culture. Theanalysis showed that Korean Americans (54.1%) hadmore male respondents than Americans did (47.1%) butfewer male respondents than Koreans did (68.7%). KoreanAmericans (48.1%) had part-time or full-time jobs moreoften than Koreans did (19.3%) but less frequently than
Americans did (47.1%). In addition, household size
among Korean Americans (4.0 members) was greater thanamong Americans (3.2 members) but smaller than amongKoreans (4.2 members).
4.2. Instrumentation and procedures
In addition to the brand equity items, the question-
naire included items of other constructs: attitude toward
brand and purchase intention, product category involve-ment and experience, and brand purchase experience. Wedeveloped 12 versions of the questionnaire for the 12brands surveyed in the three product categories. Acrossversions, we maintained the same format and order ofquestion items.
American and Korean American participants answered
the English questionnaire, whereas Korean participants
answered the Korean questionnaire for their languageconvenience. Participants viewed one randomly assignedversion of the questionnaire and then rated their brandequity evaluations for the particular brand in the question-naire. Almost the same number of responses for eachversion was obtained, and there was no significant differ-ence in the number of responses among the 12 versions.
The number of responses per version ranged from 47 to 57
for Koreans, from 18 to 35 for Korean Americans and from40 to 52 for Americans.
4.3. Measures
Brand and product category experiences were measured
with yes or no items. The item measuring product category
purchase experience was ‘‘Have you ever bought any
brand of product category X?’’; for brand purchase ex-perience, ‘‘Have you ever bought brand X?’’; and forusage and ownership, ‘‘Do you currently use/own anybrand of product category X?’’ Subsequently, brand equityitems were measured as in the pilot study. Next, brand
purchase intention, brand attitude, and product categoryinvolvement were measured as follows: Purchase intention
was measured as ‘‘I would like to buy X’’ and ‘‘I intend to
purchase X’’ (five-point scale). Attitude toward brand wasmeasured with five-item scales of ‘‘very bad/very good,’’‘‘very nice/very awful,’’ ‘‘very attractive/very unattrac-tive,’’ ‘‘very desirable/very undesirable,’’ and ‘‘extremelylikable/extremely unlikable.’’ Product category involve-ment was measured with four five-point items: ‘‘I am veryinvolved with product category X,’’ ‘‘I use (wear) product
category X very often,’’ ‘‘I am a product category X
expert,’’ and ‘‘I am not interested in [the product cate-gory].’’ The brand equity items were asked before theitems regarding brand attitude and purchase intention toreduce the halo effect common to multiattribute attitudemodels, in which subjects distort their perceptions whenexpressing their overall attitudes before they evaluate de-tails that contribute to the attitudes (Beckwith and Leh-
mann, 1975; Cooper, 1981).
4.3.1. Product experiences
Data show that a significant number of the participants
had experienced the product categories. Those who hadpersonally purchased athletic shoes, film, and color televi-sion sets were 95%, 91%, and 83%, respectively, ofAmericans; 92%, 90%, and 64% of Korean Americans;
and 93%, 96%, and 16% of Koreans. Active users of
athletic shoes, film, and color television sets were 93%,72%, and 92%, respectively, of Americans; 88%, 78%, and79% of Korean Americans; and 90%, 86%, and 95% ofKoreans. Participants who had purchased the athletic shoebrands ranged from 36% (Asics) to 94% (Nike) of Amer-icans, from 24% (LA Gear) to 88% (Nike) of KoreanAmericans, and from 26% (Puma) to 65% (Nike) among
Koreans. For film, it ranged from 7% (Agfa and Konica) to
96% (Kodak) of Americans, 9% (Agfa) to 95% (Kodak) ofKorean Americans, and 76% (Konica) to 94% (Kodak) ofKoreans. For television sets, it was 10% (Samsung) and36% (Sony) of Americans, 36% (Samsung) and 75%(Sony) of Korean Americans, and 38% (Samsung) and2% (Sony) of Koreans. This rich product experienceindicates that the participants might have developed strong
attitudes toward the brands (Fazio and Zanna, 1981; Smith
and Swinyard, 1983).
The four-item measure of product category involvement
showed acceptable reliability, 0.82, 0.85, and 0.64 forAmericans, Korean Americans, and Koreans, respectively.Product category involvement was significantly differentamong the selected products: 2.41, 3.08, and 2.66 ( F=
32.78, p< 0.0001) for film, athletic shoes, and television
sets, respectively, for Americans; 2.35, 2.94, and 2.69 ( F=
10.80, p< 0.0001) for Korean Americans; and 2.21, 2.71,
and 2.76 ( F= 46.33, p< 0.0001) for Koreans. This wide
variety of involvement strengthens the generalizability ofthe scale.B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 5

5. Analysis and results
We conducted three levels of analyses to develop a brand
equity measure (see Durvasula et al., 1993). First, we
performed an individual analysis to determine whethercommon items and dimensions were found in each sample.Second, we conducted a multigroup analysis to examinefactorial invariance of the items selected in the individualanalysis (Jo ¨reskog and So ¨rbom, 1993; Bollen, 1989). The
factorial invariance becomes the basis of cross-culturalcomparisons of the constructs (Steenkamp and Baumgart-
ner, 1998). Third, we conducted a pooled analysis to
identify culture-free universal dimensions of brand equityin the pooled sample (Leung and Bond, 1989). Discoveringthe same dimensions in the individual, multigroup, andpooled analyses supports the universality of the dimensionsacross samples. We conducted an O-factor analysis of the 17
by 17 correlation matrix for the 17 brand equity itemscorrelated across the 12 brands (Rummel, 1970).
5.1. Individual analysis
We explored whether a similar pattern of internal con-
sistency and dimensionality among brand equity items couldbe found in each of the three samples. We selected items foreach construct until no higher reliability could be achievedand obtained the same items across samples, five for
perceived quality and three each for brand loyalty, brand
associations, and brand awareness. These fourteen selecteditems showed excellent reliability: perceived quality with0.92, 0.90, and 0.84 for Americans, Korean Americans, andKoreans, respectively; brand loyalty with 0.88, 0.86, and0.87; brand associations with 0.83, 0.79, and 0.78; andbrand awareness with 0.93, 0.91, and 0.84. However,exploratory factor analysis did not produce four distinct
factors among the selected brand equity items, mainly
because of the inseparability of brand awareness and brandassociations. Only three factors (i.e., perceived quality,brand loyalty, and brand awareness/associations) were con-sistently found in each sample.
To examine the dimensionality of brand equity, we
established three sets of measurement models, one-, three-,and four-dimensional, where the four-dimensional model
comprises brand loyalty, perceived quality, brand aware-
ness, and brand associations. We used confirmatory factoranalysis and estimated the models through the LISREL 8maximum likelihood method (Jo ¨reskog and So ¨rbom, 1993).
We then compared 2 four-dimensional models: Model 1, inwhich the correlation between brand awareness and brandassociations is specified to be free, and Model 2, in whichthe correlation is constrained to be equal with unity. Ac-
cording to the results of these methods, brand awareness and
associations should be combined due to a lack of discrimi-nant validity when either (1) there is no significant c
2
difference between the two models (Bagozzi, 1980; Burnkr-
ant and Page, 1982; Anderson and Gerbing, 1988) or (2) thesquared correlation between awareness and associations is
greater than the average variance extracted for awarenessand associations (Fornell and Larker, 1981).
As we report in Part A of Table 1, the c
2difference test
between Models 1 and 2 was significant in each sample(c
2
degrees of freedom [ d.f.]=1= 115.72 for Americans, 92.12 for
Korean Americans, and 38.22 for Koreans). As shown inTable 2, however, the correlation between awareness andassociations was very high in Model 1 (0.88 for Americans,0.80 for Korean Americans, and 0.90 for Koreans). Accord-ingly, the squared correlation (0.77, 0.64, and 0.81 for
Americans, Korean Americans, and Koreans, respectively)
was greater than the average variance extracted for aware-ness (0.82, 0.77, and 0.64) and associations (0.64, 0.58, and0.54), as reported in Part B of Table 1. Therefore, despitepassing the c
2test, by failing the variance comparison test,
discriminant validity between brand awareness and associa-tions was not proven in each sample.
By combining these two dimensions, we created a new
measurement model, Model 3, composed of the three
dimensions of brand loyalty, perceived quality, and brandawareness/associations. To examine further whether com-bining awareness and associations is better than combin-ing any possible pair of the dimensions, we compared thefit of Model 3 with the fit of all remaining three-dimen-sional models (Models 4 through 8). Table 1 (Part A)shows that the c
2fit index of the other models was worse
than that of Model 3. To investigate whether three-
dimensionality is desirable, we also compared Model 3with the one-dimensional model, Model 9, in which allfour dimensions are combined into one dimension. But thec
2fit index of Model 9 was significantly poorer than any
other previous model. Therefore, we used Model 3 insubsequent analyses.
Model 3 was also supported by other values of fit (see
Table 1). Its goodness-of-fit index (GFI) was 0.87, 0.87, and
0.94 for Americans, Korean Americans, and Koreans,respectively, and the adjusted goodness-of-fit index (AGFI)was 0.82, 0.82, and 0.92. Comparative goodness of fitindexes in both comparative fit index (CFI) and incrementalfit index (IFI) were 0.93, 0.91, and 0.95. The standardizedroot mean square residual (SRMR) was 0.061, 0.060, and0.054. These fit indexes indicate an excellent level of fit of
the model (see Hu and Bentler, 1999). In addition, the
loadings of the items to their corresponding dimensionsranged from 0.58 to 0.87 for Americans, 0.50 to 0.93 forKorean Americans, and 0.63 to 0.94 for Koreans. Thesmallest t-value of the loadings was 9.31 across samples,
which indicates high significance of the loadings. Thecomposite reliability estimates, which are internal consis-tency reliability measures as evidence of convergent validity
(Fornell and Larker, 1981) were acceptable, ranging from
0.88 to 0.92 for Americans, 0.86 to 0.90 for KoreanAmericans, and 0.84 to 0.89 for Koreans. The averagevariance extracted for each dimension was greater than thesquared correlation between the dimension and any otherB. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 6

dimension and 0.50, which indicates the independence of
the dimensions (Fornell and Larker, 1981). In summary,brand loyalty, perceived quality, and brand awareness/asso-
ciations make three reliable and valid dimensions of brand
equity in each sample.
5.2. Multigroup analysis
We subsequently performed an invariance test for Model
3, which is useful in examining the equivalence of afactorial measurement or structure model across multiple
samples (see Bollen, 1989; Durvasula et al., 1993; Jo ¨reskogand So ¨rbom, 1993). In particular, we tested whether the
factor structure (i.e., item loadings to factors) was statisti-cally invariant among the three samples by comparing an
unconstrained and a constrained model. In the unconstrained
model, the factor structure is specified to vary acrosscultures, whereas the factor structure is constrained to bethe same across cultures in the constrained model. When thec
2fit difference between these models is insignificant, the
factor structure is invariant across samples. As Table 3shows, the LISREL likelihood maximum method of estima-tion produced a c
2fit index of 1124.49, with 222 d.f.for the
unconstrained model.Table 1
Dimensionality and internal consistency of the brand equity scale
d.f.= degrees of freedom, SRMR = standardized root mean square residual, GFI = goodness-of-fit index, AGFI = adjusted goodness-of-fit index, CFI =
comparative fit index, IFI = incremental fit index, and VE = variance explained for.
Part A: Dimensionality
Americans Korean Americans Koreans
c2of the four-dimensional model
Model 1: fawareness:associations = free ( d.f.= 71) 391.58 212.84 223.24
Model 2: fawareness:associations =1( d.f.= 72) 507.30 304.96 261.46
c2of the three-dimensional model (d.f. = 74)
Model 3: Awareness and associations are combined 533.09 317.81 273.59
Model 4: Brand loyalty and perceived quality are combined 1143.35 549.37 1028.03
Model 5: Brand loyalty and associations are combined 1279.53 694.33 1089.00Model 6: Perceived quality and associations are combined 1302.37 679.41 1153.37Model 7: Brand loyalty and awareness are combined 1357.37 733.28 1129.49
Model 8: Perceived quality and awareness are combined 1999.40 946.51 1426.86
c
2of the one-dimensional model (d.f. = 77)
Model 9 2906.09 1533.59 2269.89
Fit statistics
Model 3: Three-dimensional model Model 1: Four-dimensional model
Americans Korean Americans Koreans Americans Korean Americans Koreans
c2 533.09 317.81 273.59 391.58 212.84 223.24
d.f. 74 74 74 71 71 71
SRMR 0.061 0.060 0.054 0.052 0.049 0.051
GFI 0.87 0.87 0.94 0.90 0.91 0.95AGFI 0.82 0.82 0.92 0.85 0.87 0.93CFI 0.93 0.91 0.95 0.95 0.95 0.96
IFI 0.93 0.92 0.95 0.95 0.95 0.96
Part B: Internal consistency
Americans Korean Americans Koreans
Composite a Coefficient a VE Composite a Coefficient a VE Composite a Coefficient a VE
Model 3: Three-dimensional model
Brand loyalty 0.88 0.88 0.71 0.86 0.86 0.68 0.87 0.87 0.69Perceived quality 0.92 0.92 0.69 0.90 0.90 0.68 0.84 0.84 0.53Awareness/associations 0.92 0.92 0.67 0.89 0.89 0.59 0.89 0.88 0.58
Model 1: Four-dimensional model
Brand loyalty 0.88 0.88 0.71 0.86 0.86 0.68 0.87 0.87 0.69
Perceived quality 0.92 0.92 0.69 0.90 0.90 0.65 0.84 0.84 0.53
Awareness 0.93 0.93 0.82 0.91 0.91 0.77 0.84 0.84 0.64Associations 0.84 0.83 0.64 0.79 0.79 0.58 0.77 0.78 0.54B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 7

For the constrained model, the fit was c2
d.f. = 250 =
1215.63. Thus, the fit difference was not insignificant(Dc
2
d.f.=2 8= 91.24, p< 0.0001). This result suggests that
the factor structure is not invariant across cultures; therefore,brand equity evaluations cannot be compared meaningfullyacross samples due to lack of cross-cultural metric equiva-lence (Steenkamp and Baumgartner, 1998).
To locate the source of inequality within the specified
matrix and discover an invariant measurement model acrosscultures, we conducted a partial measurement invariancetest, as suggested by Byrne et al. (1989). We tested theinvariance of each factor loading by declaring the loadingalone to be invariant across samples. To detect the invar-iance, we conducted the c
2difference test with the uncon-strained model of c2
d.f.= 222 = 1124.49. The test revealed
that the metric inequivalence occurred because of threeperceived quality items (i.e., QL1, QL4, and QL5) andone awareness item (i.e., AW1). By excluding these fouritems and combining the remaining 10 invariant items, wegenerated an etic measure of brand equity. The 10 itemscomprise three brand loyalty items, two perceived quality
items, and five brand awareness/associations items (see
Appendix A). The unconstrained model ( c
2
d.f. =9 6 =
424.65), in which the 10 factor loadings were specified tovaryacross cultures, was compared with the constrainedmodel ( c
2
d.f.=1 1 6 = 455.66), in which the 10 factor loadings
were constrained to be invariant across cultures. The c2
difference was not significant ( Dc2
d.f. =2 0 = 31.01, p>Table 2
Intercorrelations among dimensions
A = Americans, KA = Korean Americans, and K = Koreans.
Model 1: Four-dimensional model
Brand loyalty Perceived quality Awareness AssociationsAK A KA K A K AK A KA K A K
Brand loyalty 1 1 1
Perceived quality 0.50 0.52 0.38 1 1 1
Awareness 0.30 0.15 0.28 0.45 0.38 0.18 1 1 1
Associations 0.38 0.29 0.30 0.52 0.37 0.29 0.88 0.80 0.90 1 1 1
Model 3: three-dimensional model
Brand loyalty Perceived quality Awareness/Associations
AK A KA K A K A K A K
Brand loyalty 1 1 1
Perceived quality 0.50 0.52 0.38 1 1 1Awareness/associations 0.33 0.19 0.30 0.48 0.40 0.24 1 1 1
Table 3
Simultaneous tests of invariance
LO1, LO2, and LO3 = brand loyalty items; QL1, QL2, QL3, QL4, and QL5 = perceived quality items; AW1, AW2 and AW3 = brand awareness items; andAS1, AS2, and AS3 = brand associations items.
Competing models c
2d.f. Dc2Dd.f.
The unconstrained model: Model 3 with factor loadings variant 1124.49 222 – –
The constrained model: Model 3 with factor loadings invariant 1215.63 250 91.24** 28
Model 3 with LO1 on brand loyalty factor invariant 1127.98 224 3.49 2
Model 3 with LO2 on brand loyalty factor invariant 1129.17 224 4.68 2Model 3 with LO3 on brand loyalty factor invariant 1125.00 224 0.51 2Model 3 with QL1 on perceived quality factor invariant 1130.99 224 6.50* 2
Model 3 with QL2 on perceived quality factor invariant 1127.81 224 3.32 2
Model 3 with QL3 on perceived quality factor invariant 1124.68 224 0.19 2Model 3 with QL4 on perceived quality factor invariant 1144.63 224 20.14** 2
Model 3 with QL5 on perceived quality factor invariant 1142.24 224 17.75** 2
Model 3 with AW1 on awareness/associations factor invariant 1139.25 224 14.76** 2Model 3 with AW2 on awareness/associations factor invariant 1128.73 224 4.24 2Model 3 with AW3 on awareness/associations factor invariant 1127.32 224 2.83 2
Model 3 with AS1 on awareness/associations factor invariant 1125.12 224 0.63 2
Model 3 with AS2 on awareness/associations factor invariant 1127.26 224 2.77 2Model 3 with AS3 on awareness/associations factor invariant 1128.32 224 3.83 2
*p< 0.05.
**p< 0.001.B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 8

0.05). Therefore, the hypothesis of an invariant pattern of
factor loadings was tenable. Table 4 reports the maximumlikelihood estimates for the 10 items. All the factor loading,
error, and factor intercorrelation estimates were significant
at the 0.0001 level, and the fit indexes were SRMR = 0.066,GFI = 0.91, CFI = 0.96, and IFI = 0.96. The reliability wasalso satisfactory: 0.88, 0.86, and 0.87 for brand loyaltyamong Americans, Korean Americans, and Koreans, respec-tively; 0.81, 0.77, and 0.84 for perceived quality; and 90,0.86, and 0.86 for brand awareness/associations.
5.3. Pooled analysis
To check the cross-cultural validity of the 10 items further,
we conducted an individual-level multicultural factor analy-
sis, as suggested by Leung and Bond (1989). This technique
provides a meaningful way to pool cross-cultural data anddiscover etic dimensions at the individual level while elim-inating response sets often found in cross-cultural data. Leung
and Bond’s procedure begins with a double-standardization
procedure. First, a within-subject procedure standardizesscores across the variables for each subject. The mean of thevariables becomes zero, and the standard deviation becomesone for each individual in any culture. This removes thecultural positioning effect, ‘‘the relative location of theresponses made by the average individual from a particularculture’’ (Leung and Bond, 1989, p. 141). Under the cultural
positioning effect, even though certain variables show no
relationships in individual cultures, when a set of the vari-ables’ mean scores is used as a data point per culture, the
relationships between the variables may appear. Second, awithin-culture standardization, in which any variable has a
zero mean and unity standard deviation within each cultural
group, eliminates the patterning effect of culture, under whicha different culture shows a different relationship betweenvariables. After these standardizations, the data are pooledacross cultures and analyzed, and the dimensions or measuresobtained are etic measures.
After we pooled data from the three samples using Leung
and Bond’s procedure, we examined three kinds of mea-
surement models for the 10 items (i.e., four-, three-, and
one-dimensional), as we did in the individual analysis. First,the four-dimensional model ( c
2
d.f.=2 9= 219.82), in which
the correlation between brand awareness and associationswas unconstrained, was compared with the other four-dimensional model ( c
2
d.f. =3 0 = 317.46), in which the
correlation between the two factors was constrained to beunity. The c
2difference ( Dc2
d.f. =1= 97.64, p< 0.0001)
was significant. But high correlation (0.89) between brand
awareness and associations suggested the inseparability ofthose two constructs. The squared correlation (0.79) be-tween the two factors was larger than the average varianceextracted for either awareness (0.70) or associations (0.58).Therefore, the four-dimensional model was not supporteddue to lack of discriminant validity between brand aware-ness and associations. Second, the three-dimensional model
of brand loyalty, perceived quality, and brand awareness/
associations showed a better fit ( c
2
d.f. =3 2 = 326.19) thanTable 4
Maximum likelihood estimates and standard errors (SE)
LO = brand loyalty factor: QL = perceived quality factor; and AWAS = brand awareness/association factor. c2
d.f.=1 1 6 = 455.66, SRMR = 0.066, GFI = 0.91,
CFI = 0.96, and IFI = 0.96.
Across-culture equivalencies Americans Korean Americans Koreans
Parameter Estimate SE Estimate SE Estimate SE Estimate SE
lLO1, LO 0.86 0.02
lLO2, LO 0.88 0.02
lLO3, LO 0.77 0.02
lQL2, QL 0.85 0.03
lQL3, QL 0.80 0.03
lAW2, AWAS 0.82 0.02
lAW3, AWAS 0.79 0.02
lAS1, AWAS 0.79 0.02
lAS2, AWAS 0.81 0.02
lAS3, AWAS 0.61 0.02
qLO1, LO1 0.26 0.02 0.38 0.04 0.26 0.02
qLO2, LO2 0.18 0.02 0.19 0.03 0.30 0.03
qLO3, LO3 0.43 0.03 0.40 0.04 0.39 0.03
qQL2, QL2 0.26 0.04 0.34 0.05 0.24 0.04
qQL3, QL3 0.37 0.04 0.41 0.05 0.32 0.04
qAW2, AW2 0.26 0.02 0.39 0.04 0.37 0.03
qAW3, AW3 0.32 0.02 0.40 0.04 0.42 0.03
qAS1, AS1 0.38 0.03 0.39 0.04 0.37 0.03
qAS2, AS2 0.25 0.02 0.34 0.04 0.44 0.03
qAS3, AS3 0.57 0.04 0.69 0.06 0.65 0.04
fLO, QL 0.49 0.04 0.37 0.04 0.54 0.05
fLO, AWAS 0.34 0.04 0.32 0.04 0.25 0.06
fQL, AWAS 0.49 0.04 0.25 0.04 0.41 0.06B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 9

any other three-dimensional model. It was also better than
the one-dimensional model. Third, the model had excellentfit indexes in the pooled sample (SRMR = 0.042, GFI =
0.96, AGFI = 0.93, CFI = 0.96, and IFI = 0.96). The factor
correlations ranged from 0.35 to 0.48, and the compositereliability ranged from 0.82 to 0.88. In summary, the pooledsample confirmed the identical factor structure for the modelas was found in the individual and multigroup analysis,which is strong evidence of the universality of the items.
5.4. Measure of multidimensional brand equity (MBE) and
its index
We suggest this 10-item measure of MBE as a scale of
consumer-based brand equity. Because of the cross-culturalinvariance, the scores of the MBE and its dimensions canbe compared cross-culturally, and the different scores maybe considered indicative of true cross-cultural differencesin the constructs.
Adding up the raw scores of the 10 items of the MBE
may not be an appropriate way to develop a MBE index,because they are not evenly distributed among the threedimensions. More important, the three dimensions maycontribute differently to brand equity. To develop the for-mula for a single MBE index, the relationships between thedimensions and brand equity should be considered. There-fore, we generated the higher-order three-dimensional mod-
el that comprises the same dimensions and loading
specifications as the MBE three-dimensional measurementmodel. The higher-order model is equivalent to the MBEmodel because the intercorrelational paths of the MBEmodel can be converted into the causal paths of the high-er-order model without adding any new path or deleting anyexisting path (Stelzl, 1986). In the higher-order model, thethree dimensions are related to a higher-order factor, which
can be named ‘‘higher-order brand equity.’’ The fit indexes
remain the same between these two different models be-cause they are statistically equivalent.
All the causal paths of higher-order brand equity to the
dimensions were significant at the 0.0001 level. The smal-lestt-value of the paths was 10.88 for Americans, 5.44 for
Korean Americans, and 7.41 for Koreans. The path coeffi-cient of brand equity for brand loyalty was 0.60 for Amer-
icans, 0.56 for Korean Americans, and 0.69 for Koreans;
Table 5
The MBE index (standard deviation) by brand
nMBE
indexaBrandloyaltyPerceived
qualityBrand
awareness/
associations
Americans
Camera films 92.36 ****
,b28.64 **** 35.76 **** 95.01 ****
Agfa 41 2.37 (0.44) 1.96 (0.73) 3.09 (0.25) 1.79 (0.83)
Fuji 40 3.14 (0.37) 2.11 (0.77) 3.40 (0.57) 3.80 (0.67)Kodak 52 3.79 (0.53) 3.27 (0.97) 3.97 (0.56) 4.06 (0.57)Konica 44 2.64 (0.40) 2.01 (0.72) 3.17 (0.37) 2.51 (0.87)
Athletic shoes 27.42 **** 15.28 **** 16.63 **** 8.63****
Adidas 51 3.08 (0.49) 2.01 (0.79) 3.44 (0.53) 3.65 (0.83)Asics 46 2.97 (0.46) 2.01 (0.67) 3.34 (0.46) 3.43 (0.98)
LA Gear 44 2.87 (0.40) 1.84 (0.89) 3.18 (0.58) 3.49 (0.75)
Nike 49 3.82 (0.51) 3.01 (0.80) 4.09 (0.54) 4.26 (0.56)Puma 50 2.85 (0.55) 1.88 (0.79) 3.14 (0.71) 3.43 (0.88)
Reebok 52 3.34 (0.52) 2.42 (0.85) 3.58 (0.70) 3.95 (0.71)
Television sets 62.13 **** 24.19 **** 25.97 **** 37.67 ****
Samsung 48 2.67 (0.53) 1.94 (0.79) 3.21 (0.61) 2.65 (1.03)
Sony 52 3.46 (0.48) 2.76 (0.88) 3.82 (0.59) 3.67 (0.59)
Korean Americans
Camera films 23.26 **** 16.23 **** 7.61*** 19.82 ****
Agfa 22 2.59 (0.52) 1.98 (0.85) 3.05 (0.62) 2.35 (1.12)
Fuji 23 3.08 (0.33) 2.00 (0.75) 3.48 (0.57) 3.60 (0.62)Kodak 21 3.65 (0.59) 3.37 (0.71) 3.69 (0.70) 3.93 (0.60)Konica 18 2.68 (0.31) 2.30 (0.66) 2.92 (0.43) 2.68 (0.58)
Athletic shoes 19.03 **** 12.82 **** 11.78 **** 4.18**
Adidas 35 3.18 (0.57) 2.17 (0.86) 3.49 (0.65) 3.81 (0.82)
Asics 20 2.95 (0.41) 2.20 (0.71) 3.25 (0.38) 3.26 (1.19)
LA Gear 31 2.79 (0.38) 1.90 (0.73) 3.10 (0.51) 3.26 (0.71)Nike 24 3.66 (0.46) 3.08 (0.70) 3.77 (0.44) 4.18 (0.64)Puma 32 2.49 (0.52) 1.52 (0.60) 2.66 (0.80) 3.40 (1.15)
Reebok 25 3.07 (0.44) 2.09 (0.81) 3.46 (0.64) 3.50 (0.81)
Televisions sets 16.90 *** 9.80** 16.62 *** 3.56
Samsung 24 2.80 (0.34) 2.13 (0.67) 3.04 (0.41) 3.17 (0.56)
Sony 26 3.44 (0.68) 2.83 (0.90) 3.75 (0.75) 3.55 (0.83)
Koreans
Camera films 8.20**** 10.16 **** 1.79 2.71 *
Agfa 55 2.52 (0.44) 2.04 (0.63) 2.95 (0.47) 2.75 (0.72)Fuji 53 2.61 (0.48) 2.07 (0.70) 3.03 (0.56) 2.93 (0.75)Kodak 53 2.87 (0.52) 2.63 (0.77) 3.11 (0.47) 2.96 (0.64)
Konica 49 2.43 (0.47) 1.98 (0.64) 2.90 (0.52) 2.58 (0.97)
Athletic shoes 7.43**** 2.69* 3.39** 9.52****
Adidas 47 2.70 (0.48) 1.87 (0.64) 3.04 (0.56) 3.55 (0.92)
Asics 53 2.49 (0.52) 1.69 (0.65) 2.91 (0.71) 3.22 (0.73)
LA Gear 50 2.32 (0.44) 1.67 (0.61) 2.94 (0.42) 2.58 (0.93)Nike 54 2.78 (0.54) 2.05 (0.81) 3.20 (0.64) 3.38 (0.73)Puma 53 2.35 (0.47) 1.69 (0.60) 2.82 (0.52) 2.81 (0.92)
Reebok 57 2.64 (0.54) 1.86 (0.74) 3.15 (0.64) 3.22 (0.79)
Television sets 4.90* 7.50** 1.83 6.96 **
Samsung 55 2.78 (0.55) 2.41 (0.76) 3.21 (0.59) 2.83 (0.75)
Sony 52 2.54 (0.54) 2.00 (0.78) 3.37 (0.60) 2.41 (0.90)Notes to Table 5:
aThe American MBE index = 0.296 (the mean of brand loyalty) +
0.414 (the mean of perceived quality) + 0.290 (the mean of brand
awareness/associations); the Korean American MBE index = 0.289 (themean of brand loyalty) + 0.491 (the mean of perceived quality) + 0.220 (themean of brand awareness/associations); and the Korean MBE index = 0.411
(the mean of brand loyalty) + 0.315 (the mean of perceived quality) + 0.274
(the mean of brand awareness/associations).
bANOV A results for mean differences among brands by pro-
duct category.
*p< 0.05.
**p< 0.01.
*** p< 0.001.
**** p< 0.0001.B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 10

0.84, 0.95, and 0.53 for perceived quality; and 0.59, 0.43,
and 0.46 for brand awareness/associations. These coeffi-cients become the weights of the dimensions when comput-
ing the MBE index. Then, the weight of a dimension is the
portion of the path coefficient of that dimension in the sumof the three path coefficients. For example, among Amer-ican consumers, the weight of brand loyalty is 0.296, whichcomes from 0.60/(0.60 + 0.84 + 0.59). Table 5 illustrates theapplication of the MBE formula on the 12 brands. Acrosscultures, Nike and Kodak were the highest in the MBEindex and in every brand equity dimension. Sony received
higher evaluations than Samsung in the American market,
but Koreans preferred Samsung to Sony. As apparentcollectivists (Hofstede, 1991), they might have shown highconsumer ethnocentrism, favoring their domestic productover the import.
Further analysis found that the MBE index was highly
correlated to a composite score computed from a simplesum of the mean scores of the three dimensions. The
correlation between the MBE index and the mean score
composite was 1.00 for Americans, 0.98 for Korean Amer-icans, and 0.99 for Koreans. The other composite score,made from a sum of the raw scores of the 10 items, was alsohighly correlated to the MBE index. The correlation be-tween the MBE index and the raw score composite was 0.97for Americans, 0.91 for Korean Americans, and 0.94 forKoreans. These results suggest that the composite score
based on the mean or raw scores can be used as an excellent
proxy for the MBE index, in particular in non-Korean ornon-American countries.
5.5. Construct validity
We compared the MBE with purchase intention and
attitude toward brands for validity purposes. Practically,
purchase intention and brand attitude have been used as
surrogates for brand equity (e.g., Agarwal and Rao, 1996).Although brand equity may not be conceptually equivalentto intention and attitude (e.g., Rangaswamy et al., 1993;Swait et al., 1993; Park and Srinivasan, 1994), a strongrelationship between the constructs has been expected(Aaker, 1991; Keller, 1993). Thus, a high correlation revealsthe construct validity of the MBE. Both purchase intention
and brand attitude measures show high reliability. The
reliability of the two-item measure of purchase intentionwas 0.90 for Americans, 0.90 for Korean Americans, and0.80 for Koreans; and the reliability of the five-itemmeasure of attitude toward brand was 0.93, 0.94, and0.90. The prediction of a highly positive correlation betweenbrand equity and purchase intention was supported withcorrelations of 0.66, 0.70, and 0.55 ( p<0 . 0 0 0 1 )f o r
Americans, Korean Americans, and Koreans, respectively.
The correlation between brand equity and brand attitude wasalso high: 0.72, 0.71, and 0.50 ( p< 0.0001) for Americans,
Korean Americans, and Koreans, respectively. These find-ings demonstrate strong construct validity of the MBE.5.6. Convergent validity with a measure of overall brand
equity (OBE)
For a convergent validity check of the MBE, using the
same participants we developed a four-item unidimensionalmeasure of brand equity, which is a measure of OBE. Wegenerated 18 candidate items for OBE on the basis of ourdefinition of brand equity. We worded each item to comparea focal brand with its counterpart, which was a brandwithout a name but with identical product characteristicssuch as physical attributes, quality, and price. Other factors
being equal, the difference in consumer response between
the focal brand and the counterpart can be interpreted as thebrand equity of the focal brand.
After multistep examinations of the 18 candidate items, 4
items survived to form the OBE. Examples of the itemsinclude ‘‘It makes sense to buy Brand X instead of any otherbrand, even if they are the same’’ and ‘‘Even if anotherbrand has the same features as Brand X, I would prefer to
buy Brand X.’’ The items are reported in Appendix A. The
reliability of the OBE was 0.90, 0.89, and 0.90 for Amer-icans, Korean Americans, and Koreans, respectively. Thecorrelation between the OBE and the MBE index was 0.60,0.63, and 0.59 ( p< 0.0001) for Americans, Korean Amer-
icans, and Koreans, respectively. This high correlationsupports the convergent validity of the MBE.
6. Discussion
The purpose of this research is to develop a psychologi-
cally sound and cross-culturally generalizable measure ofbrand equity by testing Aaker’s (1991) and Keller’s (1993)conceptualizations. Our etic measure of MBE comprises 10items representing the three dimensions of brand loyalty,
perceived quality, and brand awareness/associations.
This study has important practical and theoretical im-
plications that benefit brand equity research in several ways.First, the measure can be used to examine how consumer-based brand equity results from its potential antecedents,such as brand knowledge, purchase and consumption ex-perience, marketing activity, corporate image, and environ-mental factors. Similarly, the consequences of brand equity
may be efficiently investigated using the measure. In parti-
cular, the impact of each dimension of the MBE onconsequent variables needs to be identified. As a result,nomological relationships between brand equity and itsrelevant variables may be studied more efficiently.
Second, the measure can expedite studies of brand
name values and extensions. For example, the measurewould be useful in examining the equity of cobrands. It
may be used to measure the equity of each brand sepa-
rately, before and after cobranding, and the brand equity ofthe cobrand. In addition, the measure may be used forstudying how brand equity affects brand extension and tohelp explain how customers use the brand equity of parentB. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 11

brands to develop preference among different extensions
(Sullivan, 1998).
Third, there may exist a potential causal order among the
dimensions of brand equity. For example, the hierarchy of
effects model suggests that brand awareness and associa-tions precede perceived quality and that perceived qualityprecedes brand loyalty (Levidge and Steiner, 1961). Percep-tion of high product quality leads to brand loyalty because itis the basis of consumer satisfaction (Oliver, 1997). If astrong ordering exists among the dimensions, to manageresources more efficiently, managers should consider stra-
tegies that focus on the timing of the dimensions.
Fourth, the measure is not only valid and reliable, but
also parsimonious, which helps practitioners track brandequity of individual brands on a regular basis. By assessingthe facets of brand equity, the measure appears to be a lessconfounded way of measuring brand equity. When practi-tioners use the measure as a tool for evaluating and trackingbrand performance over time, they may understand clearly
in which area the brand succeeds or fails. Thus, they can
make an efficient allocation of resources to maintain abalance among the brand equity dimensions. By trackingthe progress of target products’ brand equity and dimen-sions over time, managers will have a better understandingof the long-run effect of their efforts. As they understandthe dynamics between marketing efforts and brand equity,managers may set a reasonable goal for building brand
equity (Yoo, Donthu, and Lee, 2000). In addition, they may
track the relationships between brand equity and businessconsequences such as market share, sales, and profits.
Fifth, the extent to which the measure is cross-culturally
generalizable is somewhat limited. Although the factorloadings are invariant across samples, the factor intercorre-lations are not, which means that different cultures placedifferent levels of importance on the dimensions of brand
equity. The higher-order model analysis shows perceived
quality is the most important factor among Americans andKorean Americans, whereas brand loyalty is the mostimportant among Koreans. This may result from culturaldifferences, implying that brand equity-related marketingstrategies need to focus more on different dimensions ofbrand equity in different cultures.
6.1. Further research
Scales are established by ‘‘conventions or agreements
among scientists about a good scaling’’ (Nunnally andBernstein, 1994, p. 24). Our study should spur researchersto revise and revalidate our scale. We suggest two majordirections for further research. First, the three dimensions ofbrand equity may be extended into subdimensions. This will
help clarify the structure of brand equity in detail. For
example, the five dimensions of Aaker’s (1997) brandpersonality construct (sincerity, excitement, competence,sophistication, and ruggedness) may be related to ourawareness/associations dimension. Also, brand loyaltymay be further divided into cognitive, affective, conative,
and action loyalty (Oliver, 1997).
Second, the measure should have higher external gen-
eralizability. Additional research should validate the mea-
sure using different types of products, such as services andindustrial goods. The measure also should be extended totest how it can be applied to corporate or organizationalequity, retail equity, and chain equity. Also, the relationshipbetween dollar-metric brand equity and our consumer-basedbrand equity requires should be examined. In addition, themeasure must to be validated among non-student samples.
Researchers may worry about using students for brand
equity research because older consumers’ psychologicalattachments to brand names may vary markedly fromstudents, for several reasons (e.g., differences in length ofexposure to the brand, nostalgic associations with thebrand, information processing abilities). In a strict sense,the scale may be generalizable across only the threesamples that participated in the study. When more cultures
are involved in further research, a more etic scale of brand
equity can be developed.
Acknowledgments
The authors would like to thank two anonymous
reviewers and Charles M. Schaninger, the buyer behavior
associate editor of the JBR, for their detailed and insightful
comments. They would also like to thank Hoi-Joong Lee,Eue-Hun Lee, Young-Hyuck Joo, and Sungwoo Jung forhelping them with the translation of the questionnaire intoKorean language; Wookhee Lee, Jihee Lee, Betty Pak,Julie Sok, Jaebeom Suh, Angie Kim, Samuel Yu, ByungKi Ahn, John Koh, Suk Bum Choi, Brian S. Lee, SungKyu Park, Jinmyung Chang, Young T. You, Oliver Im,
Ronald Chu, Shin Jae Lee, Elizabeth Kang, Chul Woo
Kim, David Lee, Joe Kim, and Thomas Song for collectingKorean American data; and Sungnan Yoo, Minhee Yoo,Sang-Hyun Kim, Myung-Kwan Park, Yoonsook Cho,Sunlim Moon, Wonhee Kim, Geumhee Rho, and HyunjinLee for collecting Korean data.
References
Aaker DA. Managing Brand Equity. New York: The Free Press, 1991.
Aaker DA. Measuring brand equity across products and markets. Calif
Manage Rev 1996;38(3):102– 20.
Aaker DA, Keller KL. Consumer evaluations of brand extensions. J Mark
1990;54(January):27– 41.
Aaker JL. Dimensions of brand personality. J Mark Res 1997;34(Au-
gust):347– 56.
Agarwal MK, Rao VR. An empirical comparison of consumer-based mea-
sures of brand equity. Mark Lett 1996;7(3):237 –47.
Alba JW, Hutchinson J. Wesley: dimensions of consumer expertise. J Con-
sum Res 1987;13(March):411– 54.
Ambler T. How much of brand equity is explained by trust? Manage Decis
1997;35(March/April):283 – 92.
Anderson JC, Gerbing D. Structural modeling in practice: a reviewB. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 12

and recommended two-step approach. Psychol Bull 1988;103(3):
411– 23.
Bagozzi RP. Causal Models in Marketing. New York: Wiley, 1980.Barwise P. Brand equity: snark or boojum? Int J Res Mark 1993;10(March):
93–104.
Beatty SE, Kahle LR. Alternative hierarchies of the attitude –behavior re-
lationship: the impact of brand commitment and habit. J Acad Mark Sci1988;16(Summer):1–10.
Beckwith NE, Lehmann DR. The importance of halo effects in multi-attri-
bute attitude models. J Mark Res 1975;12(August):265– 75.
Berry JW. Introduction to methodology. Triandis HC, Berry JW, editors.
Handbook of Cross-Cultural Psychology: Methodology (vol. 2). Bos-
ton: Allyn and Bacon, 1980. pp. 1 – 28.
Bharadwaj SG, Varadarajan PR, Fahy J. Sustainable competitive advantage
in service industries: a conceptual model and research propositions. JMark 1993;57(October):83– 99.
Bollen KA. Structural Equations with Latent Variables. New York:
Wiley, 1989.
Boulding W, Kirmani A. A consumer-side experimental examination of
signaling theory: do consumers perceive warranties as signals of qual-
ity? J Consum Res 1993;20(June):111 – 23.
Burnett JJ, Dunne PM. An appraisal of the use of student subjects in
marketing research. J Bus Res 1986;14:329– 43.
Burnkrant RE, Page TJ Jr. An examination of the convergent, discriminant,
and predictive validity of fishbein’s behavioral intention model. J Mark
Res 1982;19(November):550– 61.
Byrne BM, Shavelson RJ, Muthe ´n B. Testing for the equivalence of factor
covariance and mean structures: the issue of partial measurement invar-
iance. Psychol Bull 1989;105(3):456– 66.
Calder BJ, Philips LW, Tybout AM. Designing research for application. J
Consum Res 1981;8(September):197–207.
Cobb-Walgren CJ, Beal C, Donthu N. Brand equity, brand preference, and
purchase intent. J Advertising 1995;24(3):25 –40.
Cooper WH. Ubiquitous halo. Psychol Bull 1981;90(2):218–44.Cox EP III, Wogalter MS, Stokes SL, Murff EJ. Tipton: do product warn-
ings increase safe behavior? A meta-analysis. J Public Policy Mark
1997;16(Fall):195– 204.
Deshpande R, Hoyer WD, Donthu N. The intensity of ethnic affiliation: a
study of the sociology of Hispanic consumption. J Consum Res 1986;
13(2):214– 20.
Dodds WB, Monroe KB, Grewal D. Effects of price, brand, and store
information on buyers’ product evaluation. J Mark Res 1991;28(Au-
gust):307– 19.
Douglas SP, Craig CS. International Marketing Research. Englewood
Cliffs, NJ: Prentice-Hall, 1983.
Durvasula S, Andrews JC, Lysonski S, Netemeyer RG. Assessing
the cross-cultural applicability of consumer behavior models: a
model of attitude toward advertising in general. J Consum Res1993;19(March):626– 36.
Farquhar PH, Han JY, Ijiri Y . Recognizing and measuring brand assets.
Working Paper Series, Report Number 91-119. Cambridge, MA: Mar-keting Science Institute, 1991.
Fazio RH, Zanna MP. Direct experience and attitude –behavior consistency.
Berkowitz L, editor. Advances in Experimental Social Psychology (vol.
14). New York: Academic Press, 1981. pp. 162– 202.
Fornell C, Larker DF. Evaluating structural equation models with unobser-
vable variables and measurement error. J Mark Res 1981;18(Febru-
ary):39–50.
Geertz C. The Interpretation of Cultures. New York: Basic Books, 1973.Guadagni PM, Little JDC. A logit model of brand choice calibrated on
scanner data. Mark Sci 1983;2(Summer):203– 38.
Gupta S. Impact of sales promotions on when, what, and how much to buy.
J Mark Res 1988;251(November):342 – 55.
Hofstede G. Cultures and Organizations: Software of the Mind. Berkshire,
England: McGraw-Hill, 1991.
Hu L-T, Bentler PM. Cutoff criteria for fit indexes in conventional criteria
versus new alternatives. Struct Equation Model 1999;6:1–55.Jo¨reskog KG, So ¨rbom D. LISREL 8.02. Chicago: Scientific Software Inter-
national, 1993.
Kamakura WA, Russell GJ. Measuring brand value with scanner data. Int J
Res Mark 1993;10(March):9– 21.
Keller KL. Conceptualizing, measuring, and managing customer-based
brand equity. J Mark 1993;57(January):1– 22.
Lane V , Jacobson R. Stock market reactions to brand extension announce-
ments: the effects of brand attitude and familiarity. J Mark 1995;59(January):63 –77.
Leung K, Bond MH. On the empirical identification of dimensions for
cross-cultural comparisons. J Cross-Cult Psychol 1989;20(June):133– 51.
Levidge RJ, Steiner GA. A model of predictive measurement of advertising
effectiveness. J Mark 1961;25(October):59– 62.
Lichtenstein DR, Burton S. The relationship between perceived and objec-
tive price – quality. J Mark Res 1989;26(November):429– 43.
Mahajan V , Rao VR, Srivastava RK. An approach to assess the importance
of brand equity in acquisition decisions. J Prod Innovation Manage1994;11:221– 35.
Meredith W. Measurement invariance, factor analysis, and factorial invar-
iance. Psychometrika 1993;58(December):525– 43.
Nelson PE. Advertising as information. J Polit Econ 1974;81:729–45.Netemeyer RG, Durvasula S, Lichtenstein DR. A cross-cultural assessment
of the reliability and validity of the CETSCALE. J Mark Res 1991;
28(August):320 –7.
Nunnally JC, Bernstein IH. Psychometric Theory (3rd edn). New York:
McGraw-Hill, 1994.
Olian JD, Schwab DP, Haberfeld Y . The impact of applicant gender com-
pared to qualifications on hiring recommendations. Organ Behav HumDecis Processes 1988;41(April):180– 95.
Oliver RL. Satisfaction: A Behavioral Perspective on the Consumer. New
York: McGraw-Hill, 1997.
Owenbey S, Horridge PE. The Suinn–Lew Asian self-identity acculturation
scale: test with a non-student, Asian-American sample. Soc Behav Pers.1998;26(1):57 – 68.
Park CS, Srinivasan V . A survey-based method for measuring and under-
standing brand equity and its extendibility. J Mark Res 1994;31(May):271– 88.
Petroshius SM, Monroe KB. Effect of product-line pricing characteristics
on product evaluations. J Mark Res 1987;13(March):511 – 9.
Rangaswamy A, Burke R, Oliva TA. Brand equity and the extendibility of
brand names. Int J Res Mark 1993;10(March):61– 75.
Rosenzweig PM. When can management science research be generalized
internationally? Manage Sci 1994;40(January):28– 39.
Rossiter JR, Percy L. Advertising and Promotion Management. New York:
McGraw-Hill, 1987.
Rummel RJ. Applied Factor Analysis. Evanston, IL: Northwestern Univer-
sity Press, 1970.
Sheth JN. Are there differences in dissonance reduction behavior between
students and housewives? J Mark Res 1970;7(May):243– 5.
Shimp TA, Bearden WO. Warranty and other extrinsic cue effects on con-
sumers’ risk perceptions. J Consum Res 1982;9(June):38– 46.
Shocker AD, Weitz B. A perspective on brand equity principles and issues.
In: Leuthesser L, editor. Report Number 88-104. Cambridge, MA: Mar-
keting Science Institute, 1988. pp. 2 –4.
Simon CJ, Sullivan MW. The measurement and determinants of brand
equity: a financial approach. Mark Sci 1993;12(Winter):28– 52.
Smith RE, Swinyard WR. Attitude– behavior consistency: the impact
of product trial versus advertising. J Mark Res 1983;20(August):257– 7.
Srivastava R, Shocker AD. Brand equity: a perspective on its meaning and
measurement. Working Paper Series, Report Number 91-124. Cam-
bridge, MA: Marketing Science Institute, 1991.
Srull TK. Methodological techniques for the study of person memory and
social cognition. Wyer RS, Srull TK, editors. Handbook of Social Cog-
nition (vol. 2). Hillsdale, NJ: Lawrence Erlbaum Associates, 1984. pp.1 –72.B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 13

Steenkamp J-BEM, Baumgartner H. Assessing measurement invariance
in cross-national consumer research. J Consum Res 1998;25(June):
78–90.
Stelzl I. Changing a causal hypothesis without changing the fit: some rules
for generating equivalent path models. Multivariate Behav Res 1986;21
(July):309– 31.
Sullivan MW. How brand names affect the demand for twin automobiles. J
Mark Res 1998;35(May):154 –65.
Swait J, Erdem T, Louviere J, Dubelaar C. The equalization price: a mea-
sure of consumer-perceived brand equity. Int J Res Mark 1993;10
(March):23– 45.Wallendorf M, Reilly MD. Ethnic migration, assimilation, and consump-
tion. J Consum Res 1983;10(December):292–302.
Wells WD. Discovery-oriented consumer research. J Consum Res 1993;19
(March):489–504.
Yavas U. Research note: students as subjects in advertising and marketing
research. Int Mark Rev 1994;11(4):35– 43.
Yoo B, Donthu N, Lee S. An examination of selected marketing mix ele-
ments and brand equity. J Acad Mark Sci 2000; 28(Spring):195– 211.
Zeithaml VA. Consumer perceptions of price, quality, and value: a
means–end model and synthesis of evidence. J Mark 1988;52(July):
2 –22.
Appendix A. Brand equity items and their means (standard deviations)
Item Americans Korean Americans KoreansTen-item MBEBrand loyalty
LO1. I consider myself to be loyal to X.
a2.26 (1.05) 2.27 (1.00) 2.00 (0.84)
LO2. X would be my first choice. 2.39 (1.08) 2.34 (1.09) 1.99 (0.84)
LO3. I will not buy other brands if X is available at the store. 2.22 (0.99) 2.17 (0.96) 2.00 (0.84)
Perceived quality
QL2. The likely quality of X is extremely high. 3.38 (0.72) 3.22 (0.77) 3.04 (0.63)
QL3. The likelihood that X would be functional is very high. 3.57 (0.69) 3.38 (0.76) 3.06 (0.63)
Brand awareness/associations
AW2. I can recognize X among other competing brands. 3.49 (1.21) 3.49 (1.19) 3.13 (1.14)
AW3. I am aware of X. 3.71 (1.20) 3.71 (1.13) 2.59 (1.00)AS1. Some characteristics of X come to my mind quickly. 3.14 (1.20) 3.15 (1.19) 2.64 (1.11)
AS2. I can quickly recall the symbol or logo of X. 3.37 (1.28) 3.31 (1.25) 3.19 (1.18)
AS3. I have difficulty in imagining X in my mind. (r)
b3.41 (1.20) 3.45 (1.15) 3.13 (1.04)
Four-item OBE
OBE1. It makes sense to buy X instead of any other brand,
even if they are the same.2.58 (0.90) 2.63 (0.91) 2.38 (0.86)
OBE2. Even if another brand has the same features as X,
I would prefer to buy X.2.63 (0.92) 2.78 (0.93) 2.35 (0.85)
OBE3. If there is another brand as good as X, I prefer to buy X. 2.58 (0.90) 2.72 (0.91) 2.15 (0.74)
OBE4. If another brand is not different from X in any way,
it seems smarter to purchase X.2.64 (0.86) 2.79 (0.94) 2.24 (0.77)
aX indicates a brand name.
b(r) indicates reversed scoring.B. Yoo, N. Donthu / Journal of Business Research 52 (2001) 1–14 14