 

5i“. .v..n. . .

2.. .3

(is.

.. 3. 51..

a...

1.7.1..
.n. H
a I...
.7

‘11... » V.
1..., Ja

E311“,

‘ . X41
.6...

.
.

”.7.” NFL;

w

«w

.. . .. .1
a «P». Jab?
Hahn‘s gmwﬁmﬁrw
s: .5” my ..

._

EM
.

an.
#3.. arm.
a Iraqi“?!
r .. . av...
ghﬂr.

(1.8., ‘ .

.f. 5.3.?1lb‘
.1. .. .v Ar a
was: K»

h.
. 5.41.;

.vad‘niwﬁi:

i. .3:
:4.

‘y :4! 3:.“ 1.

“.1. .
.542...

s.

.5

 

 

_ v

5h

nuh

:1 LIBRARY
.2 o o g Michigan State
University

This is to certify that the
dissertation entitled

EXAMINING THE CULTURAL ANTECEDENTS OF
PERCEIVED SOURCE CREDIBILITY AND THEIR
INFLUENCES ON ASIAN AMERICAN CONSUMERS'
RESPONSES TO ADVERTISING

presented by

MARIKO MORIMOTO

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Mass Media

 

 

@«Q %

Major ProfesMSﬁnature
July 8, 2005

 

Date

MSU is an Aﬁinnative ActiorVEqual Opportunity Institution

.u-o-l-b-0-0-I-I--I-O-

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ﬂ
2/05 c:/ClRC/DmDuo.Indd-p.18

EXAMINING THE CULTURAL ANTECEDENTS OF
PERCEIVED SOURCE CREDIBILITY AND THEIR INFLUENCES ON
ASIAN AMERICAN CONSUMERS’ RESPONSES TO ADVERTISING

By

Mariko Morimoto

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHIOLOSOPHY
Mass Media

2005

ABSTRACT

EXAMINING THE CULTURAL ANTECEDENTS OF
PERCEIVED SOURCE CREDIBILITY AND THEIR INFLUENCES ON
ASIAN AMERICAN CONSUMERS’ RESPONSES TO ADVERTISING

By

Mariko Morimoto

While the construct of source credibility has been a popular topic in advertising
research, very few studies have examined cultural factors that may inﬂuence this
construct. The purpose of the current study is to examine the cultural antecedents of
perceived model credibility and the potential effects on Asian American consumers’
responses to advertising. Speciﬁcally, Asian Americans’ levels of ethnic identiﬁcation
and cultural congruency associations are examined to understand their relationships with
the constructs of perceived model congruency and credibility. In addition, the study
examines how Asian Americans’ cognitive processing of perceived congruency and
credibility affect advertising responses.

The ﬁndings reveal that for Asian American consumers, model congruency
appears to be a signiﬁcant predictor of perceived model credibility and advertising
outcomes, while other cultural antecedents including the strength of ethnic identiﬁcation,
do not have a signiﬁcant impact on Asian Americans’ advertising evaluation process.
Also, among three factors of perceived model credibility, trustworthiness seems to play
an important role in determining the degree of perceived model credibility for Asian
American consumers. Suggestions for future research and contributions to the ﬁeld are

also discussed.

Copyrighted by
MARIKO MORIMOTO
2005

ACKNOWLEDGMENTS

Throughout this Ph.D. program, many people have supported me in various
ways. I thank them all. In particular, the following people have been extremely
supportive during my years at Michigan State University.

First, I would like to acknowledge the support and guidance that my dissertation
committee members have offered to me. Dr. Carrie La Ferle, my advisor and dissertation
committee chair, has been an outstanding and wonderful mentor since we ﬁrst met in
2001. She has spent numerous 'hours to work with me. Words cannot express how much I
truly appreciate her dedication to train me as a scholar and a teacher. Without her
guidance I would not have been able to complete my study. I truly hope that our
collaboration will continue in the future. Dr. Steve Edwards and Dr. Folu Ogundimu have
been constantly encouraging and providing me with thoughtful feedback and advice on
my research. From these discussions, many new ideas were generated. It was such a
pleasure to be able to work with them. Dr. Teresa Mastin has been extremely supportive
and helpful not only as a committee member, but also as a friend. She was always there
for me when I needed help or advice. I thank her for the wonderful friendship. Dr. Brenda
Wrigley has been my role model --- both as a scholar and as a person. There were days
when I lost conﬁdence in my research. But she has always encouraged me, convincing
me that I could do it. Thank you for your friendship and believing in me.

I would like to express my appreciation to the colleagues in Mass Media PhD.
Program and the staﬁ‘ members. Especially, Susan Chang has been a wonderful friend and

a colleague. We shared so many great times and survived the program together. I am

iv

lucky to have met her at Michigan State. Vicki Van Hurley and her family have been
always open and welcoming me with their warm and big heart. Marina Choi, Jay Newell,
James Ramos, Ki-Young Lee, and Carrie Trimble shared their wisdom with me to be
successful in the program. Thank you for making my time at MSU rewarding.

I would also like to express my appreciation to my family, Satoru and Sachiko
Morimoto, for their unconditional love and continuous support. Thank you for having the
faith in me. Eddie, my precious Japanese Chin, deserves to be credited here. He is the one
who has been seeing me struggling so many nights to complete this dissertation. You are
the best dog that I can ever ask for.

Lastly, I would like to conclude this section with a quote from my favorite artist.
This has been very encouraging especially when things were rough. I shall never forget

this, even aﬁer I leave from Michigan State and start a new academic career.

“ The desire for knowledge is the driving force for everything. It is never too late to want

to learn ” by Ace

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. viii
LIST OF FIGURES ................................................................................. x
Introduction ........................................................................................... 1
CHAPTER 1 — Literature Review ................................................................. 4
Perceived Model Credibility ........................................................................ 4
Trustworthiness ............................................................................ 5
Expertise .................................................................................... 7
Attractiveness ............................................................................... 8
Possible Cultural Antecedents of Perceived Model Credibility ............................... 10
Self, Culture, Ethnicity, and Distinctiveness ......................................... 10

Self and Culture ................................................................. 10

Race and Ethnicity .............................................................. 15

Distinctiveness ................................................................. l 8

Schema Incongruity ........................................................... 21

Situational Ethnicity, Ethnic Self-Awareness, and Ethnic Identiﬁcation ......... 25
Situational Ethnicity and Ethnic Self-Awareness ......................... 25

Ethnic Identiﬁcation ........................................................... 27

The Role of Cultural Congruency ..................................................... 35
Congruency Theory ........................................................... 35

Similarity vs. Familiarity ..................................................... 37

Congruency Research in Advertising ........................................ 39

CHAPTER 2 — Hypotheses ........................................................................ 46
Ethnic Identiﬁcation ...................................................................... 46
Perceived Similarity ..................................................................... 47

Ad Schema Incongruity ................................................................. 47

Model Congruency ....................................................................... 48

Model Credibility ........................................................................ 49
Attitudes toward Ad, Brand, and Purchase Intentions .............................. 51

Conceptual Model: Cultural Antecedents of Perceived Source Credibility.....52

CHAPTER 3 — The Study ......................................................................... 53
CHAPTER 4 - Method ........................................................................... 55
Research Design .......................................................................... 55
Stimuli ..................................................................................... 57
Measures .................................................................................. 58

CHAPTER 5 — Results ............................................................................ 61

Sample Characteristics .................................................................. 61
Manipulation Checks .................................................................... 63
Overview of the Main Findings ....................................................... 64
Hypothesis Testing ....................................................................... 68

Post Hoc Analyses of Cultural Antecedents of Perceived Source Credibility and

Aad ....................................................................................... 83
CHAPTER 6 — Discussion ........................................................................ 86
Model Congruency Effect on Advertising Evaluations of Asian Americans. . ....86
Absence of Model Congruency Effects on Caucasian Americans ................. 88

Effects of Other Cultural Antecedents of Perceived Model Credibility .......... 90
Perceived Model Credibility and Inﬂuence of Each Component .................. 95
CHAPTER 7 -- Suggestions for Future Research and Conclusion ......................... 97
Limitations of the Current Study ............................................... 97
Suggestions for Future Research .................................................... 99
Implications ............................................................................ 101
Practical Implications ........................................................ 101

Academic Implications ................................................. 102

Conclusion ............................................................................... 103
APPENDIX ........................................................................................ 105
Sample Advertisements ................................................................ 105
REFERENCES .................................................................................. 109

vii

LIST OF TABLES

Table 1

Experimental Design ............................................................................... 57
Table 2

Number of Cases per Manipulation ............................................................... 61
Table 3

Asian and Caucasian American Sample Characteristics ....................................... 62
Table 4a

ANCOVA Results on Perceived Model Credibility (All Sample) ............................ 64
Table 4b

ANCOVA Results on Perceived Model Credibility (Caucasian Sample) .................... 65
Table 4c

ANCOVA Results on Perceived Model Credibility (Asian Sample) .................... 65
Table 5a

ANCOVA Results on Attitudes toward Ad (All Sample) ...................................... 66
Table 5b

ANCOVA Results on Attitudes toward Ad (Caucasian Sample) .............................. 67
Table 5c

ANCOVA Results on Attitudes toward Ad (Asian Sample) ................................... 67
Table 6

Results of Descriptive Statistics and Correlation for H2 ....................................... 69
Table 7

Results of Descriptive Statistics and Correlation for H3 ..................................... 69
Table 8

Result of Regression for H3 ........................................................................ 70
Table 9

Correlations between Similarity and Credibility Components (H33) ........................ 70
Table 10

Results of Descriptive Statistics and Correlation for H4 ...................................... 71

viii

Table 11

Result of Regression for H4 ....................................................................... 72
Table 12
Results of Descriptive Statistics and Correlation for H4b ................................... 72
Table 13
Results of Descriptive Statistics and Correlation for H5 ...................................... 73
Table 14
Results of Descriptive Statistics and Correlation for H6 ....................................... 73
Table 15
Descriptive Statistics for Attitude toward Ad ................................................... 74
Table 16
Descriptive Statistics for Perceived Model Credibility for H8 ............................... 75
Table 17
AN COVA Results for H8a-c ...................................................................... 77
Table 18
Model’s Race as a Mediator (H8d) .............................................................. 78
Table 19
Similarity as a Mediator of Ethnic Identiﬁcation and Credibility ........................ 79
Table 20
Tests of Similarity as a Mediator (H10) .......................................................... 81
Table 21
Correlations between Perceived Model Credibility and Aad, Ab, and PI ................... 81
Table 22
Goodness-of-ﬁt Test Results for H12 ............................................................ 83
Table 23
Post Hoc Analysis of Cultural Antecedents of Perceived Credibility ........................ 84
Table 24
Post Hoc Analysis of Cultural Antecedents of Aad ............................................ 85

LIST OF FIGURES

Figure 1
Conceptual Model: Cultural Antecedents of Perceived Source Credibility .................. 52
Figure 2
Similarity as a Mediator between Ethnic Identiﬁcation and Credibility .................... 79
Figure 3
Similarity as a Mediator between Ethnic Identiﬁcation and Aad ............................ 80
Figure 4
Relationships between Aad, Ab, and PI ......................................................... 82

INTRODUCTION

The US. society has continued to become increasingly multicultural and diverse
over the past few decades. In particular, the growth of ethnic minority groups and their
buying power is notable. Of these minority groups that have been experiencing a
signiﬁcant population growth, Asian Americans have long been neglected in the ﬁeld of
advertising research. While the growth of the general population in the US. has
experienced an increase of 13 percent between 1990 and 2000 (US. Census Bureau
2001), the Asian American population has doubled from six million to 12 million in the
same time period (Raymond 2001). Currently this group consists of 12.5 million, which
is approximately 4.4 percent of the US. population (US. Census 2003), and the increase
appears to be continuing. The US. Census (2001) predicts that by 2015, the Asian
American population will reach 17.4 million and will consist of 5.5 percent of the overall
population.

The characteristics of this consumer group are quite attractive to marketers.
Asian Americans enjoy the highest level of average income. About 40 percent of Asian
American households have incomes higher than $75,000 while 35 percent of the
non-Hispanic Caucasian families have the same income level (U .8. Census 2003). Also,
according to estimates from the Selig Center at the University of Georgia, Asian
Americans achieved a buying power of $296.4 billion in 2002 (Gardyn and Fetto 2003).
As consumers, Asian Americans tend to be highly educated, brand-conscious, and prefer
premium goods (Kaufman-Scarborough 2000). Considering these facts, it is difﬁcult for

marketers to ignore this ethnic group.

In response, marketers try to accommodate their promotion campaigns to
different ethnic and cultural groups by using cultural cues with which the target group can
identify their value and beliefs (de Mooij 1998). Previous research on advertising
evaluations also indicates that consumer attitudes toward ads can inﬂuence their attitudes
toward the advertised brand and purchase intention (Batra and Ray 1986; MacKenzie,
Lutz and Belch 1986). Featuring endorsers from the same ethnic group as the target
consumer group in advertising is one method to incorporate cultural cues. However, very
few studies have been conducted examining the tactics used by marketers to attract Asian
American consumers or to assess their responses to advertising. The majority of studies
on Asian Americans and advertising use the method of content analysis to examine
representations of Asian Americans in print and TV advertisements (Taylor and Lee 1994;
Taylor, Lee and Stern 1995; Taylor and Stern 1997; Morimoto and La Ferle 2002).
Although we have recently begun to see some scholars paying attention to the reactions
of Asian Americans to advertisements (Lee, Choi and Lee 2003), more research in this
area is necessary to better understand and appeal to this affluent ethnic consumer group.

In addition to the practical contribution to the industry, the current study seeks to
explore how cultural factors and the race of models presented in advertising inﬂuence
Asian American consumer evaluations. Source credibility, the extent to which a source is
perceived believable, has been studied by a number of researchers to understand its effect
on persuasion and attitude change (Hovland and Weiss 1951; Bimbaum and Stegner
1979). In the context of advertising evaluations, source credibility can serve as a

predictor of consumer attitudes toward advertising (Lafferty and Goldsmith 1999;

Goldsmith, Lafferty and Newell 2000). Many studies on source credibility and its effect
have been conducted thus far, but the majority of these studies have treated source
credibility as an independent variable and focused mainly on the trustworthiness,
expertise, or attractiveness of the source (Pompitakpan 2004). Although it is crucial to
recognize the inﬂuence of perceived credibility on attitudes toward advertising, to fully
understand the inﬂuence of this construct on persuasion, more attention should be paid to
other factors inﬂuencing source credibility, such as model-product congruency or
model-consumer congruency.

Moreover, when studying ethnic consumers, one should recognize the
importance of cultural inﬂuences on their attitudes and behaviors such as racial
congruency with models or strength of ethnic identiﬁcation. Pompitakpan (2004) argues
that very little has been studied about the potential inﬂuence of culture on perceived
source credibility. Therefore, the current study seeks to contribute to the source
credibility and ethnic advertising literatures through an examination of the underlying
cognitive process of Asian Americans’ reactions to advertisement. Speciﬁcally, Asian
Americans’ levels of ethnic identiﬁcation, in relation to schema incongruity and
perceived similarity, will be examined along with cultural congruency factors to
understand their relationships with perceived source credibility. It is hoped that the
current study will contribute to adding knowledge to the existing source credibility and
ethnic advertising literatures, as well as provide guidance for advertisers to create more

effective advertising strategies targeted to the Asian American community.

CHAPTER 1
LITERATURE REVIEW
Perceived Model Credibility

Source credibility has been examined by many social scientists in the past
several decades in order to understand its effect on message persuasiveness. In a review
of the credibility literature over the past ﬁve decades, Pompitakan (2004) found that the
majority of studies undertaken indicate that perceived source credibility has some degree
of inﬂuence on communication effectiveness. In advertising, such credibility is often
associated with a model or an endorser (Friedman and Friedman 1979; Patzer 1983;
Freiden 1984; Kahle and Homer 1985; Kamins, Brand, Hoeke and Moe 1989; Kamins
1990; Ohanian 1990; Homer and Kahle 1990; Ohanian 1991; Kamins and Gupta 1994),
the advertiser (MacKenzie and Lutz 1989), the advertisement (MacKenzie and Lutz
1989), corporate credibility (Lafferty and Goldsmith 1999; Goldsmith, Lafferty and
Newell 2000; Newell and Goldsmith 2001), or web credibility (Choi and Rifon 2002). In
the current study, the literature on source credibility in relation to endorser/spokesperson
credibility is the most relevant and is reviewed in the following section.

Scholars have identiﬁed mainly three components of source credibility, and
proposed two models of source credibility. The source credibility model includes two
components, expertise, the source’s ability to make valid assertions about the object or
issue, and trustworthiness, a belief of receivers that the source makes honest assertions
(Hovland and Weiss 1951; Hovland, Janis and Kelley 1953). The source attractiveness

model has the third component, attractiveness which refers to the perceived attractiveness

of the source (Joseph 1982; Patzer 1983; Kahle and Homer 1985; Ohanian 1991; Kamins
1990). Source attractiveness in the context of message effectiveness (and communication)
is said to depend on the source’s familiarity, likability, similarity and overall
attractiveness to the receiver (McGuire 1985; Ohanian 1990). The following section will
review each of these credibility components in more detail.

Trustworthiness

One of the earliest studies concerning source credibility and its relationship with
communication effectiveness was conducted by Hovland and Weiss (1951). This study
identiﬁed trustworthiness as one factor inﬂuencing perceived source credibility. Ohanian
(1990) deﬁned trustworthiness as the degree of the receiver’s conﬁdence in and
acceptance of the communicator.

Trustworthiness is also considered to be related to the number of products a
celebrity endorses. Tripp and her colleagues ( 1994) examined how the number of
products celebrities endorse affects consumer evaluations of ads and purchase intentions.
They found that as the number of products a celebrity endorses increases, the level of
perceived trustworthiness decreases, and the number of products endorsed by the same
celebrity also tends to negatively affect attitudes toward advertisements. The authors
explained the phenomenon by suggesting that the greater number of endorsements made
by a celebrity results in a decline of their salience/distinctiveness, and as a result, the
endorsement effect including trustworthiness may be attenuated (Tripp, Jensen and
Carlson 1994).

Personal selling literature explains the relationship between perceived source

trustworthiness and credibility. Liu and Leach (2001) studied consumer satisfactions of
consultative salespeople, and customers’ level of trust with a sales-person was positively
related to the level of satisfaction with the salesperson. Jones, Moore, Stanaland, and
Wyatt (1990) focused on the effect of races and gender of salespeople on perceived
credibility by African American and Caucasian customers. They found that Aﬁican
American customers tended to perceive African American salespeople as more
trustworthy, and, thus more credible salespeople. The researchers also argued that
homophily, the degree to which individuals who interact are similar in terms of identity,
can enhance the perceived trustworthiness of the communicator. This argument can be
extended to the issue of congruency between a communicator and a receiver; if the
receiver perceives his/her ethnic identity congruent with that of the communicator, he/she
may consider the communicator more trustworthy. In the context of the current study, it
can be argued that if Asian Americans perceive homophily and/or congruency between
their race and the model’s race, they may rate Asian American models more trustworthy
and therefore more credible than Caucasian models.

While previous studies provide examples of positive relationships between
perceived source trustworthiness, credibility, and attitudes toward advertisements and/or
endorsers/salespeople, there have been very few studies that discovered a direct effect of
trustworthiness on purchase intention. For example, Ohanian (1991) discovered no main
effect of trustworthiness on the audience purchase intention of the advertised product
regardless of the product usage. It appears that perceived trustworthiness is interrelated

with the other components of source credibility, and leads to changes in attitudes and

behaviors (Patzer 1983). Liu and Leach’s (2001) study on business-to-business
salespeople’s credibility and its effect on customer satisfaction suggests that perceived
expertise of sales people affects the perceived trustworthiness. Hence, when measuring
perceived credibility, it is recommended to investigate all of the three components of the
construct rather than measuring the construct by simply asking whether or not the source
is trustworthy.
Expertise

Bimbaum and Stegner (1979) deﬁned source expertise as the perceived
correlation between the communicator’s reputation and the outcomes of veriﬁcations, and
such reputation is said to be inﬂuenced by the source’s training, experience, knowledge,
and ability. In the proposed study, perceived expertise of the model may be linked to the
model’s race depending on the product type the model advertises. Such perceived
expertise of the model may lead to a more favorable evaluation of the ad, brand, and
purchase intention.

Friedman and Friedman (1979) conducted a study on endorser effectiveness in
relation to product type. Using Kelman’s (1961) internalization theory which argues that
individuals are likely to lean toward the attitude or behavior advocated by a
communicator if they believe or value the attitude/behavior of the communicator, the
researchers hypothesized that when the endorser in an ad was perceived as an expert of
the advertised product, consumers would be likely to rate the ad more favorable. The
ﬁndings indicated that an ad of a vacuum cleaner was rated most highly by subjects when

the home appliance expert endorsed a vacuum cleaner, while a costume jewelry ad was

rated the highest when a celebrity endorsed the product (Friedman and Friedman 1979).
Another study conducted by O’Mahony and Meenaghan (1998) who also used
internalization theory (Kelman 1961) as the theoretical ground, revealed that consumers
expected congruency between the celebrity endorsers’ expertise and the type of products
they endorse. These examples indicate that when there is congruency between the
expertise of the endorser and the advertised product, the advertising rating is likely to be
more favorable. In terms of purchase intent, Ohanian’s (1991) celebrity endorsement
study suggests that consumer evaluations of perceived expertise of the endorser are
signiﬁcantly related to their intention to purchase the product. Also, Till and Busler
(1998) examined the match-up effect of endorsers and their expertise, and reported that
while both attractiveness and expertise positively affect attitude toward the advertised
brand, expertise appears to be a more important factor that drives the match-up effect.
Regarding the current study, it is expected that ethnic minority models will be regarded to
have more expertise in products related to their race and ethnicity, such as food, skin
foundation, or hair coloring and care products. Therefore, an interaction effect between
the product type and perceived credibility including expertise can be hypothesized.
However, the relationship should be inﬂuenced by the level of product involvement, and
this construct is discussed in a later section.
Attractiveness

The third component, attractiveness, has been a popular research topic in
advertising. Previous ﬁndings indicate that attractive endorsers tend to be liked more, and

are likely to have a positive inﬂuence on the product they advertise. The attractiveness of

the endorser becomes more important when the advertised brand does not have a strong
brand image (Joseph 1982). Hence, in the case of a new brand introduction, the use of
attractive models/endorsers may be an effective strategy to leave a positive impression in
consumers’ minds. Also, the level of attractiveness can inﬂuence not only advertising
evaluations and behavioral changes, but also the other components of source credibility
such as expertise and trustworthiness, and liking toward the endorser (Patzer 1983).

Kahle and Homer (1985) manipulated the level of attractiveness, along with the
level of subject involvement, and found a strong main effect of attractiveness on
evaluations for the advertised product and purchase intention. They also argue that when
attractive endorsers are featured in an ad, brand recall tends to be higher than the case of
unattractive endorsers (Kahle and Homer 1985). However, Ohanian (1991) has found that
perceived attractiveness was not related to purchase intention.

Some scholars paid attention to the match-up effect between celebrity endorsers
and product type and discovered that for an attractiveness-related product, the use of an
attractive celebrity endorser signiﬁcantly improved the perceived credibility and attitude
toward ads (Kamins 1990; Kamins and Gupta 1994; O’Mahony and Meenaghan 1998). A ‘
more recent study conducted by Till and Busler (2000) revealed that there was a
signiﬁcant main effect of endorser’s attractiveness on respondents’ attitudes toward
brands as well as their purchase intentions. In particular, for the product category that
helps to enhance personal attractiveness such as cologne, the effect of the endorser’s
physical attractiveness on brand attitude and purchase intent was signiﬁcantly greater

compared to the non-attractiveness related product like pens (Till and Busler, 2000). For

the most part, it appears that the effect of endorser attractiveness plays a positive and
important role in enhancing favorability of ads, brands, and purchase intention.

Based on the previous empirical ﬁndings, it is possible to argue that perceived
source credibility that consists of a model’s expertise, trustworthiness, and attractiveness
plays an important role in consumers’ cognitive processing of advertisements, including
Asian American consumers. While these three traditional factors certainly can inﬂuence
the degree of perceived source credibility, it is also important to take other cultural factors
into consideration when studying ethnic minority individuals and how they process
information. The following sections will discuss such potential cultural antecedents of
perceived model credibility and explicate the key concepts.

Possible Cultural Antecedents of Perceived Model Credibility
Self, Culture, Ethnicity, and Distinctiveness
Self and Culture

The concept of self has been acknowledged as an important and relevant factor
in the study of consumer behavior (Sirgy 1982; Hogg, Cox and Keeling 2000; Reed
2002). As one’s self is central to his/her thinking (Triandis 1989), possibly affecting
attitudes and behaviors, it is essential to examine the self-concept and self-identity in
considering the relationship with the cultural antecedents of perceived model credibility.

Rosenberg (1979) deﬁned the self-concept as “the totality of an individual’s
thoughts and feelings having reference to himself as an object “(p. 7). Based on this
deﬁnition, consumer behavior scholars have attempted to develop several perspectives of

the self concept in order to examine its relationship with consumption behaviors. One

10

such perspective applicable to the current study is the social-cognition approach which
regards the self-concept as the overall self-referent knowledge in memory (Reed 2002).
In particular, Sirgy’s (1982) self-image/product-image congruity theory clearly represents
this approach. The construct of self-image, the degree of value associated with a speciﬁc
self-concept (Sirgy 1982), plays an important role in the theory. According to the theory,
product attributes or product images tend to activate a self—schema involving the same
images, and these image attributes, along with the value placed on the product, are likely
to be inﬂuenced by the evoked self-schema (Sirgy 1982). In other words, the link
between self-image and product attributes would invoke the strength of association in
memory between a particular image and a product, and the consumer’s value intensity
with the self-image level (Reed 2002). Speciﬁcally, this self-image and product image
relationship results in one of the following situations: 1) positive self-congruity, a
comparison between a positive product image perception and a positive self-image belief;
2) positive self-incongruity, a comparison between a positive product image perception
and a negative self-image belief; 3) negative self-congruity, a comparison between a
negative product-image perception and a negative self-image belief; and 4) negative
self-incongruity, a comparison between a negative product-image perception and a
positive self-image belief (Sirgy 1982). Among these situations, it is argued that positive
self-congruity is likely to positively correlate with the strength of purchase intention the
highest, followed by positive self-incongruity, negative self-congruity, and negative
self-incongruity (Sirgy 1982). In other words, when a consumer has a positive image

about the product and such a product image is positively reﬂected on his/her own image

11

through the use or ownership of the product, his/her purchase intention is likely to be
high.

In the process of the self-image/product-image congruity theory, it is indicated
that one’s self-esteem and self-consistency (self maintenance) needs tend to serve as
mediators. Using this theory, scholars have argued that consumers would lean toward
positively valued products in order to maintain their positive self-image, and would be
more motivated to purchase/consume image-congruent products to enhance self-esteem
and create identity (Sirgy 1982; Shavitt 1990; Hogg et al. 2000). In the ﬁeld of
advertising research, Wong and Zinkhan (1995) found that when there was congruency
between consumers’ self-concepts and advertising attributes, their brand evaluations and
purchase intention tended to be more favorable than the incongruent situation. Applying
the ﬁndings of Wong and Zinkhan (1995) to the current study, it is possible to argue that
Asian American consumers may lean toward racially-congruent models and ﬁnd them
more credible than racially-incongruent models because the racially-congruent models
could trigger their self-schema and enhance their self-identity. In addition, Asian
American consumers may ﬁnd racially-congruent models more credible if they advertise
racially-charged products because the value and image associated with the product match
with the image of the models, which is also similar to their own image. In a sense, we can
argue that one’s self-concept and self-image play a central and signiﬁcant role in the
information processing and the evaluation of advertisements because one tends to refer
back to his/her own self in order to process information and make decisions.

As individuals are believed to have multiple characteristics, the self-concept is

12

considered multidimensional (Gergen 1971). In the case of ethnic consumers, one agent
that may have a strong potential inﬂuence on the formation of the self-image,
self-concept, and/or self-identity is culture.

Historically, scholars have attempted to deﬁne culture from various perspectives.
In this study, culture is explicated based on the deﬁnition provided by Lehman, Chin, and
Schaller (2004) which says that culture is a collection of discrete behavioral norms and
cognitions commonly held by individuals within deﬁnable population (p. 690). Carpenter
(2000) argues that because culture consists of shared elements by the people within, it can
inﬂuence individuals’ perceptions of self, thoughts, feelings, and behaviors. Also, such
shared elements of culture help distinguish the people within from others outside the
cultural boundary (Lehman et a1. 2004). According to these examples, culture deﬁnes
unique characteristics of people within. Therefore, it is important to closely examine the
potential cultural factors that may affect ethnic consumer behavior and advertising
evaluations since various ethnic groups have different and unique cultures that may
trigger exclusive consumer behavior and advertising information processing system. In
particular, factors related to share beliefs and values that are likely to inﬂuence
advertising evaluation systems and consumer behavior, including perceived model
credibility, will be referred as cultural antecedents in the current study.

Triandis (1989) argued that culture tends to inﬂuence every level of an
individual’s attitude and behavior through one’s self perception. According to the
researcher, an individual is assumed to have three main aspects of self: 1) the private self,

which involves the individual’s traits, states, or behaviors; 2) the public selﬁ which

13

corresponds with the generalized other’s view of the self; and 3) the collective self,
cognitions related to a view of the self found in collective situations such as family,
coworkers, and society; an individual is likely to differently sample these aspects of self
in different cultures and contexts (Triandis 1989). Combining Tajfel’s (1979) social
identity theory which states that a part of the individual’s self-concept is derived from
his/her knowledge of the membership in a social group along with the values and
emotional signiﬁcance associated with the membership, Triandis (1989) argued that in a
collectivistic culture such as Asian culture, one is likely to sample the collective self
which corresponds to assess the self by the generalized other (Triandis 1989). With
regards to Asian American consumers, although they live in America which is often
classiﬁed as an individualistic culture according to Hofstede’s (1997) typology, they may
be more inclined to sample the collective self due to their cultural heritage compared with
the general American population, and therefore, recognize themselves as a member of the
Asian American consumer group. Such self-recognition as Asian American may enhance
their identity as Asian Americans.

Also, the self-concept in the cultural context is important for understanding the
inﬂuence of an endorser on Asian American consumers because of the characteristics of
uncertainty avoidance. According to Hofstede (1997), uncertainty avoidance is deﬁned as
the degree to which an individual tolerates uncertainty and ambiguity. Asians are
generally considered to be high in uncertainty avoidance (Hofstede 1997), and Asian
Americans may have a higher degree of uncertainty avoidance compared with general

Caucasian Americans if they strongly perceive themselves as Asians. Because they may

14

be more prone to be high in uncertainty avoidance, the role of source credibility in the
information processing may play an important and critical role in their advertising and
brand evaluations, as well as purchase intentions. For example, Asian American
consumers may ﬁnd a model from the same ethnic background more credible because
they regard the model as more similar to themselves and familiar, which reduces
uncertainty. Hence, when studying ethnic consumers and source credibility issues, it is
advisable to pay attention to the inﬂuence of the self-concept and identity in relation to
culture.
Race and Ethnicity

Several cultural factors can be considered in relation to Asian Americans’
self-identity. In particular, ethnicity and race may be important elements that inﬂuence
their self-concept. Many scholars have attempted to explicate the concepts of race and
ethnicity in social science. Traditionally, race and ethnicity have been considered separate
constructs. For example, Marger (1999) states that ethnicity is characterized by unique
cultural traits, sense of community, and ascribed membership with a distinct territory.
Other scholars argue that ethnicity refers to a group with common national or religious
backgrounds (Ogden, Ogden and Schau 2004). On the other hand, it is often argued that
race refers to biological and physical characteristics of a person (Kibria 1998). While
ethnicity appears to suggest a sense of belonging to cultural, religious, and geographical
groups, race has been mainly deﬁned by an individual’s physical characteristics and
appearances.

However, some scholars point out that the distinction between these two

15

constructs is not clear. Khanna (2004) argues that one’s racial identity could be
inﬂuenced by cultural exposures that are often associated with ethnicity. Stephan and
Stephan (1989) suggest that physical appearance may inﬂuence one’s perception on
his/her ethnicity. For example, the study of Khanna (2004) that examined ethnic identity
of multiracial Asians indicates that ethnic identity of Asian-white subjects is likely to be
inﬂuenced by their Asian physical characteristics. Hence, for the current study, the
combined approach to race and ethnicity will be applied. Speciﬁcally, ethnicity is deﬁned
as a group with cormnon national/geographic origin, culture, religion, and/or physical
characteristics such as skin color in the current study. Inclusion of physical characteristics
in the deﬁnition is important because members of a minority group are more likely to feel
familiar to a person from another minority group than a person who is Caucasian, even if
they are not from the same ethnic group (Appaih 2001). Agbayani-Siewert (2004) also
argues that membership in a racial group is similar with shared cultural values and
attitudes. Because one’s physical appearance appears to inﬂuence his/her cognitive
processing of cultural cues, it is crucial that the operational deﬁnition of ethnicity reﬂects
both cultural and physical characteristics.

Based on this deﬁnition, Asian Americans in this study are deﬁned as people
with the ancestry rooted in Asian countries other than India, the Middle East, and the
former members of Soviet Union (Taylor and Lee 1991). For example, American citizens
or permanent residents with the ancestry rooted in China, Korea, Japan, Taiwan, Hong
Kong, Vietnam, Laos, Cambodia, the Philippines, Singapore, and Thailand may be

considered Asian Americans. Because the proposed deﬁnition of ethnicity indicates the

16

shared physical characteristics, individuals with East and South East Asian backgrounds
will be targeted for the study.

In the current study, the deﬁnition of Asian culture is based on the region where
similar values and/or religion are held by the people. In this region (East Asia),
Confucianism is the basis for the beliefs and behaviors of the people. According to the
doctrine, every member in the society has a role to play, and the key relationships are
established around the family; obedience to parents, elder people, and sovereign are
expected (Keller and Kronstedt 2005). Relationships and connections are highly regarded
in Confucian society, which can be also found as a characteristic of collectivistic culture.
People in Confucius culture value harmony with others, and such a tendency is reﬂected
in a group-oriented nature of people in this region. Because these behavioral and
attitudinal tendencies are shared by people in this region due to the commonly held value
of Confucianism, the current study deﬁnes Asian culture by the geographical region.

Ethnicity can be operationalized in various ways based on the proposed
deﬁnition. Stephan and Stephan (1989) introduce several variables reﬂecting one’s ethnic
identity, such as religion, and cultural and religious ceremonies as well as food
consumption, and Xu, Lotz, and Almeida (2004) have developed a scale based on these
variables. Webster (1994) also identiﬁed language use and reference group inﬂuence as
indicators of ethnic identity. Ogden et al (2004) state that the national/regional origin
should be taken into consideration when studying ethnicity and ethnic consumers. In
addition, self-identiﬁcation measures, such as asking subjects to describe who they are

(McGuire et a1. 1987), may be helpful in assessing ethnic identity because they are

17

believed to better reﬂect the inner beliefs of individuals about their vieWpoint on cultural
and ethnic realities (Desphande et al. 1986; Ogden et al. 2004). These variables are
considered to be representative of one’s ethnic identity and the strength of ethnic
identiﬁcation.

Ethnic identity and ethnic identiﬁcation should be treated separately, according
to several scholars such as Ogden et al. (2004) and Stayman and Deshpande (1989).
Stephan and Stephan (1989) deﬁne ethnic identity as “the identiﬁcation of an individual
or group of individuals with a particular ethnic group (p. 507).” Hence, ethnic identity
represents one’s association to an ethnic group (Ogden et al. 2004). Meanwhile, ethnic
identiﬁcation is often associated with the strength of one’s ethnic affiliation (Green 1997:
Appaih 2000; Ogden et al. 2004). For the current study, both aspects of ethnicity will be
incorporated into the concepturalization of ethnicity.
Distinctiveness

In terms of one’s ethnic identity and its relation to Asian American consumers’
cognitive processing, McGuire’s (1985) distinctive theory can be used as a guide for
understanding the mechanism of ethnic identity and its inﬂuence on perceived model
credibility. The researcher argued that for individuals who are members of ethnic
minority groups, such as Black, Hispanic, Asian, and Native Americans, they are likely to
be more aware of their ethnicity compared with members of the majority group because
of the distinctive traits which tend to be more central to their self-concept (McGuire
1984; McGuire et al. 1978; Aaker, Brumbaugh and Grier 2000). Forehand, Despande, and

Reed (2002) reported that individuals are more likely to self-report their social identity,

18

such as ethnicity, when they are distinctive in social environments. Applying these
arguments, it is possible to argue that since Asian Americans only consist approximately
4 percent of the overall US. population (US. Census Bureau 2001), it is likely that they
perceive their distinctive ethnic traits more than those in the majority group. As a result,
their ethnic-identity may be strongly inﬂuenced by their ethnicity. Then, the self-identity
could inﬂuence potential cultural antecedents of perceived model credibility such as
strength of ethnic identiﬁcation.

In the context of marketing, distinctiveness implies that when the proportion of
minority group members is small in the general population, ethnically-targeted cues
and/stimuli are more likely to work effectively to appeal to the minority group members
(Deshpandé and Stayman 1994; Aaker et al. 2000). For example, Deshpande and
Stayman (1994) compared Hispanics and Caucasian American consumers in two cities,
San Antonio (Hispanics as a dominant group) and Austin (Caucasians as the majority) to
investigate how the perceived ethnic distinctiveness inﬂuences the evaluations of
spokespersons. The results suggest that in both cities, ethnic minority participants evoked
their ethnic identity more spontaneously than members of the majority group. In addition,
when consumers’ ethnic salience became greater due to the numerical minority status,
their evaluations of racially congruent models in ads and the advertised brands were more
positive. Speciﬁcally, Hispanic consumers in Austin where they were a minority group
tended to mention their ethnicity more spontaneously than in those in San Antonio where
the Hispanic population is considered a majority; they were more likely to rate the

Hispanic spokesperson more favorably than Hispanics in San Antonio (Deshpande and

19

Stayman 1994).

In addition, Aaker et a1. (2000) argued that consumer distinctiveness can serve as
a moderator in targeted versus non-targeted marketing efforts. These researchers found
that when minority consumers, such as African Americans and homosexuals, were
exposed to advertising stimuli that were tailored for their unique characteristics, they
rated the ads more favorably than non-targeted ads (Aaker et al. 2000). It is noteworthy to
point out that in these studies, distinctiveness has been considered an enduring state
where individuals who belong to minority groups were considered to consistently
perceive distinctiveness due to their relatively small size in the greater population.

Other scholars have proposed that it is ethnic self-awareness that moderates the
strength of ethnic identiﬁcation and the target market effects, rather than ethnic
distinctiveness (Stayman and Deshpandé 1989; Forehand and Deshpande 2001). This line
of research states that while ethnic self-awareness is a temporary salience that can be
cued by ethnic distinctiveness, the strength of ethnic identiﬁcation is an ongoing
association between one’s sense of self and their ethnicity; therefore, ethnic
self-awareness is activated only when people are ethnically cued (Forehand and
Deshpande 2001). As a result, ethnic self-awareness enhances the target market effects in
the case of high ethnic identiﬁers; however, it does not necessarily affect one’s strength
of ethnic identiﬁcation (Forehand and Deshpande' 2001). Although this argument can be
considered an alternative explanation to the role of self-awareness, empirical studies
suggest that ethnic and social distinctiveness tend to inﬂuence the levels of ethnic

identiﬁcation for Hispanics and Caucasians (Deshpande and Stayman 1994) and Asian

20

Americans (Forehand et al. 2002). Therefore, it can be hypothesized that the intensity of
Asian American ethnic identiﬁcation is likely to be inﬂuenced by their ethnic
distinctiveness.
Schema Incongruity

Another factor that may inﬂuence the effect of Asian American ethnic
identiﬁcation on perceived model credibility and advertising evaluations is schema
congruity. 'Shank and Abelson (1979) deﬁne a schema as a cognitive structure with
speciﬁc knowledge about stimuli, and plays an important role in encoding and retrieving
memory information (Dimofte, Forehand and Deshpandé 2004). Originally proposed by
Mandler (1982), schema congruity theory argues that when an individual processes
information to evaluate an object/issue, one of the three situations can take place. First,
when the individual’s expected or prior values are congruent with the perceived value of
the object of the evaluation, he/she is likely to accept the value easily because it is
perceived as a match, therefore, familiar (Mandler 1982; Stayman, Alden and Smith
1992). In this case, the knowledge formation leading to an evaluation of the object is
relatively less elaborated because such a match between the individual’s prior expectation
and the perceived value does not require a strong effort to process the information
(Peracchio and Tybout 1996). Combined with congruency theory (Osgood and
Tannenbaum 1955), such congruity is likely to result in a favorable evaluation of the
object because people prefer objects that match with their expectations (Mandler 1982;
Meyer-Levy and Tybout 1989).

Another possibility is schema incongruity. Mandler (1982) broke down the

21

concept of schema incongruity into two levels: moderate and extreme schema incongruity.
Schema incongruity is a state when there is a partial mismatch between an individual’s
prior expectation about the object and the actually perceived value of the object (Mandler
1982; Meyers-Levy and Tybout 1989). Mandler (1982) argues that the degree to which an
individual can resolve anomalities or a mismatch between their expectation and the actual
value of the object with ease determines the level of incongruity.

Moderate schema incongruity is believed to result in a more favorable evaluation
of the object compared with schema congruity or extreme incongruity because in this
situation, an individual is likely to be able to resolve incongruity by attempting to process
information intensively (Mandler 1982). Such elaboration of knowledge can be perceived
as interesting and rewarding, and therefore, is likely to result in a favorable evaluation of
the object (Peracchio and Tybout 1996). On the other hand, in the case of extreme schema
incongruity, Mandler (1982) argues that it is likely to generate a negative attitude or
evaluation because the accommodation to anomalities is likely to produce anxiety or
frustration. Therefore, to maximize a positive evaluation of an object, it is preferable to
achieve the moderate level of schema incongruity when an individual processes
information about an object.

Previous studies in the ﬁeld of marketing examined the effect of moderate
schema incongruity, and discovered that this state maximized consumers’ positive
attitudes toward products and brands (Meyers-Levy and Tybout 1989; Stayman et al.
1992; Meyers-Levy, Louie and Curren 1994). For example, Meyers-Levy and Tybout

(1989) discovered that moderate incongruity between consumers’ perception of soft drink

22

attributes and the actual product values led to a more favorable product evaluation
compared with the cases of congruity and extreme incongruity. In terms of brand
evaluations, Meyers-Levy et al. (1994) manipulated company brand names for a nutrition
book, and found that when there was a moderate mismatch between the product and the
brand name, consumers were likely to evaluate the brand more favorably compared with
congruity and extreme incongruity. Stayman et al. (1992) examined the level of congruity
between consumers’ pre- and post-product trial perceptions, and found support for
Mandler’s (1982) argument on moderate schema incongruity effect on positive
evaluations. Based on these examples, we can argue that if Asian Americans perceive
moderate incongruity between their prior knowledge about an object, such as
advertisements, and the actual attributes of the object, such as featured models, they may
evaluate the object more favorably. Although previous content analyses of Asian models
in American advertising indicated that Asian Americans were slightly overrepresented as
a proportion of their size in the overall population (Taylor and Lee 1994; Taylor and Stern
1997; Morimoto and La Ferle 2002), their appearance in mainstream advertising is still
not prominent compared with other ethnic groups. Therefore, Asian Americans may
generally expect to see Caucasian models in advertising. When their expectation is
incongruent with what they actually see in ads, such as seeing an Asian American model,
their evaluation of the model may be regarded as interesting and a pleasant surprise, and
therefore, leading to a more favorable attitude toward the model in the ad.

There is also another possibility of the positive effect of moderate schema

incongruity on one’s ethnic identity. In the advertising literature, Dimofte, Forehand and

23

Deshpande (2004) examined how ad schema incongruity inﬂuences the ethnic awareness
of Hispanics and Caucasian Americans and their advertising evaluations. The researchers
manipulated a voice-over of a TV commercial, and created a version with the English
voice-over and another version with the Spanish voice-over, and examined how the
incongruity between consumers’ prior expectation about typical advertising (feattuing
mainly English voice-overs) and the actual voice-over in the manipulations inﬂuenced
participants’ ethnic awareness and their attitude toward the spokesperson in the TV
commercials (Dimofte et al. 2004). The results indicate that the exposure to incongruent
ad schema (TV commercial with Spanish voice-over) increased the self-ethnic report rate
for both Hispanics and Caucasians, but the increase rate was greater for the Hispanic
participants, and the brand recall rate for the ad schema incongruent manipulation was
greater than that of the congruent manipulation (Dimofte et al. 2004). The researchers
also argue that the incongruity between the existing ad expectation and the actual ad
attributes can draw more attention to speciﬁc ad characteristics such as ethnic cues used
in the ad; because processing the information on such ad-schema incongruity requires
more involvement, the elaboration in the information processing is likely to increase
ethnic awareness (Dimofte et al. 2004).

A similar phenomenon may take place for Asian American consumers and their
ethnic identiﬁcation. While the strength of ethnic identiﬁcation is generally considered an
ongoing state that may not be necessarily inﬂuenced by advertising cues (Forehand and
Deshpandé 2001), in other literatures, situational ethnicity (Stayde and Deshpande

1989), which is generally regarded as a temporary state, is measured by the same scale

24

used for ethnic identiﬁcation. Therefore, it can be also argued that ethnic identiﬁcation
has a base level that can be considered ongoing or/and enduring, but the base may be
enhanced by situations and cues. In other words, the intensive ethnic cue processing (in
this case, model’s race in an ad) resulting from the incongruity between Asian American
consumers’ expectation about models’ race in advertising in general and the actual
portrayal of an Asian American model in the ad may enhance their strength of ethnic
identiﬁcation particularly when they are exposed to ads featuring Asian American models.
As these examples indicate, the level of schema congruity may serve as a moderator for
the strength of ethnic identiﬁcation, as well as the evaluation of the model and the overall
impact of credibility on advertising outcome measures.

Situational Ethnicity, Ethnic Self-Awareness, and Ethnic Identiﬁcation

Situational Ethnicity and Ethnic Self-Awareness

In ethnic advertising research, three ethnicity-related constructs have been
identiﬁed: situational ethnicity or felt ethnicity (Stayman and Deshpandé 1989; Forehand
and Deshpandé 2001), ethnic self-awareness (Forehand and Deshpande' 2001), and ethnic
identiﬁcation (Whittler and DiMeo 1991; Appaih 2000; Appaih 2001; Green 1999;
Whittler and Spira 2002).

Situational ethnicity is based on an assumption that variability of ethnic identity
may depend on the immediate social situation and is related to the individual’s perception
of that situation (Okamura 1981). Stayman and Deshpandé (1989) distinguished the
difference between situational (or felt) ethnicity and ethnic identiﬁcation, stating that

while ethnic identiﬁcation refers to who one is, situational ethnicity is about how one

25

feels in and about a situation. Combined with distinctiveness theory, it can be argued that
if an individual perceives him/herself as a ethnic minority member, the degree of
situational (or felt) ethnicity is going to increase. Stayman and Deshpandé (1989) suggest
that the interaction effect between consumers’ self-designated ethnicity choice and their
social situation appeared to inﬂuence situational ethnicity. While situational ethnicity is
more concerned about an individual’s self-perception depending on the context he/she is
in, ethnic self-awareness (discussed in the following paragraph) tends to be more focused
on the effect of situational ethnicity on cognitive information processing and behaviors.
Forehand and Deshpande (2001) introduced the concept of ethnic self-awareness,
a temporary state in which an individual is more sensitive to information related to
his/her ethnicity, to better understand the inﬂuence of temporary ethnicity on attitude and
behavioral changes. Unlike situational ethnicity which assumes that the main inﬂuential
factor is the environment in which the individual is located, ethnic self-awareness
assrunes that this state depends on many factors, including the social situation. For
example, the strength of ethnic identiﬁcation, which is considered an enduring state
(discussed later in the section) and ethnic cues in advertising, have been considered the
potential determinants of ethnic self-awareness in advertising research (Forehand and
Deshpande 2001; Forehand, Deshpande and Reed 2002). In addition, related to the idea
of advertising ones as a potential determinant of ethnic self-awareness, Dimofte et al.
(2004) argue that moderate schema incongruity between the prior expectation about
seeing Caucasian models in general advertising and the actual exposure to ethnic

minority models in an ad may also inﬂuence ethnic self-awareness. As a result, the

26

elaboration of ethnic self-awareness may lead to more favorable advertising evaluations.
Ethnic Identification

Ethnic identiﬁcation has been rigorously studied in advertising research to
understand ethnic consumer responses to advertising stimuli (Whittler and DiMeo 1991;
Appaih 2000; Appaih 2001; Green 1999; Whittler and Spira 2002). Unlike situational
ethnicity and ethnic self-awareness which are often considered temporary states, ethnic
identiﬁcation is generally considered a personality trait (Deshpande, Hoyer and Donthu
1986; Forehand and Deshpande 2001; Dimofte et al. 2004). The rationale for this
approach is that one is born a member of an ethnic group, and does not usually leave the
group, and the degree of assimilation to the mainstream society depends on his/her
relationships with in- and out-groups (Marger 1999). Previous studies have stated that the
strength of ethnic identiﬁcation is likely to inﬂuence ethnic self-awareness (Deshpande’ et
al. 1986), and the marketing and consumer behavior literature also shows ethnic
identiﬁcation as an indicator of consumers’ attitudes toward products and behavioral
changes (Deshpande et al. 1986; Williams and Qualls 1989; Donthu and Cherian 1992;
Webster 1994; Chung and Fischer 1999; Lee, Fairhurst and Dillard 2002; Xu, Shim, Lotz
and Almeida 2004).

However, there is another approach to ethnic identiﬁcation and its likelihood of
being inﬂuenced by cultural cues. Kelman (1961) argued than when an individual
responds to social influence, there are mainly three processes one is likely to take
depending on the characteristics of the social inﬂuence: 1) compliance; 2) identiﬁcation;

3) internalization. The theory of identiﬁcation is the most applicable to the current study.

27

Identiﬁcation occurs when an individual adopts a behavior of another individual or a
group because the adopted behavior is associated with an acceptable and satisfactory
self-image; during the process, the individual assesses the level of similarity and
relevance with another individual or group (Kelman 1961; Appaih 2000). Different
identiﬁcation forms may take place depending on situations. For example, classical
identification takes place when an individual fully or partially accepts the inﬂuence from
the agent (Kelman 1961). In terms of advertising evaluations, the positive identiﬁcation
process could take place if a person feels familiar and similar to the models presented in
ads. Another form of identiﬁcation, a reciprocal role relationship, could also explain
consumer responses to advertising stimuli. In a reciprocal role relationship, both parties
are deﬁned with reference such as friendships and group memberships, and if the
individual ﬁnds a particular relationship satisfying and acceptable for self-deﬁnition,
he/she is likely to accept the inﬂuence (Kelman 1961). Based on this assumption, we can
argue that an individual may accept the inﬂuence of another person, such as a model in an
ad, if he/she ﬁnds the other person relevant and similar to him/herself. This example also
indicates that the degree of identiﬁcation is actually inﬂuenced by cues such as the
inﬂuence from a communicator This example suggests the possibility of ethnic
identiﬁcation being partially enhanced by environments or cultural cues, although
individuals are likely to have a certain base level of ethnic identiﬁcation that would not
be inﬂuenced by cultural cues.

Such identiﬁcation processes are not only between individuals and endorsers but

also between individuals and the social group to which they belong. Kelman (1961)

28

suggested that identiﬁcation may also take place when an individual attempts to maintain
their relationship with a group with which their self-deﬁnition is associated. This type of
ethnic identiﬁcation is likely to be an enduring state discussed by Deshpande et al. (1986).
For ethnic consumers, one way to self-deﬁne could be to recognize their ethnicity. Parker,
Sussman, Crippens, Elder, and Scholl (1998) state that the term, ethnic identity, refers to
perceptions of shared unique cultural traditions that are persistent across generations, and
ethnic identiﬁcation is the degree of an individual’s sense of belonging and group pride,
as well as prescribed norms, values, and social behaviors. Hence, consumers’ levels of
ethnic identiﬁcation may be represented by how strongly they feel associated with their
own ethnic group.

When examining one’s ethnic identiﬁcation, it can be argued that there are
mainly three ways of how one identiﬁes with his/her ethnicity based on the proposed
deﬁnition of ethnicity since the deﬁnition of ethnicity includes one’s physical
characteristics (race), values, attitudes, behaviors. First, one may identify with his
ethnicity due to the physical and/or racial appearance. For example, Asian Americans
may identify with an Asian model in an ad because of the similar physical appearance.
The second possibility is that one may identify with his/her ethnicity because he/she feels
that he/she belongs to the ethnic group due to similar values or attitudes. The
identiﬁcation process may take place for Asian Americans if they perceive an Asian
model representing the similar values or beliefs. The third case is that an individual may
identify with his/her ethnicity through similarly-conducted behaviors. For instance, Asian

Americans may identify with an Asian model because the actions or behaviors

29

represented by the model are similar to theirs.

In the current study, the hypotheses related to ethnic identiﬁcation include all of
these three levels of ethnic identiﬁcation, but some hypotheses place more emphasis on
physical appearance while some other hypotheses focus more on attitudes or feelings.
However, while speciﬁc hypotheses weigh more on a certain aspect of ethnic
identiﬁcation, such as model’s physical appearance, one’s ethnic identiﬁcation process
may involve other criteria such as assuming the model has a similar religious background
or values. Such individual differences in perception of an endorser or a model in an
advertisement are likely to depend on each person’s unique criteria. In this study, while
actual responses toward cultural cues presented in an ad, such as models from different
ethnic backgrounds, the criteria and perceptions used in the process of ethnic
identiﬁcation are not measured.

In particular, scholars have suggested that the strength of ethnic identiﬁcation,
along with ethnic origins, can explain differences and similarities among ethnic
consumers. Williams and Qualls (1989) argued that it is important to take the intensity of
ethnic identiﬁcation into account because if we neglect the intensity, the data from weak
ethnic identiﬁers may contaminate the analysis. The majority of these previous studies
with an emphasis on the strength of consumer ethnic identiﬁcation have been mainly on
either African American consumers (Williams and Qualls 1989; Whittler and DiMeo
1991; Green 1997; Appaih 2000; Appaih 2001), Hispanics (Desphande, Hoyer and
Donthu 1986; Donthu and Cherian 1992; Webster 1994), or international consumers (Tan

and Farley 1987; Lee et al. 2002). Since there is a lack of literature on Asian American

30

consumers and ethnicity, the following literature review will mainly focus on studies
concerning other ethnic minority groups in the United States.

In marketing, ethnic identiﬁcation has been acknowledged as an important
predictor of consumer behavior. For example, scholars have argued that consumption
patterns are strongly inﬂuenced by the strength of ethnic identiﬁcation. Deshpande et al.
(1986) and Donthu and Cherian (1992) found that the stronger the level of ethnic
identiﬁcation of Hispanic consumers, the more brand-loyal they are and more likely to
buy brands targeted to their ethnic group. Another study of Hispanic consumers
conducted by Webster (1994) indicates that high Hispanic ethnic identiﬁers tend to show
more of the ancestral Hispanic cultural characteristics such as male-dominance in
decision-making compared with weak Hispanic ethnic identiﬁers. Examining Asian
American consumers, Xu et al. (2004) indicate that those with a greater sense of Asian
ethnicity tend to be engaged in culture-speciﬁc consumption such as eating ethnic food
and enjoying ethnic movies and music compared with low Asian American ethnic
identiﬁers. Therefore, it is possible to argue that high Asian ethnic identiﬁers may tend to
react positively toward ethnic-oriented brands and advertising.

Other studies have actually examined the effect of ethnic identiﬁcation on
consumer responses toward advertisements featuring ethnic and racial cues such as
models from different ethnic groups. Similar positive consumer reactions toward the ads
with ethnic-oriented cues have been found. These studies suggest the function of ethnic
identiﬁcation as a mediator between a model’s race and consumer advertising/brand

evaluations and purchase intentions. Whittler (1989) found that African American college

31

students were more likely to purchase a product (laundry detergent) advertised by an
African American actor while highly racially-prejudiced white college students with
negative racial attitudes toward racial minority groups were less likely to obtain
additional information when they were shown the same advertisement. Along the similar
research stream, Whittler and DiMeo (1991) have examined the level of racial prejudice
held by white participants and the inﬂuence on perceived similarity of the models, brand
recognition, as well as the inﬂuence of model’s race in the ad. The results indicate that
high racially-prejudiced whites perceived an African American model in an ad as less
similar; however, low racially-prejudiced whites did not show any difference in perceived
similarity between white and African American models. In addition, high
racially-prejudiced whites tended to identify with the white model more than the African
American model; when the high racially-prejudiced whites saw the ad with the African
American model, their brand-narne recognition also declined (Whittler and DiMeo 1991).
Recent studies also show this trend.

Studying African American adolescents, Appaih (2000) found that African
Americans perceived African American models in ads more similar to themselves than
white models, and they were more likely to identify with African American characters
than with white characters. The researcher also indicated that Aﬁican Americans were
more likely to believe that ads with African American characters were intended more for
them than ads with white characters, and were likely to evaluate the African American
models and the ads more favorably than ads with Caucasian models (Appaih 2000).

Hence, a similar phenomenon can be hypothesized for Asian American consumers

32

regarding their evaluations of racially-congruent models. Asian Americans may be more
likely to identify with Asian models in advertisements. They may also think that the ads
are more targeted toward themselves when Asian models are featured in advertisements.
Along similar lines, Appaih (2001) closely examined the strength of ethnic
identiﬁcation as a key variable in the cognitive processing of advertising stimuli by
ethnic consumers. The ﬁndings show that participants with strong black ethnic
identiﬁcation perceived themselves more similar to and identiﬁed with African American
characters in ads than weak black ethnic identiﬁers. In addition, Green (1997) studied
African American females and their reactions to ads featuring models from difference
ethnic groups, and also took media placement and model display into account. Green
(1997) discovered that the strength of ethnic identiﬁcation played a signiﬁcant mediating
role in generating favorable attitudes toward racially-charged product ads (skin
foundation) with African American models compared with ads featuring Caucasian
models. She concluded that it appeared that high African American ethnic identiﬁers
tended to prefer ads featuring African American models in dominant positions more when
they were placed in racially-targeted media, while weak ethnic identiﬁers showed more
favorability toward ads featuring Caucasian models in dominant positions when placed in
the mainstream media (Green 1997). Green (1997) argued that weak ethnic identiﬁers
tended to favor Caucasian or Caucasian-dominant models along with the mainstream
media because they may have identiﬁed more closely with the dominant American
culture and were likely to consume mainstream media products where Caucasians were

portrayed often in a dominant position. A more recent study conducted by Whittler and

33

Spira (2002) suggests that African American consumers who highly identify with their
culture evaluated the product and ad featuring an African American model more
favorably than with a Caucasian model. According to the ﬁndings from these previous
studies that focused on African Americans in general, we can argue that for Asian
Americans, the strength of ethnic identiﬁcation may also be a key factor that can
inﬂuence their responses toward advertising and brands.

Moreover, ethnic identiﬁcation may serve as a predictor of perceived model
credibility. According to Kelman (1961), an individual may adopt the message from a
communicator because he/she ﬁnds the person relevant and/or similar to them. An
empirical study by Whittler and Spira (2002) supports this argument. The racial cue
(model’s race)'presented in an ad motivated high African American ethnic identiﬁers to
process the ad message in a biased way, leading them to rate the product advertised by the
Aﬁ'ican American model more favorably (Whittler and Spira 2002). In the context of the
current study, Asian American consumers may identify themselves with models from the
same ethnic background because of the similar cultural background and appearance, and
therefore, the messages from the Asian American models may be well-accepted by the
Asian American consumers. The study of Appaih (2001) found that both Asian American
and Hispanic consumers tended to perceive themselves more similar to African American
models than white models in the ads, and they rated the ad with the African American
models more favorably. The researcher stated that the ﬁndings may be because the
adolescents identiﬁed themselves more with African American youth culture and positive

racial stereotypes such as “urban, hip, cool, athletic, and trendsetters “(Appaih 2001, p.

34

44). As these scholars argue, it appears that the strength of ethnic identiﬁcation inﬂuences
ethnic consumers’ responsiveness to racial cues presented in advertisements. Additionally,
these examples indicate that ethnic identiﬁcation may mediate the relationship between
model’s race and product type and the resulting perceived model credibility which may
inﬂuence advertising/brand evaluations and purchase intentions. Hence, the current study
seeks to investigate the potential causal relationship between Asian ethnic identiﬁcation
and perceived model credibility and its effect on advertising evaluations. However,
another important variable that may inﬂuence perceived model credibility is cultural
congruency.
The Role of Cultural Congruency
Congruency Theory

The concept of cultural congruency may be helpful to better understand the
relationship between the strength of ethnic identiﬁcation and perceived source credibility.
Previous literature on source credibility in an advertising context has raised the issue of
congruency as an inﬂuencing agent (Kamins 1990; Kamins and Gupta 1994). Likewise,
the current study plans to examine the role of congruency in Asian Americans’ cognitive
processing of advertisements. In particular, this study argues that cultural congruency
may enhance the effect of perceived model credibility and advertising evaluations.

The idea of congruency as a predictive agent of attitude change suggests that
attitude changes, or evaluation changes are likely to take place when they are congruent
with the existing frame of reference (Osgood and Tannenbaum 1955). Osgood and

Tannenbaum (1955) argued that existing attitudes toward an object or issue prior to

35

reception of the message play an important role in determining the level of congruity and
its inﬂuence on attitude changes. When there is a distance between the preexisting
attitude toward the object/source/issue and the actual location of the object/source/issue,
some pressure toward congruity is likely to take place (Osgood and Tannenbaum 1955).
For example, when an individual has a favorable attitude toward an endorser in an ad,
and has a neutral attitude toward the product type advertised, it is likely that his/her
attitude toward the advertised product becomes congruent with the attitude toward the
endorser. In relation to source credibility, Osgood and Tannenbaum (1995) state that the
more favorable the attitude toward the communicator becomes, the greater the effect of a
positive assertion on an object/issue associated with the communicator, which results in a
more favorable attitude toward the object/issue. On the other hand, if the communicator
favorably rated by the receiver makes a negative assertion about the issue/object, the
resulting negative attitude change is likely to be enhanced (Osgood and Tannenbaum
1955)

Applying this idea to advertising evaluations of Asian Americans, it is possible
to argue that if Asian American consumers have an existing attitude toward their own
ethnic identiﬁcation, their attitudes may inﬂuence the evaluations of an ad featuring
ethnic minority models. Speciﬁcally, one may be able to predict that high Asian
American ethnic identiﬁers tend to rate ads with Asian models more favorably because
the ad rating is congruent with their level of ethnic identiﬁcation. As this example
suggests, the congruency effect may have a notable inﬂuence on advertising evaluations

of Asian Americans.

36

When measuring these cultural antecedents, it is recommended to include both
individual-level items, such as one’s own cultural values, and group-level items, such as
one’s membership in a geographical region or race to better understand ethnic consumer
behavior. This is particularly important to better understand Asian American consumers
because of their tendency to be group-oriented, which is based on collectivism (Hofstede
1997) and Conﬁicianism (Keller and Kronstedt 2005).Because of the heavy emphasis on
relationships and group-orientation in Asian culture, individuals in Asian culture are more
prone to social pressure to follow the group norms and rules (Ang, Cheng, Lim and
Tambyah 2001). Such group norms are likely to come from culture, and Triandis (1989)
argues that culture inﬂuences every level of an individual’s thinking and behavior,
including his/her attitudes toward consumption and ads. Therefore, it is essential to
combine both individual and group level variables to study ethnic minority consumers,
particularly Asian American consumers.

Similarity vs. Familiarity

In order to investigate the effect of congruency, it is necessary to address its
possible indicators. In terms of communication effectiveness, similarity and familiarity
are often discussed as potential indicators that inﬂuence message effectiveness through
perceived attractiveness of the source. According to McGuire (1985) and Ohanian (1990),
message effectiveness is likely to depend on the source’s similarity, familiarity, as well as
attractiveness, to the receiver. In particular, perceived similarity appears to be an
important indicator of cultural congruency that may have a strong impact on perceived

model credibility and advertising evaluations.

37

Based on the proposed deﬁnition of ethnicity, the approach of the current study
to the concept of similarity reﬂects the essence of physical characteristics, values,
attitudes, and behaviors. Tversky and Gati (1978) describe similarity as common features
between an individual and others, while the state of dissimilarity can be explained by
distinctiveness between the individual and others. Whittler and DiMeo (1991) measured
similarity with an actor in an ad with items that reﬂect lifestyle, appearance, cultural
background, and values. The current study also incorporates the approach of Whittler and
DiMeo (1991) to examine perceived similarity. In other words, perceived similarity
between models presented in ads and consumers may be determined by perceived
resemblance based on lifestyle, physical appearance, cultural background, and/or values.

Although somewhat related, the idea of familiarity differs from similarity.
Yzerbyt, Judd, and Comeille (2004) state that one’s familiarity with an object or issue is
represented by the knowledge about the object or the issue. Such familiarity may be
achieved by perceived resemblance between consumers and endorsers. However, while
perceived familiarity does not necessarily require the resemblance (or similarity), it is
based on previous knowledge about the object or the issue. In terms of this study, while
Asian Americans may perceive an Asian model familiar because of the perceived
similarity between themselves and the model which may have been built on their prior
knowledge about the ethnic characteristics, they may not necessarily perceive the model
similar because of perceived familiarity with the model due to the previous knowledge
about the model’s characteristics. Therefore, the focus is placed upon similarity as one of

the cultural antecedents of perceived model credibility and advertising evaluations.

38

Congruency Research in Advertising

In discussions of the role of congruency in advertising evaluations, two types of
congruency are relevant in advertising research: (1) congruency between personal
attributes of the audience and attributes featured in advertisements such as models and
endorsers (Whittler and DiMeo 1991; Green 1997; Appaih 2002; Chawla and Nataraajan
1995; Gerristen, Korzilius, van Meurs and Gijsbers 2000); and (2) congruency between
advertising attributes including endorsers’ attributes and product advertised (Kahle and
Homer 1985; Cohen 1992; Green 1997). Identiﬁcation theory (Kelman 1961) helps to
again explain this congruency phenomenon. For example, if an individual’s
characteristics are congruent with advertising attributes such as model/endorser, he/she is
more likely to identify his/herself with the ad because such an attribute is similar and
relevant to the person (Kelman 1961; Basil 1996; Appaih 2000). In such a case, source
similarity and familiarity may enhance the congruent effect either in a positive or
negative direction on perceived source credibility and/or advertising evaluations. In terms
of congruency between one’s own characteristics and advertising stimuli, several studies
have shown support for this argument.

Zhang and Gelb (1998) examined appeals in American and Chinese ads to assess
their relationship with the cultural values present in each country. The researchers also
took product use conditions (socially visible use or private use) into account. The
ﬁndings suggest that when advertising appeals are congruent with cultural values, the
match is likely to result in favorable consumer attitudes toward the ads (Zhang and Gelb

1998). Similar results were found in a study that focused on Mexicans; advertisements

39

that portrayed local cultural values were rated more favorably and the resulting purchase
intention was greater compared with advertisements with inconsistent cultural norms
(Gregory and Munch 1997). Zhang and Gelb (1998) also pointed out that although
cultural congruency is preferable to achieve favorable consumer attitudes in general,
employing culturally incongruent advertising stimuli may still be acceptable depending
on the product use, as long as there is congruency between the cultural appeal and
product use conditions (Zhang and Gelb 1998). For example, it may be effective to use
individualistic appeals in ads targeted toward Chinese audiences who are considered to be
more collectivistic by nature, if the product use is individualistic (Zhang and Gelb 1998).
Therefore, product type is also an important variable to assess when examining
congruency issues in ads and will be discussed in a later section.

Another study conducted by Lwin and Wee (2000) focused on how audiences
from three countries, namely Australia, Myanmar, and Singapore, evaluated TV
commercials with various combinations of a spokesperson’s race and the use of different
English accents. Overall, the results indicate that Caucasian spokespersons having
Western accents received the highest score in terms of advertising evaluations. Based on
Kelman’s (1961) identiﬁcation theory, the researchers stated that the result may be due to
the overwhelming media exposure to Caucasians world-wide that could increase the
familiarity with Caucasians; therefore, the audience may have identiﬁed well with
Caucasian models (Lwin and Wee 2000). Moreover, a study of Dutch consumers
responses to TV commercials with English texts and voice-over reveals that in general,

the use of English language in TV commercial was not appreciated by Dutch consumers,

4O

and the resulting advertising responses were negative (Gerritsen, Korzilius, van Meurs,
Gijsbers 2000). It appears that congruency between the cultural cues in ads and
consumers’ beliefs about these cues have a signiﬁcant impact on their responses to
advertising.

Several other studies have attempted to examine the effect of congruency
between personal attributes of consumers, such as the race of an individual, and a model’s
race presented in ads. These studies have attempted to examine the effect of
similarity/relevance with endorsers on advertising evaluations, and found that African
Americans tended to perceive African American models in ads more similar than
Caucasian models, and their responses to the ad with African American models were
more favorable than ads with Caucasian models (Whittler and DiMeo 1991; Green 1997;
Appaih 2000, Appaih 2001, Whittler and Spira 2002). Feldman (1984) also found that
when congruency existed between the audience’s race and communicator’s characteristics
including race, the audience was more likely to accept the information. Moreover,
congruency between a consumer’s race and a model’s race appears to have a strong
impact on a consumer’s response to advertising even for members of an ethnic majority
group. Chawla and Nataraajan (1995) studied Caucasian consumers’ evaluations of ethnic
food advertisements. They manipulated the endorser’s race (Asian/Caucasian) and gender
in an ad of a soy sauce brand. Based on McGuire’s (1985) familiarity effect on attitude
changes along with source credibility (Hovland et al. 1953), the researchers discovered
that when the race of the endorser was Caucasian, the level of perceived model credibility

became higher compared with the case where the endorser’s race was Asian (Chawla and

41

Nataraajan 1995). This example suggests that it is possible for congruency between
personal characteristics and the advertising cues to overpower congruency between the
advertising cues and the advertised product type. Therefore, because Asian Americans
may ﬁnd Asian American models more familiar than Caucasian models due to the similar
physical characteristics and/or cultural backgrounds, Asian American models may induce
a more favorable attitude of Asian American consumers toward the ad and the brand than
non-Asian models. Likewise, Asian Americans may regard Asian American models more
familiar than Caucasian models, and therefore, they may perceive Asian American
models more credible than Caucasian models.

However, a model’s race and the type of product have also been shown to affect
consumer responses to advertising. Several studies have been conducted about the
inﬂuence of the image of an endorser or a spokesperson in an ad on consumer responses
to advertisements. The match-up effect, proposed by Kahle and Homer (1985) to
understand the effect of celebrity endorsers, argues that the effectiveness of
communication can be enhanced by a match between the message and/or appeal
communicated by the celebrity endorser’s image and the advertised product. Based on
this argument, Kamins and Gupta (1994) found that the more congruent the endorser’s
image and product type, the more likely that viewers perceive the spokesperson as more
believable and attractive and the more favorable consumers’ attitudes toward the
advertised product because there was consistency between the advertised product and the
spokesperson’s image. A similar ﬁnding was reported by O’Mahony and Meenaghan

(1998), stating that when the celebrity endorser’s expertise matches with the product

42

advertised, consumer evaluations of the ad are more favorable than when there is not a
match between the two agents.

Green’s study (1997) on African American female consumers’ reactions to ads
with African American and Caucasian American models suggests that there is an
interaction effect of product type; when African American models were featured in
racially-charged product ads such as skin foundation, the advertising evaluation of
Aﬁ'ican American females were more positive than Caucasian models. Another study
which examined this type of congruency focused on Caucasian consumers. Cohen (1992)
found that when Asian American models advertised a technology-related product, the
advertising rating by Caucasian consumers was greater compared with other product
types. In this case, a stereotype of Asians being more competent in the ﬁeld of technology,
and its congruency with the product type may have affected perceived model credibility
(particularly the expertise component of perceived model credibility), and therefore,
positively inﬂuence the favorability of the ad. Hence, it is possible to suggest that two
types of congruency, in addition to the other cultural variable of strength of ethnic
identiﬁcation are likely to mediate the effect of perceived model credibility. In fact,
speciﬁc relationships with each of the credibility components can be predicted.

With regards to the relationship between congruency and perceived source
trustworthiness, it is possible to argue that a positive congruency effect may lead to a
positively-perceived trustworthiness of the source. As discussed earlier, Jones et al.
(1990) found that when the race and gender of the salespeople were congruent with

customers (in this case, African American), the customers tended to perceive the sales

43

associates more trustworthy than Caucasian sales associates. Hence, in this study, Asian
Americans may ﬁnd Asian models in ads more trustworthy than Caucasian models
because there is a match between their own race and the model’s race.

A similar positive congruency effect may take place on perceived attractiveness
of Asian American models. Brewer (1979) argued that individuals tend to compare and
categorize others as well as themselves either as in-group or out-group based on criteria
such as race, age, and lifestyle, and they tend to evaluate in-group people who are similar
to themselves more favorably than out-group people who are dissimilar. Because there is
a larger social distance between an individual and members of the out-groups, he/she is
likely to rely on positive bias toward members of their in-group when they make .
comparisons and/or evaluations (Brewer 1979; Green, 1999). Also, McGuire (1985)
argued that source attractiveness in relation to message effectiveness and communication
tends to depend on the source’s familiarity, similarity, and overall attractiveness. If Asian
American consumers ﬁnd Asian models more familiar and similar to themselves, they
may ﬁnd Asian models attractive, resulting in favorable perceived source credibility and
advertising evaluations. Based on this assumption, it is possible to argue that due to the
social distance between Asian Americans and Caucasian Americans, Asian Americans
may be more prone to have a positive bias on Asian American models in advertisements
if they have a stronger sense of ethnic identity. Or, they may simply become positively
biased toward Asian models because of the similar physical appearance, resulting in a
higher rating for perceived model attractiveness.

In terms of the relationship between the source’s expertise and congruency effect,

44

it is likely that congruency between advertising cues such as model’s race and product
type positively inﬂuences the resulting advertising evaluations. Green (1997) found when
African American models were featured in an ad for skin foundation, a racially-charged
product, the resulting advertising evaluations and purchase intentions of African
American consumers tended to be more favorable than those for ads with Caucasian
models. It appears that in this case, because race is relevant in selecting the product,
consumers tend to make the connection between the race of the models and the product
type, and ﬁnd the models having more expertise in the product category when their race
is similar to that of the models. As a result, the perceived model credibility may be
increased due to the congruency effect, and positive attitude changes may take place.
Based on these previous studies, we can argue that Asian Americans may ﬁnd
the model of the same race more trustworthy, with greater expertise, and more attractive
and therefore more credible when the model is featured with race-related products such
as cosmetics. In other words, Asian American consumers may ﬁnd an Asian American
model in the ad more credible if there is congruency between their own personal
attributes and the model in the ad, as well as congruency between the advertised product
and the model’s expertise, trustworthiness, and attractiveness. Based on these

assumptions, the following hypotheses on perceived model credibility are proposed.

45

CHAPTER TWO
HYPOTHESES

Ethnic Identiﬁcation

According to Kelman’s (1961) identiﬁcation theory, a positive identiﬁcation
process takes place when an individual ﬁnds familiarity and/or similarity with an object.
In particular, if an individual has a reciprocal role relationship with his/her group, and
ﬁnds the relationship acceptable for deﬁning him/herself, he/she tends to accept the
inﬂuence of the group (Kelman 1961). Hence, Asian Americans may go through a
positive identiﬁcation process when they see an Asian model in an advertisement.
Additionally, if Asian American consumers have a reciprocal role relationship with their
ethnic group, they may be more prone to accept the inﬂuence of another person from the
group, such as an Asian American model in an advertisement, if he/she ﬁnds the model
relevant and similar to him/herself. Previous studies also found that the degree of ethnic
identiﬁcation tends to inﬂuence the advertising responses of ethnic minority consumers,
but not necessarily for members of the majority group (Whittler and DiMeo 1991; Green
1997; Appaih 2001). Hence, the following hypotheses are proposed.

HI: Asian American consumers will rate the ad with Asian models more
favorably than the ad with Caucasian models.

HIa: Caucasian consumers will not rate the ad with Caucasian models more
favorably than the ad with Asian models.

H2: Asian American consumers’ ethnic identification will be positively

correlated with their attitudes toward advertisements with Asian American models.

46

Perceived Similarity

The perceived similarity with a model in an advertisement is considered to
affect a consumer’ ethnic identiﬁcation process (Whittler and DiMeo 1991; Green 1997;
Appaih 2000) and perceived source credibility (McGuire, 1985; Ohanian, 1990). In
particular, a positive identiﬁcation process that results in acceptance of the inﬂuence
from the endorser in an ad may be due to perceived similarity with the endorser. Also,
regarding communication effectiveness, perceived source attractiveness, one of three
components of source credibility, can be inﬂuenced by perceived similarity with the
source (McGuire 1985). Likewise, the other two components of source credibility,
source expertise and trustworthiness, may be inﬂuenced by perceived similarity with the
source. Therefore, the additional hypotheses related to perceived similarity can be stated.

H3." Perceived similarity with the model in the ad will be positively correlated
with perceived source credibility.

H3a: There is a positive relationship between perceived similarity and i)
attractiveness; ii) trustworthiness; and iii) expertise.

H4: Perceived similarity with the model in the ad will be positively correlated
with the advertising evaluation.

H4a: Perceived similarity with the model in the ad will be positively correlated
with ethnic identification.
Ad Schema Incongruity

Mandler (1982) argues that when there is a moderate mismatch between one’s

prior expectation about an issue/object and the actual value of the issue/object, the

47

resulting attitude toward the issue/obj ect is likely to be more favorable compared with the
cases of schema congruity and extreme schema incongruity. In the context of the current
study, if Asian American consumers have a prior expectation of seeing Caucasian models
in advertising in general, their attitudes toward an advertisement with Asian American
models may be more favorable. This may occur because they may ﬁnd this partial
mismatch between their expectation about seeing Caucasian models in advertising and
actually seeing an Asian American model in an advertisement as a positive situation.

Also, it is possible to argue that consumer ethnic identiﬁcation may inﬂuence the
degree of advertising schema incongruity because high ethnic identiﬁers may be more
sensitive and conscious about not seeing Asian American models in general advertising,
and when they actually see Asian models in ads, they may ﬁnd them as a pleasant
surprise. Also, advertising schema incongruity may result in positive evaluations of
advertisements with Asian models by Asian American consumers because of the pleasant
surprise to see Asian models in advertisements. Based on these examples, the following
hypotheses are proposed.

H5: Asian Americans’ ethnic identiﬁcation will be positively correlated with
advertising schema incongruity.

H6: Advertising schema incongruity will be positively correlated with Asian
American consumers ’evaluations of advertisements with Asian American models.
Model Congruency

Osgood and Tannenbaum’s (1955) congruency theory states that when there is

congruency between an individual’s expectation and/or frame of reference and the actual

48

object, a positive attitude change is likely to occur. In the current study, two types of
congruency exists: 1) congruency between consumers’ personal attributes and attributes
featured in advertisements (Whittler and DiMeo 1991; Green 1997; Appaih 2001); and 2)
congruency between advertising attributes such as models’ attributes and the advertised
products (Kahle and Homer 1995; Cohen 1992; Green 1997). In the context of the current
study, we can argue that when Asian American models are used in the ad of
racially-charged products, Asian American consumers may evaluate the models and the
ad more favorably due to the congruency between the model’s race and the product type,
as well as congruency between their racial/ethnic background and models’ race. On the
other hand, previous ﬁndings imply that members of the majority group do not
necessarily show preference for ads featuring racially-congruent models, regardless of the
product type (Whittler and DiMeo 1991). Therefore, the following hypotheses are
proposed:

H7: Asian American consumers will ﬁnd the ad with Asian models more
favorable than Caucasian models when the models appear in the ad for a
racially-charged product.

H7a: Caucasian American consumers will not ﬁnd the ad with Caucasian
models more favorable than the ad with Asian models when the models appear in the ad
for a racially-charged product.

Model Credibility
In this study, cultural congruency relationships are hypothesized as inﬂuencing

factors on the perceived credibility of the models. Speciﬁcally, congruency between

49

Asian Americans’ own ethnicity and the models’ race may affect the three factors of
source credibility, attractiveness, expertise, and trustworthiness due to the similarity
between these two groups. In addition, perceived expertise may be inﬂuenced by
congruency between the models’ race and the product type; when a product is based on
and targeted toward their ethnicity and race, consumers may ﬁnd models with their own
race more credible as they may perceive the models’ race as a representation of expertise.
Based on these arguments, the following set of hypotheses are proposed regarding model
credibility.

H8: Asian (Caucasian) Americans will ﬁnd Asian (Caucasian) models in an ad
more credible than Caucasian (Asian) models.

H8a: Asian Americans will ﬁnd Asian models in an ad more attractive than
Caucasian models.

H8b: Asian Americans will ﬁnd Asian models in an ad as having more expertise
than Caucasian models.

H8c: Asian American will ﬁnd Asian models in an ad more trustworthy than
Caucasian models.

H8d.‘ Model :9 race will mediate the relationship between Asian American
consumers ’ethnic identiﬁcation and perceived model credibility.

In addition, one’s strength of ethnic identiﬁcation may inﬂuence perceived
model credibility and serves as an inﬂuencer in the cognitive processing of Asian
American consumers. For example, if one goes through a positive identiﬁcation process,

he/she may be more likely to pay attention to racially congruent cues such as models’

50

race due to the physical and cultural similarity shared between themselves and models
presented in the ad. In addition, such positive identiﬁcation may enhance perceived
model similarity, leading to more perceived credibility for the Asian model in the
advertisement. The mediation effect of perceived similarity may also inﬂuence the
relationship between ethnic identiﬁcation and resulting advertising responses. As a result,
the following hypotheses can be proposed.

H9: Perceived model similarity will mediate the relationship between Asian
Americans ’ethnic identification and perceived source credibility.

H10: Perceived model similarity will mediate the relationship between Asian
Americans ’ethnic identification and advertising evaluations.

Previous studies show that source credibility inﬂuences consumers’ evaluations
of advertisements (Kamins 1990; Kamins and Gupta 1994; Tripp et al. 1994;
O’Mahony and Meenaghan 1998), brands (Till and Busler 1998), and purchase
intention (Ohanian 1991). Therefore, the following hypothesis is proposed.

H1]: The degree of perceived model credibility is positively correlated with
Asian American consumer attitudes toward a) advertisements, b), advertised brand, and
c) purchase intention.
Attitudes toward Ad, Brand, and Purchase Intentions

Previous literature suggests that consumer attitudes toward an advertisement
may inﬂuence their attitudes toward the advertised brand and the subsequent purchase

decisions (Batra and Ray 1986; MacKenzie, Lutz and Belch 1989).

51

H12: Attitudes toward the advertisement will inﬂuence attitudes toward the
brand and this will inﬂuence attitudes toward purchase intentions.

Moreover, the level of product involvement will be treated as a covariate in this
study as previous researchers argue that product involvement may inﬂuence perceived
source credibility as well as the resulting advertising evaluations (Gotlieb and Sarel 1991;
Mazursky and Shul 1992).

Conceptual Model: Asian American Consumers ’Cognitive Processing
The cognitive processing of Asian American consumers can be explained in the

following model.

Figure 1: Conceptual Model: Cultural Antecedents of
Perceived Source Credibility

 

 

 

 

 
 
 

 

 

 

3
Advertising\ HS : $323:
Schema the ad
H5 Incongruity
H4

 

 
     

 

  

 

 

 

  

 

 

 

    
    

Ethnic ( Source toward
Identiﬁcation H4 Credibility
:1
- l-19 _Trustworthlnes-
l ‘l Similarity H3 ' Expertlse

 

Attractiveness

Purchase
Intention

    
 

[Model’s Race]\
[ Product J/ H1 H7
Type

52

 

 

 

 

CHAPTER THREE
THE STUDY

The purpose of the current study is to examine the inﬂuence of the cultural
antecedents such as the strength of ethnic identiﬁcation, similarity, and cultural
congruency of model, product, and consumer, on perceived model credibility. The study
also examines the potential effects of these variables on Asian American consumers’
responses to advertising. In addition, the study will examine how Asian American
consumers’ cognitive processing of perceived credibility led by the cultural antecedents,
will affect advertising responses.

Despite previous research efforts to investigate source credibility and its effect
on persuasion, the majority of such studies have treated the construct as an independent
variable, and focused mainly on trustworthiness, expertise, or attractiveness of the source
(Pompitakpan 2004). However, exploring other potential factors that may inﬂuence
source credibility will provide a better understanding of the mechanism of consumers’
information processing and their reaction to advertising stimuli. Moreover, when studying
ethnic consumers, the cultural inﬂuences on their attitudes and behaviors represented by
model/product congruency and the strength of ethnic identiﬁcation should not be
neglected. Therefore, the underlying process of perceived model credibility will be
proposed to assess the overall inﬂuence of relevant variables on responses to advertising.
Speciﬁcally, Asian Americans’ levels of ethnic identiﬁcation will be examined along with
cultural congruency factors to understand their relationships with perceived source

credibility.

53

Since endorser credibility can inﬂuence consumer advertising evaluations
(Ohanian 1990), it is important for marketers to understand the mechanism of source
credibility in order to produce effective advertisements. Given changes in the US.
population and the trend of global marketing, it is essential to better understand the
credibility construct based on cultural variables. Each ethnic community group is unique
and different from others. For example, Asian Americans have been considered the
“model minority,” and typically have a higher level of income and education compared
with the other ethnic minority and majority groups (Hurh and Kim 1989). Thus, the same
outcomes and/or processes regarding perceived source credibility should not be assumed
without systematical testing. The proposed study is also unique and important because the
cultural variables being examined in this study have never been assessed within one
model. Hence, it is hoped that the current study will contribute to enhance the current

knowledge about source credibility and its interaction with ethnicity and culture.

54

CHAPTER FOUR
METHOD
Research Design

The current study used a 2 (model’s race: Asian/Caucasian) x 2 (product type:
racially-charged or not) x 2 (participants’ race: Asian American/Caucasian) factorial
experimental design (see Table 1). Four print ads where model’s race and product type
were manipulated were created for the study. A pretest was undertaken for the selection of
brand name, models, and products to develop the manipulation and questionnaire. Forty
six Asian and Caucasian students attending Michigan State University (MSU) were
recruited for the pretest by the method of snowball sampling. As an incentive, a drawing
for a $20 online gift certiﬁcate was held.

For the main study, female Asian and Caucasian American students attending
MSU were recruited with the assistance of the university’s registrar’s office. Female
subjects were apprOpriate for the current study because of the possibility of using skin
foundation as a racially-charged product category (Green 1997), which is an irrelevant
product category for men in general.

The selection of Asian American females from the registrar’s ofﬁce was based
on self ethnicity reports gathered by the university. In addition to the ethnicity
classiﬁcation provided by the registrar’s office, the actual questionnaire also had a section
where respondents could self-report their ethnicity to ensure accurate identiﬁcation. The
completed questionnaire with one of four ad manipulations, as well as a self-addressed

and stamped envelope was mailed to 1,058 Asian American subjects from the registrar’s

55

ofﬁce. Two weeks after the original packages were sent, reminder postcards were sent to
Asian American female subjects by the registrar’s ofﬁce. A total of 143 Asian American
responses were collected in three weeks, which resulted in a response rate of 13.5 percent.
In addition, data from 121 female Caucasian American students were collected as a
control group by visiting several classes offered by the Department of Advertising, Public
Relations, and Retailing at MSU. The survey was administrated in class, and the data
were collected from all of the students in the classes visited. However, only responses
from female Caucasian and Asian American students were used for the data analysis.
Therefore, the total initial number of participants was 264.

The method of a mail questionnaire to recruit female Asian Americans was used
rather than administrating the survey on site (such as, administrating the survey at a
cultural/ethnic organization) because Asian Americans who belong to such organizations
may have a greater degree of ethnic identiﬁcation compared with those who do not
participate in such organizations. In order to avoid smaller variance in ethnic
identiﬁcation, mailing the questionnaire with the assistance of the registrar’s ofﬁce was
helpful to equally obtain data from both high and low Asian American ethnic identiﬁers.
Moreover, a similar “Caucasian” organization did not exist. As an incentive, the
participants who completed and returned the survey were entered into a drawing with a

chance to win one of four $50 gift certiﬁcates.

56

Table 1: Experimental Design

 

 

 

 

 

 

 

 

 

 

Asian model] Asian model/ Caucasian model/ Caucasian
Racially-charge Non Racially-charged model/
(1 product racially-charged Product Non
product racially-charged
product
Asian
American
Consumers
Caucasian
Consumers
Stimuli

Four full-color magazine ads where model’s race and product type were
manipulated were created for the main study. The models, the brand name, and the
product type were pre-tested. A total of 27 Asian and 19 Caucasian students attending
MSU volunteered to participate in the pretest. They were asked to rate the attractiveness
of a series of both Asian and Caucasian American models, and the degree to which
product categories (out of 14 potential product categories) sounded racially-charged.
Racially-charged products were deﬁned as products that an individual’s race would be
important to select the product, such as the physical features of skin color or the texture
of hair. Skin foundation, for example, can be considered as a racially-charged product
because one’s race is likely to inﬂuence the color selection. For racially-charged products,
items such as skin foundation, shampoo, and hair-coloring products were used for the
pretest. For non-racially-charged products, products such as perfume, jewelry, an iron,
and soft drinks were tested in the pretest.

For pretesting the models, all models were dressed similarly and were

57

photographed in a similar pose. The models chosen for the main study (one Asian and
one Caucasian American) were selected based on similar ratings of attractiveness
measured by a 7-point Likert scale proposed by Ohanian (1990). The participants rated
the models similarly attractive (Asian model M=5.29; Caucasian model M=5.08; t=1.73,
df=43, p>.05). The brand name, A/Z, was selected as the most racially-neutral brand
name (M=2.86) out of 10 potential brand names. For product selection, skin foundation
was selected as the most racially-charged product (M=5.8l), and an iron (M=1.40) was
selected as a non-racially-charged product (t=33.43, df=43, p<.001). The pretest results
indicated that no two product categories were equally in terms of product involvement;
therefore, this variable was treated as a covariate in the following hypothesis testing.
Examples of the four ads created for the ﬁnal survey are shown in Appendix.

Measures

Independent Variables: Ethnic identiﬁcation was measured with the 18-item Likert-type
scale developed by Xu et al. (2004). A factor analysis with a varimax rotation was
conducted, and the ﬁrst factor consisted of ﬁve attitudinal items such as “I have a strong
sense of belonging to my own ethnic group,” “I feel good about my cultural or ethnic
background,” and the second factor consisted of seven behavioral items such as “I prefer
to hang out with friends from my own ethnic group,” “I prefer watching TV channels or
shows that are relevant to my ethnic group,” and “I frequently read newspapers and
magazines that are targeted to my ethnic group (Xu et al. 2004). In order to have an
ethnic identiﬁcation measure that reﬂects both attitudinal and behavioral aspects, these

two factors were combined and used as a 12-item ethnic identiﬁcation scale (a=.85).

58

Forty-eight percent of the total variance was explained by these two factors. In addition,
perceived model similarity was measured with a four-item Likert-type scale originally
created by Whittler and DiMeo (1991). The scale included questions such as “the model’s
overall lifestyle is similar to mine,” “Cultural background is similar to mine,” and “Basic
values are similar to mine,” and the scale was found to be reliable (a=.79).

Ad schema incongruity was measured with a Likert-type scale created for this
study, which is anchored by four items such as “Seeing Asian American models in ads is
very common (reverse-coded),” “It is unusual to see Asian American models in
advertising,” “I am surprised to see Asian Americans in advertisements,” and “In general,
advertisements rarely feature Asian Americans.” The higher the score was, the more that
ad schema incongruity existed in respondents’ mind because the higher score indicated
that respondents did not expect to see Asian models in advertisements, which may have
been regarded as a pleasant surprise for Asian American consumers. The scale was found
to be reliable (a=.77).

Model credibility was measured by a 15-item semantic differential scale
developed by Ohanian (1990), which included items such as attractive/unattractive,
trustworthy/untrustworthy and knowledgeable/unknowledgeable. The overall scale was
found to be reliable (a=.88) and internally consistent with three factors with ﬁve items
each: attractiveness (a=.82), trustworthiness (a=.87), and expertise (a=.86).

Dependent Variables: Two of the three dependent variables, attitude toward ad (6 items,
a=.82) and attitude toward brand (9 items, a=.92) were measured by a 7-point semantic

differential scale using items developed by MacKenzie et al. (1986). Purchase intention

59

was measured by a combination of three items from the 7-point scale developed by
Putrevu and Lord (1994) and three original items created for the current study (a=.86).
Also, the level of product involvement was measured by a combination of the 5-item
semantic differential scale by Zaichokowsky (1986) and one item created for this study
(a=.93), and used as a covariate because previous literature suggests that product
involvement can inﬂuence source credibility and the variable has an effect on advertising
evaluations (Gotlieb and Sarel 1991; Mazursky and Shul 1992). Demographic items that
were measured include age, income, self-report ethnic identity and the strength of ethnic
identity, education level, major, and nationality.
Other Measures

For the purpose of the manipulation check of model’s race, participants were
asked to identify the racial background of the model featured in the ad. To rate model’s
attractiveness, Ohanian’s (1990) model attractiveness scale was used. Also, two
dichotomized questions that ask whether or not participants have seen the model or the
brand previously were included in the main survey. In addition, participants were asked to
rate the degree to which they perceive the product portrayed in the ad was

racially-charged by using a seven-point semantic differential scale.

60

CHAPTER FIVE
RESULTS

Sample Characteristics

Of the 143 responses from Asian American females, 27 participants identiﬁed
themselves as Asian Indians. Since the operational deﬁnition of Asian Americans
proposed for the this study did not include Asian Indians as Asian Americans, these cases
were discarded from the ﬁnal analysis. Therefore, the sample size was down to 237 (116
Asian Americans; 121 Caucasian Americans). All eight conditions had a cell size greater
than 20, which is essential to ensure statistical power (Table 2).

Table 2: Numbers of Cases per Manipulation

 

 

 

 

 

 

 

 

Asian model/ Asian model/ Caucasian Caucasian

Racially-charged Non model/ model!

product racially-charged Racially-charged Non

product Product racially-charged
product

Asian 32 31 27 26
American
Respondents
Caucasian 25 3O 33 32
Respondents

 

 

The majority of participants were in their 208, with the largest group being age
20 and under (49. 8 percent), followed by 21 to 30 years of age (45.1 percent).
Approximately 92 percent of the subjects were full-time undergraduate students. Ninty
percent of all participants were US. citizens, and 10 percent were permanent residents
(greencard holders). Seventy-three percent of the participants were born in the US, with

approximately 45 percent being third generation or beyond.

61

Regarding Asian American ethnic groups, Chinese Americans were the largest
group (27.6 percent), followed by Korean Americans (22.4 percent), Vietnamese
Americans (20.7 percent), Filipino Americans (13.8 percent), Japanese Americans (10.3
percent), and others (5.2 percent). Eighty-four and a half percent of the Asian American
participants were US. citizens, and 15.5 percent held permanent resident status. About
ﬁfty percent of the Asian American participants were born in the US, while 50 percent
was born abroad. In terms of generation, 45.7 percent of the Asian Americans answered
that they were the ﬁrst generations, followed by those indicating second generations (19
percent). See Table 3 for details on demographic information.

Table 3: Asian and Caucasian American Sample Characteristics

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Asian Americans Caucasian Americans
Legal Status
US. Citizen 84.5% US. Citizen 96.7%
Permanent 15.5% Permanent Resident 2.5%
Resident
Born in the US.
Yes 50.4% Yes 95.8%
No 49.6% No 4.2%
Generation
1St Generation 45.7% 1st Generation 7.5%
2"d Generation 19.0% 2"d Generation 10.0%
3“I Generation + 13.8% 3rd Generation + 85.8%
Not a US. Citizen 13.8% Not a US. Citizen 2.5%

 

 

 

 

62

 

Manipulation Checks

All participants who saw the ad with the Asian model correctly identiﬁed the
model’s race, while one out of 119 participants who saw the ad with the Caucasian model
incorrectly identiﬁed the model’s race; therefore, this misidentiﬁed case was discarded
from the actual analysis, and the ﬁnal sample size was 236 (116 Asian Americans and
120 Caucasian Americans). A t-test was conducted to check the product type
manipulation, and skin foundation (M=5.81) was perceived more racially-charged
compared with an iron (M=1.40, t=33.43, df=236, p<.01). Another set of t-tests on
product type was performed on each ethnic group, and both Asian Americans (skin
foundation M=5.69, an iron M=1.25, t=28.46, df=115, p<.01) and Caucasian Americans
(skin foundation M=5.91, an iron M=1.25, t=28.46, df—‘l l9, p<.01) found the products to
be signiﬁcantly different.

In addition to the manipulation check on product type, participants’ level of
product involvement for the selected products was also tested using a t-test, and the result
indicated that participants were more involved with skin foundation (M=4.08) than an
iron (M=3.54, t=2.7l, df=231, p<.01). Speciﬁcally, Caucasian American participants
were more involved with skin foundation (M=4.33) than an iron (M=3.69, t=2.24, d%117,
p<.01) while there was no statistical difference between skin foundation and an iron in
terms of the level of product involvement for Asian American participants. Therefore,
product involvement was used as a covariate in the actual hypothesis testing.

Another manipulation check was performed on perceived model attractiveness.

Both Asian and Caucasian models were perceived similarly attractive (Asian M=5.29,

63

Caucasian M=5.08, t=1.74, df=232, p>.05). Therefore, it is concluded that the
manipulation checks were successful.
Overview of the Main Findings

To provide an overview of the main ﬁndings, two sets of ANCOVA tests were
performed. The ﬁrst set includes three ANCOVA tests using model’s race and product
type as the independent variables, and ethnic identiﬁcation, ad schema incongruity, and
perceived similarity as the covariates, and perceived model credibility as the dependent
variable. First, an ANCOVA test was performed on all samples (Table 4a). Two additional
ANCOVA tests were administrated on Caucasian (Table 4b) and Asian (Table 40) samples
separately.

Table 4a: ANCOVA Results on Perceived Model Credibility (All Sample)

 

 

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic 1, 221 3.01 .08
Identification
Perceived 1, 221 23.78 .01 *
Similarity
Ad Schema 1, 221 .12 .73
Incongruity
Model’s Race 1, 221 1.91 .17
Product Type 1, 221 .04 .84
Model’s Race x 1, 221 1.23 .27
Product Type

 

* Signiﬁcance at p<.01 level

64

Table 4b: ANCOVA Results on Perceived Model Credibility (Caucasian Sample)

 

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic l, 109 2.25 .14
Identification
Perceived 1, 109 28.27 .01*
Similarity
Ad Schema 1, 109 .14 .71
Incongruity
Model’s Race 1, 109 6.63 .01*
Product Type 1, 109 .97 .33
Model’s Race x l, 109 6.99 .01*
Product Type

 

Table 4c: ANCOVA Results on Perceived Model Credibility (Asian Sample)

* Signiﬁcance at p<.01 level

 

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic 1, 104 .05 .82
Identification
Perceived 1, 104 51.12 .01*
Similarity
Ad Schema 1, 104 1.66 .20
Incongruity
Model’s Race 1, 104 .41 .52
Product Type 1, 104 .10 .75
Model’s Race x 1, 104 .30 .58
Product Type

 

 

 

* Signiﬁcance at p<.01 level

As these results indicate, perceived similarity appears to have a strong effect on
perceived model credibility for all samples. When both Asian and Caucasian respondents
perceived the model in the ad similar to themselves, the rating for model credibility

tended to be higher. For the Caucasian sample, model’s race also has a main effect on

65

perceived model credibility. Generally, Caucasian respondents regarded the Asian model

more credible than the Caucasian model. More speciﬁc ﬁndings will be discussed in the

later result section of hypothesis testing.

Similarly, another set of ANCOVA tests were performed using attitudes toward

ad as the dependent variable, model’s race and product type as the independent variables,

and ethnic identiﬁcation, ad schema incongruity, and perceived similarity as the

covariates. First, an ANCOVA test was performed on all samples (Table 5a). Then two

additional ANCOVA tests were administrated on Caucasian (Table 5b) and Asian (Table

5c) samples separately.

Table 5a: ANCOVA Results on Attitudes toward Ad (A11 Sample)

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic 1, 220 .39 .53
Identification
Perceived 1, 220 26.91 .01*
Similarity
Ad Schema 1, 220 .35 .55
Incogruity
Model’s Race 1, 220 .09 .771
Product Type 1, 220 10.47 .01*
Model’s Race x 1, 220 .59 .44
Product Type

 

 

 

66

* Signiﬁcance at p<.01 level

Table 5b: ANCOVA Results on Attitudes toward Ad (Caucasian Sample)

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic 1, 109 .65 .43
Identification
Perceived l, 109 9.75 .01*
Similarity
Ad Schema 1, 109 .001 .99
Incongruity
Model’s Race 1, 109 .001 .98
Product Type 1, 109 2.16 .14
Model’s Race x l, 109 .81 .37
Product Type

 

 

 

* Signiﬁcance at p<.01 level

Table 5c: ANCOVA Results on Attitudes toward Ad (Asian Sample)

 

 

 

 

 

 

 

 

 

 

Source df F Sig.
Ethnic l, 104 .46 .50
Identiﬁcation
Perceived 1, 104 14.43 .01*
Similarity
Ad Schema 1, 104 4.04 .05"
Incongruity
Model’s Race 1, 104 .31 .58
Product Type 1, 104 10.17 .01*
Model’s Race x 1, 104 .08 .78
Product Type

 

 

 

*Signiﬁcance at p<.01 level

** Signiﬁcance at p<.01 level

For attitudes toward ad, perceived similarity has a main effect in all of the three
samples. Interestingly, for the Asian American sample, ad schema incongruity and
product type seems to have an effect on their attitudes toward ad. In the following section,

each hypothesis that was stated earlier will be tested. However, several of the ﬁndings

67

will not be signiﬁcant since ethnic identiﬁcation is not signiﬁcant in these ANCOVA test
results.
Hypothesis Testing

H1 tested the effect of model’s race on advertising evaluations. Two sets of
ANCOVA were performed, with model’s race as the independent variable, attitude toward
the ad as the dependent variable, and the level of product involvement as a covariate. The
results indicated that for attitude toward the ad, there was a statistically signiﬁcant
difference in terms of the mean scores of the ad evaluation between the Asian model
(M=4.13) and the Caucasian model (M=3.63) (F=6.42, df=1, 227, p<.05). In other words,
Asian respondents liked the ads with the Asian model more than the ads with the
Caucasian model. On the other hand, there was no signiﬁcant difference between the
rating for the ad regardless of the model’s race on Caucasian American subjects
(Caucasian model M=4.11, Asian model M=4.17; F=.11, df=1, 227, p>.05). This ﬁnding
was consistent with the hypothesis and the ﬁndings from the previous study by Whittler
and DiMeo (1991). Therefore, H1 was supported by the data. Asian Americans liked the
ads better with the Asian model while Caucasian Americans had no difference between
either model presented on attitudes toward the advertisements.

H2 examines the correlation between Asian Americans’ ethnic identiﬁcation and
their attitude toward advertisements with Asian American models. A bivariate correlation
was conducted to test the relationship between ethnic identiﬁcation and advertising

evaluations only on Asian American respondents who saw the ad with the Asian model.

68

Table 6: Results of Descriptive Statistics and Correlation for H2 (n=62)

 

 

 

 

 

 

Variables Ethnic Identification Attitude toward Ad
Mean 4.07 4.15
Standard Deviation 1.11 1.10
Pearson’s Correlation -.017

Significance (2-tailed) .90

 

 

 

 

The result was not statistically signiﬁcant. Therefore, H2 was not supported by
the data. There was no positive correlation between Asian Americans’ ethnic
identiﬁcation and their advertising evaluations of the ad with the Asian model.

H3 tested the correlation between perceived model similarity and perceived
model credibility. A bivariate correlation was performed to test the relationship on all of
the cases.

Table 7: Results of Descriptive Statistics and Correlation for H3 (n=232)

 

 

 

 

 

 

Variables Perceived Similarity Perceived Credibility
Mean 3.24 4.64
Standard Deviation 1.29 .73
Pearson’s Correlation .50

Signiﬁcance (2-tailed) .001*

 

 

 

 

* Signiﬁcance at p<.01 level

The result indicates a very strong positive correlation between the two variables

with the statistical signiﬁcance level of p<.01. Additionally, a regression was conducted
to test the causality between perceived similarity as the independent variable and

credibility as the dependent variable. The ﬁt of the regression model (R2) was .25,

69

meaning that 25 percent of the variance was explained by the regression equation, and the
ANOVA result showed the overall signiﬁcant causal relationship between perceived
similarity and credibility (F=77.0, df=l, 229, p<.01). Also, the beta weight value is
positive (B=.28, Table 8).

Table 8: Result of Regression for H3

 

 

 

 

Model 8 Standard t p
Error

(Constant) 3.77 .11 32.78 .001*

Perceived .28 .03 8.77 .001*

Model

Similarity

 

 

 

 

 

 

* Signiﬁcance at p<.01 level
In addition, correlations were administrated using the three components of
source credibility and perceived similarity (Table 9).

Table 9: Correlations between Similarity and Credibility Components (113a)

 

 

 

 

 

Attractiveness Trustworthiness Expertise
Mean 5.18 4.48 4.24
Standard Deviation .91 .90 .99
Pearson’s .42* .40* .36*
Correlation

 

 

 

 

 

* Signiﬁcance at p<.01 level (2-tailed)

The results indicate that all of the credibility components highly correlated with
perceived similarity with the signiﬁcance level of .01 or smaller. Therefore, H3 supported
by the data. Perceived model similarity is positively correlated with perceived model

credibility.

70

H4 tested the relationship between perceived model similarity and advertising
evaluations. According to the result from the correlation, there is a positive correlation
between perceived model similarity and advertising evaluations (r=.32, p<.01).

Table 10: Results of Descriptive Statistics and Correlation for H4 (n=232)

 

 

 

 

 

 

Variables Perceived Similarity Attitude toward Ad
Mean 3.24 4.03
Standard Deviation 1.29 1.04
Pearson’s Correlation .32

Significance (2-tailed) .001*

 

 

 

 

* Signiﬁcance at p<.01 level

In addition to the correlation analysis, another regression was performed with
perceived model credibility as the independent variable and attitude toward ad as the
dependent variable. Eleven percent of the variance was explained by the regression
model. The ANOVA result shows the overall signiﬁcance (F=26.71, df=l, 228, p<.01).
The beta weight of the independent variable was also positive and signiﬁcant (Table 11).
Therefore, H4 was supported. There was a positive relationship between perceived model

similarity and advertising evaluations.

71

Table 11: Result of Regression for H4

 

 

 

 

Model B Standard t p
Error

(Constant) 3.18 .18 18.10 .001*

Perceived .26 .05 5.17 .001*

Model

Similarity

 

 

 

 

 

 

* Signiﬁcance at p<.01 level
H4a tested the relationship between ethnic identiﬁcation and perceived similarity.
A bivariate correlation was conducted to test the hypothesis.

Table 12: Results of Descriptive Statistics and Correlation for H4b (n=235)

 

 

 

 

 

 

Variables Ethnic Identiﬁcation Perceived Similarity
Mean 4.39 3.25
Standard Deviation 1.04 1.29
Pearson’s Correlation .07

Signiﬁcance (2-tailed) .32

 

 

 

 

According to the result presented in Table 12, there was no signiﬁcant
correlation between ethnic identiﬁcation and perceived model similarity. Therefore, H4a
was not supported by the data. There was not a positive correlation between ethnic
identiﬁcation and perceived similarity.

H5 hypothesized the relationship between Asian Americans’ ethnic identiﬁcation
and ad schema incongruity. To test H5, a bivariate correlation was performed on ad
schema congruency Asian Americans’ ethnic identiﬁcation and ad schema incongruity.
The result was not statistically signiﬁcant (r=.06, p>.05). Hence, Asian Americans’ ethnic

identiﬁcation was not positively correlated with ad schema incongruity.

72

Table 13: Results of Descriptive Statistics and Correlation for H5 (n=ll4)

 

 

 

 

 

 

 

 

Variables Ad Schema Congruity Ethnic Identiﬁcation
Mean 4.95 4.15
Standard Deviation 1.27 1.11
Pearson’s Correlation .058

Signiﬁcance (2-tailed) .57

 

H6 predicts a positive correlation between ad schema incongruity and Asian
Americans’ attitude toward the ad with the Asian model. Ad schema incongruity was
measured with a four-item scale where the higher score the more likely ad schema
incongruity existed. In this case, ad schema incongruity is believed to exist in Asian
Americans’ mind when they do not expect to see Asian models in advertisements in
general. Some degree of ad schema incongruity is said to trigger a pleasant surprise when

ethnic minority members actually see models from their ethnicity in advertisements,

leading to more favorable advertising evaluations (Dimofte et al. 2004).

Table 14: Results of Descriptive Statistics and Correlation for H6 (n=62)

 

 

 

 

 

 

 

 

Variables Ad Schema Congruency Attitude toward Ad
Mean 4.90 4.15
Standard Deviation 1.33 1.09
Pearson’s Correlation .14

Signiﬁcance (2-tailed) .27

 

As Table 14 suggested, the correlation coefﬁcient was not statistically signiﬁcant
(r=.14, p=.27). Therefore, H6 was not supported. There was not a positive correlation

between ad schema incongruity and Asian Americans’ attitude toward the ad with the

73

 

 

Asian model.
H7 tests the simple effect of product type only on Asian American respondents.
An ANCOVA was executed with model’s race and product type as the independent

variables, attitudes toward ad as the dependent variable, and product involvement as a

covariate.

Table 15: Descriptive Statistics for Attitudes toward Ad

 

 

 

 

 

Asian American Respondents Caucasran Amencan
Respondents
Ad Asian Asian Caucasia Caucasia Asian Asian Caucasia Caucas
Type model x mode n model n model model x model 11 model x ian
skin 1 x an x skin x an iron skin x an skin model
Foundati iron foundatio foundati iron foundatio x an
on n on n iron
N 32 29 26 25 25 29 33 3 1
Mean 4.38 3.86 3.92 3.33 4.17 4.17 4.30 3.91
Std. 1.08 1.07 .80 .96 .97 1.08 .84 1.20
Deviat-
ion

 

 

 

 

 

 

 

 

 

 

 

The result suggested that Asian American subjects evaluated the skin foundation
ad with the Asian model more favorably (M=4.38) compared with the skin foundation ad
with the Caucasian model (M=3.92), and the mean difference between the two conditions
was statistically signiﬁcant (F=6.58, df=1, 225, p<.05). The control group of Caucasian
American participants was also examined using the same procedure (H7a). The result of
the ANCOVA indicated that there was no statistical difference between the mean scores
of Caucasian advertising evaluations regardless of the model’s race and product type
featured in the ads (F=l .14, df=1, 225, p>.05). This ﬁnding was expected as the previous

literature suggested that Caucasian Americans tended not to show the difference in

74

advertising responses regardless of the model’s race featured in the advertisements
(Whittler and DiMeo 1991). Therefore, H7 was supported by the data. While Asian
American consumers found the ad with the Asian model more favorable than Caucasian
models when the models were featured in the ad for a racially-charged product; however,
Caucasian American consumers did not show preference for the racially-charged product
ad with racially-congruent models.

H8 examines the effect of model’s race on perceived model credibility. Two sets
of ANCOVA were conducted on each ethnic group with the level of product involvement

as a covariate.

Table 16: Descriptive Statistics for Perceived Model Credibility for H8

 

Asian American Participants

 

Asian Model (n) Caucasian Model (11)

 

Constructs Cred Attract Trust Expert Cred Attract Trust Expert

Mean 4.73 5.22 4.66 4.30 4.38 4.87 4.05 4.18
(62) (62) (62) (62) (49) (51) (49) (50)

.93 .97 1.09 1.17 .62 .97 1.01 .67

 

 

Std.
Deviation

 

 

 

 

 

 

 

 

 

Caucasian American Participants

 

Asian Model (n) Caucasian Model (n)

 

Constructs Cred Attract Trust Expert Cred Attract Trust Expert

Mean 4.87 5.36 4.90 4.35 4.53 5.27 4.24 4.11
(55) (55) (55) (55) (64) (63) (64) (64)

.67 .87 .95 .91 .55 .82 .57 .81

 

 

Std.
Deviation

 

 

 

 

 

 

 

 

 

 

 

75

For Asian American participants, there was a statistical difference between mean
scores of perceived model credibility between Asian and Caucasian models (Asian
M=4.73, Caucasian M=4,38, F=4.46, df=1, 224, p<.05), and Asian American subjects
perceived the Asian model in the ad more credible than the Caucasian model.
Interestingly, for Caucasian American participants, a similar ﬁnding was obtained.
Caucasian American participants also perceived the Asian model as more credible
(M=4.87) than the Caucasian model in the ad (M=4.53, F=9.57, df=1, 224, p<.01).
Therefore, while the hypothesis was supported by the data for the Asian American
participants, it was not supported by the data for the Caucasian American participants.
Asian Americans found the Asian model in the ads more credible than the Caucasian
model, and so too did the Caucasian respondents.

H8a-c test the effect of model’s race on the three factors that form the credibility
construct for Asian Americans. Three sets of ANCOVA were administrated with model’s
race as the independent variable, attractiveness, trustworthiness, and expertise, as the

dependent variables, controlling for the level of product involvement.

76

Table 17: ANCOVA Results for H8a-c

 

 

 

 

 

 

Mean Square df F Sig
Credibility 2.75 1, 108 4.46 .037*
(Product (4.10) (6.67) (.01 1)"
InvolvementL
Attractiveness 2.96 1, 110 3.18 .07
(Product (1.72) (1.85) (.18)
Involvement)
Expertise .19 1, 108 .16 .69
(Product (5.06) (4.27) (.041)*
Involvement)
Trustworthiness 8.99 1, 109 10.99 .001"
(Product (6.42) (7.87) (.006)**
Involvement)

 

 

 

 

 

 

*Signiﬁcance at p<.05 level

** Signiﬁcance at p<.0 1 level

Regarding H8a-b, Asian American participants perceived the Asian model
neither more attractive (F=3.18, djEl , 229, p>.05) nor having more of an expertise (F =.16,
djEl , 228, p>.05) than the Caucasian model in the ads. However, the result suggested that
the Asian American participants did perceive the Asian model in the ad more trustworthy
(M=4.66) than the Caucasian model (M=4.05, F=10.99, df=1, 229, p<.01).

Additionally, H8d hypothesizes that the model’s race will mediate the
relationship between Asian Americans’ ethnic identiﬁcation and perceived model
credibility. To test this mediation effect, the method proposed by Baron and Kenny
(1986) was used. According to Baron and Kenny (1986), in order to claim the mediation
effect of a variable, three conditions must be met: 1) the independent variable must affect
the mediating variable; 2) the independent variable must affect the dependent variable;
and 3) the mediating variable must affect the dependent variable. Also, the addition of the

mediator to the relationship between the independent and dependent variables should

77

reduce the effect of the independent variable (Edwards and La Ferle 2003).

Since model’s race is a dichotomous categorical variable, it was dummy-coded
(1=Asian model, 0=Caucasian model) to be entered into the regression equations. For the
second regression, a logistic regression method was used since the dependent variable
(model’s race) is categorical.

Table 18: Model’s Race as a Mediator (H8d)

 

 

 

 

 

 

 

 

 

 

 

Independent Dependent B R’ t Sig.
Variable(s) Variable

Ethnic Credibility .072 .009 .65 .32
Identiﬁcation

Ethnic Model’s race -.13 .005 .59 .877
Identiﬁcation

Ethnic Credibility .08 .05 1.17 .24
Identiﬁcation .34 2. 1 2 .03
Model’s race

 

The results indicated that ethnic identiﬁcation did not have a signiﬁcant effect on
the mediator, model’s race (B=—.13, p>.05), while model’s race presented itself as a
signiﬁcant predictor of perceived model credibility (B=.33, p<.05). The third regression
equation which tested both ethnic identiﬁcation and model’s race on perceived credibility,
R2 increased from .009 to .05, but only model’s race remained as the signiﬁcant
predictors of perceived model credibility. Because ethnic identiﬁcation did not predict the
mediator, it is concluded that model’s race did not serve as a mediator between ethnic
identiﬁcation and perceived credibility. Therefore, H8 and H8c were supported. Asian
American consumers found Asian models in an ad more credible overall, and it appears

that trustworthiness was driving the ﬁnding.

78

H9 tests the mediation effect of perceived similarity between Asian Americans’
ethnic identiﬁcation and perceived credibility.

Figure 2: Similarity as a Mediator between Ethnic Identiﬁcation and Credibility

 

Slmllarlty

 

 

 

 

Ethnlc I Percelved
Identlflcatlon I '| Model Credlblllty

 

Three regression analyses were conducted with ethnic identiﬁcation as the
independent variable and perceived model credibility as the dependent variable, using the

data from Asian American participants.
Table 19: Similarity as a Mediator of Ethnic Identiﬁcation and Credibility

 

 

 

 

Independent Dependent f1 R’ t Sig.
Variable(s) Variable

Ethnic Credibility .071 .009 1.01 .32
Identiﬁcation

Ethnic Similarity .13 .01 1.05 .30
Identiﬁcation

Ethnic Credibility .03 .28 .42 .67
Identiﬁcation .31 6.34 .001*
Similarity

 

 

 

 

 

 

 

 

*Signiﬁcance at p<.01 level

The results indicated that ethnic identiﬁcation did not have a signiﬁcant effect on
the mediator, similarity (B=.13, p>.05). Likewise, ethnic identiﬁcation did not have a
signiﬁcant effect on perceived model credibility, either (B=.07, p>.05). The third
regression equation which tested both ethnic identiﬁcation and similarity on perceived

credibility, R2 increased from .04 to .28, and only similarity remained as the signiﬁcant

79

predictors of perceived model credibility. Because ethnic identiﬁcation failed to
signiﬁcantly predict both the mediator and the dependent variable, it is concluded that
similarity did not serve as a mediator between ethnic identiﬁcation and perceived
credibility. Therefore, H9 was not supported. Similarity did not mediate the relationship
between Asian Americans’ ethnic identiﬁcation and perceived model credibility.

H 10 predicts the mediating effect of perceived similarity in the relationship
between Asian Americans’ ethnic identiﬁcation and attitudes toward ad. To test the
mediation effect, three regression analyses were conducted.

Figure 3: Similarity as a Mediator between Ethnic Identiﬁcation and Aad

 

Slmllarlty

 

 

 

 

Ethnlc _ Attltude toward
Identlflcatlon Ad

 

 

 

 

 

 

The results of three regression analyses suggest the similar pattern seen in testing
for H9. The ﬁrst regression equation tests the effect of ethnic identiﬁcation on advertising
evaluations, and ethnic identiﬁcation does not appear to be a signiﬁcant indicator of
attitudes toward ad (B=-.02, t=-.21, p>.05). The second equation does not suggest
signiﬁcance of ethnic identiﬁcation on the mediator, similarity, either. Lastly, the
mediator is added to the ﬁrst regression equation. The results show that while R2 is

increased from .001 in the ﬁrst equation to .14 in the third regression equation, and

80

similarity is the signiﬁcant predictor of attitudes toward ad (B=.28, t=4.22, p<.01). Based
on these analyses it can be stated that because the independent variables of ethnic
identiﬁcation failed to predict the mediator and attitudes toward ad, there was no
mediation effect of perceived similarity. Therefore, H10 was not supported. Similarity did

not mediate the relationship between Asian Americans’ ethnic identiﬁcation and attitudes

 

 

 

 

toward ad.
Table 20: Tests of Similarity as a Mediator (1110)

Independent Dependent B R’ t Sig.
Variable(s) Variable
Ethnic Aad -.02 .001 -.21 .83
Identiﬁcation
Ethnic Similarity .13 .01 1.05 .30
Identiﬁcation
Ethnic Aad -.06 .14 -.67 .51
Identiﬁcation .28 4.22 .001 "'
Similarity

 

 

 

 

 

 

 

 

*Signiﬁcance at p<.01 level

H11 predicts positive correlations between Asian Americans’ perceived model
credibility and their attitudes toward a) advertisements; b) advertised brands; and c)
purchase intention.

Table 21: Correlations between Perceived Model Credibility and Aad, Ab, and PI

 

 

 

 

 

 

 

 

 

Attitude toward Ad Attitude toward Purchase
Brand Intention
N 1 10 111 112
Pearson’s '41,, .42... .47:
Correlation
Signiﬁc‘m .001 .001 .001
(2-tailed)

 

*Signiﬁcance at p<.001

For all of the variables, perceived model credibility appears to have

81

signiﬁcantly positive correlations at the signiﬁcance level of p<.001 (Table 21). Hence,
the data support H11. The degree of perceived model credibility is positively correlated
with Asian American participants’ attitudes toward a) advertisements; b) advertised
brand; and c) purchase intentions.

Lastly, H12 tests the relationships between attitudes toward ad, attitudes

toward brand, and purchase intentions. The relationship can be illustrated in the

 

 

 

following.
Figure 4: Relationships between Aad, Ab, and PI
57* .47*
Attltude toward Attltude toward Purchase
Ad ‘ Brand _’ lntentlon

 

 

 

 

 

 

 

*Signiﬁcance at p<.01 level

A path analysis using all participants was performed. Attitudes toward ad was a
signiﬁcant predictor of attitudes toward brand (r=.57, p<.01), and attitudes toward brand
had a signiﬁcant path coeﬁicient (r=.47, p<.01). The ﬁt of the model was assessed by
using the chi-square goodness-of-ﬁt test. The chi-square value was 41.56 (df=25, p<.05).
As chi-square values tend to be signiﬁcant as the sample size becomes larger, results
from additional goodness-of-ﬁt tests should be taken into consideration when assessing a
model’s ﬁt (Byme 2001). The additional test results of NF I and CFI both presented the

values greater than .95 (Table 22). These values range from zero to 1.00, and a value

82

greater than .90 suggests a well-ﬁtting model (Byme 2001).

Table 22: Goodness-of-ﬁt Test Results for H12

 

:6 df p x’Ip NFI CFI
41.56 25 .02 1.66 .97 .99

 

 

 

 

 

 

 

 

 

Therefore, H12 is supported by the data. Attitudes toward the advertisement will
inﬂuence attitudes toward the brand, and this inﬂuences attitudes toward purchase
intentions.

Post Hoc Analyses of Cultural Antecedents of Perceived Model Credibility and Aad

To better understand what drives Asian Americans’ perceived model credibility
and resulting attitudes toward ad, two regression analyses were conducted. First, a
multiple regression was administrated with ethnic identiﬁcation, ad schema incongruity,
and similarity as independent variables, and perceived model credibility as a dependent
variable. The regression equation can be stated as Credibility = 3.13+.02X‘(ethnic
identiﬁcation) +.07X2(ad schema incongruity) + .32X3(similarity) +e. 29 percent of the
variance was explained by this model, and the equation appeared to have an overall

signiﬁcant ﬁt (F=14.50, df=3, 107, p<.01).

83

Table 23: Post Hoc Analysis of Cultural Antecedents of Perceived Credibility

 

 

 

 

 

 

 

 

 

 

Model B Standard t p
Error

(Constant) 3.13 .40 7.91 00]“

Ethnic .02 .06 .32 .752

Identiﬁcation

Ad Schema .07 .05 1.28 .21

Incongruity

Similarity .32 .05 6.49 .001*

 

As the results presented in Table 23, the only signiﬁcant predictor of perceived
model credibility was similarity, and both ethnic identiﬁcation and ad schema incongruity
had very small effects on perceived model credibility for Asian Americans.

Another regression was conducted using the same independent variables as the
ﬁrst regression, and attitudes toward ad as the dependent variable. The regression
equation is Attitude toward ad = 2.49.-.07X‘(ethnic identiﬁcation) +.15X2(ad schema

incongruity) + .30X3(similarity) +e. 17.3 percent of the variance was explained by this

*Signiﬁcance at p<.01 level

model, and the equation has an overall signiﬁcant ﬁt (F=7.48, df=3, 107, p<.01).

A close examination of the regression coefﬁcients reveal that in this model,

while ethnic identiﬁcation still remains as a non-signiﬁcant predictor of attitudes toward

ad, both ad schema incongruity and similarity appear to predict attitudes toward ad.

84

 

Table 24: Post Hoc Analysis of Cultural Antecedents of Aad

 

 

 

 

 

Model B Standard t p
Error

(Constant) 2.49 .55 4.56 .001*

Ethnic -.O7 .08 -.80 .43

Identification

Ad Schema .15 .07 2.00 .047"

Incongruity

Similarity .30 .07 4.52 .001*

 

 

 

 

 

 

 

*Signiﬁcance at p<.01 level
”Signiﬁcance at p<.05 level

The results from these two post hoc regression analyses suggest the powerful
inﬂuence of perceived similarity on both perceived model credibility and advertising
evaluations for Asian Americans, as well as the non-signiﬁcant inﬂuence of Asian

American ethnic identiﬁcation.

85

CHAPTER SIX
DISCUSSION

The current study attempted to investigate the inﬂuence of ethnic identiﬁcation
and cultural congruency with the model, product, and consumer, on perceived model
credibility and Asian American consumers’ advertising evaluations. By doing so, the
current study sought to contribute to better learning the underlying mechanisms of Asian
American consumers’ cognitive processing of advertising stimuli with cultural/ethnic
cues.

While Asian Americans begin to be recognized as an important and growing
consumer group in the U.S., very few studies concerning this consumer group have been
conducted thus far. The data collected for this study supported some of the hypotheses,
and it is hoped that these ﬁndings will help both researchers and practitioners to achieve a
better understanding of this ethnic consumer group as well as to create effective
communication campaigns targeted toward the group. This section focuses on several
notable and important ﬁndings that characterize the signiﬁcance of the study.

Model Congruency Eﬂect on Advertising Evaluations of Asian Americans

One of the major ﬁndings in this study is the effect of model congruency on
perceived model credibility and advertising evaluations of Asian American consumers. In
general, Asian American participants in this study tended to perceive the Asian model as
more credible than the Caucasian model (H8), and preferred the ads featuring the
racially-congruent model (H1), particularly when the product type was racially-charged

(H7). This ﬁnding is similar to the previous ﬁndings concerning other ethnic minority

86

groups such as African Americans (Green 1997). Also, the phenomenon seemed to be
clearer particularly when the participants perceived similarity between the model and
themselves. In other words, the more Asian American participants perceived the
similarity between themselves and the models presented in the ad, the more likely that
they saw the models as credible (H3 and H3a). In addition, the degree of perceived model
credibility positively correlated with advertising favorability (H4). The results from the
post hoc analyses on cultural antecedents of perceived credibility and advertising
responses also revealed that among three major cultural antecedents examined in the
current study, namely, ethnic identiﬁcation, ad schema incongruity, and perceived
similarity, similarity was a signiﬁcantly strong driving factor of these dependent
variables.

As these results suggest, the congruency effect between the advertising stimuli
of the model’s race, and consumers’ race and/or ethnicity greatly inﬂuenced advertising
outcomes of Asian American consumers. The ﬁndings from the current study correspond
with the previous ﬁndings on other ethnic minority consumer groups and their advertising
evaluations (Appaih 2000; Green 1997; Dontshu and Cherian 1992; Whittler and Spira
2002). In this regard, although replications should be conducted to conﬁrm this
phenomenon, it is possible to argue that among the cultural variables examined in this
study, congruency plays a major role in predicting advertising evaluations of Asian
American consumers.

The in-group bias theory, originally proposed by Brewer (1979) may be helpful

to understand this phenomenon. The theory argues that individuals tend to compare and

87

categorize others as well as themselves either as in-group or out-group based on criteria
such as race, age, and lifestyle, and they tend to evaluate in-group people who are similar
to themselves more favorably than out-group people who are dissimilar (Brewer, 1979;
Wilder, 1981). As a result, Asian American respondents in this study may have perceived
the Asian model as more credible due to the perceived similarity, and the following
advertising evaluations tended to be more positive compared with the ads featuring the
Caucasian model.

In the current study, the scale used to measure perceived model credibility
includes items on physical appearances, values, cultural backgrounds and lifestyle. For
the pair of the models selected in the main study, both Asian and Caucasian models
appear to be in their 205, and their dress style also appears to be similar to that of the
respondents. In addition to the perceived similarity due to physical appearances (such as
race), the participants may have perceived the Asian model to be more similar than the
Caucasian model because of the lifestyle or values portrayed in the ad. Combined with
their in-group bias, the congruency effect that resulted from perceived similarity may
have achieved a higher degree of perceived model credibility and favorability of ads
featuring the Asian model. Needless to say, more studies are called for to learn more
about the relationship between in-group bias and model congruency associated with
Asian American consumers.

Absence of Model Congruency Eﬂects on Advertising Outcomes of Caucasian Americans

This study also examined the advertising evaluations of female Caucasian

American consumers as a control and comparison group. Interestingly, while the

88

congruency effect between advertising cues and Asian Americans’ consumer
characteristics apparently has a strong impact on the resulting perceived model credibility
and advertising evaluations, model congruency does not appear to have an inﬂuence on
Caucasian Americans’ perceived model credibility and advertising outcomes (Hla and
H7a). This phenomenon was consistent with Caucasian American participants throughout
hypothesis testing.

One possible explanation to this trend can be related to the lack of ethnic
distinctiveness among Caucasian Americans in the American society. Previous
researchers have argued that ethnic or cultural cues in advertising tend to be more
effective for members of ethnic minority groups because the salience of their ethnicity is
enhanced due to the numerical minority status (Deshpande and Stayman 1994).
Regarding Caucasian American consumers in general, they are still a numerical majority
group in the U.S., and as Appaih (2000) argues, they may be less likely to be affected by
ethnic or cultural cues in advertising due to the lesser degree of ethnic salience compared
with other ethnic minority groups in the US.

Another interesting ﬁnding on Caucasian Americans’ responses is that they
found the Asian model more credible than the Caucasian model (H8). This may be due to
positive Asian American stereotypes held by some members of the ethnic majority group.
One of such positive stereotypes of Asian Americans is the image of Asian Americans as
the model minority (Oyserman and Sakamoto 1997; Abreu, Ramirez, Kim and Haddy
2003). Caucasian Americans in this study may have regarded the Asian model as a

D

representation of the model minority, thinking that Asian Americans are hardworking and

89

successful in academic or economic structures in the US. (Oyserman and Sakamoto
1997). Often, Asian Americans are portrayed as industrious and serious in advertisements
(Taylor and Lee 1994). Caucasian respondents in this study may have been exposed to
such portrayals of Asian Americans in general advertisements and therefore, they may
have regarded the Asian model more credible than the Caucasian model.

Moreover, it is possible to argue that Caucasian respondents in the current study
may have perceived the Asian model more similar to themselves than the Caucasian
model because of the portrayed lifestyle, clothing, make-up, or age of the model. The
model selection was based on the level of perceived attractiveness of both Asian and
Caucasian models, and these variables, such as perceived lifestyle and appearance, were
not tested in the model selection process. Therefore, it is suggested that future research
will pay more attention to this outcome to learn about the perceptions of the majority on
minority groups and how they are related to the advertising evaluations featuring ethnic
minority models.

Eﬁects of Cultural Antecedents of Perceived Model Credibility

While similarity, which can be considered an indication of congruency, was a
signiﬁcant indicator of perceived model credibility and resulting advertising evaluations
of Asian Americans in the current study, the other hypothesized cultural antecedents,
namely ethnic identiﬁcation and ad schema incongruity, were not signiﬁcant predictors of
perceived model credibility. The post hoc analysis also revealed that ad schema
incongruity was a signiﬁcant predictor of advertising outcomes, which is similar to the

previous study by Dimofte et al. (2004). However, ethnic identiﬁcation did not present

90

itself as a signiﬁcant indicator of advertising evaluations.

Several explanations are possible in these cases. In terms of ad schema
incongruity and its effect on advertising evaluations but not on perceived model
credibility (H6), the ﬁndings of the current study are plausible because all of the items
used to measure ad schema incongruity asked about their expectations of Asian American
models’ appearances in advertisements in general, and did not ask questions speciﬁcally
about the Asian model presented in the stimuli. Therefore, while this construct had a
signiﬁcant indication of advertising evaluations, it may have not had an impact on
perceived model credibility. Future studies should consider including items about
respondents’ expectation to see an Asian model in speciﬁc ads used in a study, in addition
to items on general expectations to see Asian models in advertisements.

Meanwhile, the non-signiﬁcant inﬂuence of Asian Americans’ ethnic
identiﬁcation on perceived model credibility (H4a) and advertising responses (H2) was
the opposite of the previous ﬁndings of the effect of ethnic identiﬁcation of other ethnic
minority groups (Green 1997; Appaih 2000; Desphande, Hoyer and Donthu 1986;
Donthu and Cherian 1992). This tendency may be related to the sample characteristics in
this study.

First, the Asian sample was all university students, and they were probably more
educated than the general Asian population, and likely to come from affluent families. In
other words, these respondents tend to ﬁt into the typical “model minority” stereotype
(Oysterman and Sakamoto 1997). Therefore, they may not necessarily feel a need to be

outward spoken about their ethnicity compared with other ethnic minority groups such as

91

African Americans and Hispanic Americans. In addition, because of their nature of
collectivism (Hofstede 1997) and conformity motivation (Ang et al. 2001), Asian
Americans in the current study may have been more leaning toward the mainstream
group, and may have discouraged to be outspoken about their ethnicity and stand out
from the mainstream group. Such a tendency may have inﬂuenced the lack of ethnic
identiﬁcation effect on perceived model credibility and advertising evaluations of Asian
Americans.

Moreover, the Asian American population in the US. is quite small compared
with African Americans or Hispanics whose ethnicity tends to be more salient in
consumer behavior and advertising evaluations. As the population of Asian Americans
increases, they may become more outspoken about their ethnicity, and it may inﬂuence
their advertising evaluations and consumptions. Therefore, more research in the ﬁrture is
necessary as the population status of Asian American is likely to change due to their
growth rate.

Another issue is related to Asian Americans’ immigration background. The
history of Asian immigration is shorter and the ﬂux of Asian immigration has taken place
more recently compared with African Americans. For example, in the current study, about
50 percent of the Asian American sample was non-US. born (although American now or
permanent resident), and 46 percent considered themselves as the ﬁrst generation in the
US. Compared with other ethnic minority groups, particularly African Americans who
are likely to consider themselves as beyond ﬁrst or second generations, Asian Americans

have a shorter history of being “American.” Hence, such a different immigration

92

background may have contributed to the non-signiﬁcant inﬂuence of Asian “Americans’”
ethnic identiﬁcation of perceived model credibility and resulting advertising evaluations.
Also, Agbayani-Siewert (2004) argues that Asian American groups are distinguished ﬁom
other groups by characteristics including the level of acculturation. Acculturation also
inﬂuences the transmission of beliefs and behaviors to other members of the cultural
population (Lehman et al. 2004). Combined with Asian Americans’ collectivistic nature,
the level of acculturation has a great potential to inﬂuence not only their own beliefs,
attitudes, and behaviors, but also those of the other members who share the same culture.
Therefore, acculturation can be proposed as a potential predictor of perceived model
credibility and advertising evaluations of Asian Americans.

Acculturation is deﬁned as “the process of learning and adopting cultural traits,
different from the ones with which the person was originally reared” (Ownbey and
Horridge 1997, p. 3). This concept is often discussed in the study of ethnicity in
advertising and marketing literature. Previous empirical studies suggest that the level of
acculturation inﬂuences individuals’ various behaviors and perceptions ranging from
consumer behavior (Wallendorf and Reilly 1983; Faber, O’Guinn and McCarty 1987;
Kara and Kara 1996; Ownbey and Horridge 1997), and media consumption (Lee and Tse
1994) to advertising evaluations (Lee 1993; Khairulla, Tucker and Tankersley 1996).
According to the statistics from the US. Census Bureau (2000), of 12 million Asian
Americans, approximately 8 million of them were foreign-bom, and apparently, the
proportion of the ﬁrst generation is extremely high among Asian Americans.

One of the indicators of the level of acculturation is the ability to speak the

93

English language (Lee 1993), and in fact, across different age groups, more than half of
Asian Americans speak other languages than English (U .8. Census Bureau 2000). If an
Asian American is not ﬂuent in English but ﬂuent in other Asian languages, he/she may
be considered less acculturated to the US. society. On the other hand, if an Asian
American can use only English, it is likely that he/she is beyond ﬁrst or second
generation, and more acculturated to the US. society. Therefore, the ability to use the
English language as well as other languages may serve as a good indicator of
acculturation. Based on these characteristics shared by Asian Americans in the sample, in
the future studies, other cultural variables, such as the degree of acculturation, should be
taken into consideration when investigating the cognitive processing of Asian American
consumers’ advertising responses in addition to ethnic identiﬁcation.

Another issue on ethnic identiﬁcation is related to the scale used in the current
study. For this study, both attitudinal and behavioral factors were combined to create a
general ethnic identiﬁcation scale. However, when the mean scores of these factors were
compared, the behavioral factor had a lower score (M=3.25) than the attitudinal factor
(M=5.40, t=18.03, df=114, p<.01). The behavioral items included questions such as “I
like ethnic movies or movies that portray my culture’s background,” and “I prefer
watching TV channels or shows that are relevant to my ethnic group.” However, there
may be an infrastructure issue because there is a limited number of Asian radio/TV/movie
programs in general, and therefore, the Asian American respondents may not have access
to these outlets, resulting in a lower score of the behavioral ethnic identiﬁcation factor. As

a result, there may have been a discrepancy between the attitudinal and behavioral factors

94

of the ethnic identiﬁcation, leading to different outcomes from those of other ethnic
minority groups.

Such a discrepancy between attitudes and behaviors may be an indication of an
identiﬁcation (attitude) or recognition of being Asian, but not a subsequent need to act or
behave differently to stand out, which has been discovered in previous research on
Aﬁican American or Hispanic consumers. Therefore, it is recommended that future
studies use the attitude/behavioral ethnic identiﬁcation scale on different ethnic minority
groups including Aﬁican and Asian Americans to test these different ideas.

Perceived Model Credibility and the Inﬂuence of Each Component

Perceived model credibility is important to understand Asian American
consumers’ advertising evaluation process. The data ﬁom this study showed a positive
correlation between perceived model credibility and attitudes toward the ad, brand, and
purchase intention (H11). In addition to the overall credibility construct, three
components of model credibility, namely, attractiveness, trustworthiness, and expertise,
were tested on the advertising outcome measures. Each component was hypothesized to
have an inﬂuence on attitudes toward the ad (H8a—c). In particular, the effect of product
type along with model’s race, model expertise was expected to have a positive inﬂuence
on Asian American consumers’ attitudes toward the ad (H8b). Despite this prediction,
model trustworthiness was the only signiﬁcant indicator of perceived model credibility
(H8c). While this ﬁnding was initially surprising, considering the characteristics of Asian
culture using Hofstede’s (1997) cultural dimensions, the ﬁnding certainly makes sense to

a greater extent.

95

Among Hofstede’s (1997) four cultural dimensions, the concept of uncertainty
avoidance explains the best about the salience of model trustworthiness among Asian
American respondents in this study. In general, Asian culture is considered to score high
on uncertainty avoidance, and individuals from this type of culture tend to have a lower
tolerance level for ambiguity (de Mooij 1998). In this study, the relationship between
perceived model similarity and credibility was found to be signiﬁcant. Therefore, it is
possible to argue that Asian Americans tend to ﬁnd Asian models in ads more trustworthy
than Caucasian models because of the perceived similarity and their tendency of high
uncertainty avoidance. As a result, they tended to rate ads with Asian models more
favorable than ads with Caucasian models. Combining with the effect of model
congruency and induced model trustworthiness, the use of Asian models (endorsers) may
have resulted in a higher favorability of advertising campaigns associated with the
endorsers for Asian American consumers. The results from the current study suggest that
perceived similarity and model (source) trustworthiness are the two most important
considerations for the assessment of Asian American advertising evaluation process. In
this regard, it is believed that the current study has made a signiﬁcant contribution for
making the ﬁrst step to clarify the mechanism of Asian American consumers’ advertising

responses.

96

CHAPTER SEVEN

SUGGESTIONS FOR FUTURE RESEARCH AND CONCLUSION

For effective communication to consumers, it is essential to recognize the
importance of source credibility in advertising. In this regard, the area of research on
endorsers’ credibility is likely to remain popular both in the industry and academia. While
many studies have explored this area, it appears that cultural antecedents of source
credibility deserve more attention as the American society is becoming increasingly
multicultural. In this sense, it is believed that the current study has played an important
role in adding the cultural dimension to the existing source credibility research. However,
there are several issues that need to be addressed as considerations for future research.

Limitations of the Current Study

First, similar to the previous study on Asian American consumers’ ethnic
identiﬁcation on advertising evaluations (Morimoto and La Ferle 2005), due to the
product selected (skin foundation as the most racially—charged product among all of the
products provided in the pretest), the data from only female subjects were used in the
current study. To increase generalizability of the ﬁndings, it is suggested that future
studies explore other racially-charged product categories that are applicable and relevant
to both males and females.

As Morimoto and La Ferle (2005) stated, the current study also treated Asian
American and Caucasian American consumer groups as homogeneous. Particularly for
Asian American consumers, this practice can be considered appropriate for the time being,

since the proportion of Asian American population in the US. is still small; however as

97

the US. society and the Asian communities become more diverse, it is essential to start
treating “Asian Americans” as the heterogeneous group. Each segment that makes up the
Asian American group, such as Chinese and Japanese, has different and unique cultures
and characteristics, and thus, recognizing this uniqueness is likely to be the key for the
most successful advertising campaigns.

Along similar lines, the current study used self-identiﬁcation as the indicator of
respondents’ ethnic background, in addition to the initial selection administrated by the
registrar’s ofﬁce. This step was necessary as the ethnicity classiﬁcation used by the
registrar’s office treated Asian Indians also as Asian Americans, and approximately 10
percent of the responses from the registrar’s office mailing was from Asian Indian
females. Interestingly, the results of self-identiﬁcation in the main survey indicate that
these Asian Indians also considered themselves as Asian Americans. As this example
suggests, it is recommended to redeﬁne Asian Americans, and in the future, researchers
studying this diverse consumer group should also begin paying attention to Asian Indians
as a subgroup of Asian Americans.

In terms of data collection, the current study used two methods: a mail survey
for female Asian American respondents and a regular on-site survey for female Caucasian
Americans. Ideally, the data collection method should be consistent; however, due to the
limited budget and time constraints, the data from Caucasian Americans were collected
by visiting undergraduate advertising courses. This may have resulted in a smaller
variance in the data from Caucasian American participants, leading to some

non-signiﬁcant ﬁndings from this group because these participants may share similar

98

characteristics as they choose to major in the same ﬁeld. Future studies should avoid such
inconsistency in data collection as much as possible.
Suggestions for Future Research

In order to expand the understanding of cultural antecedents of model credibility
and its relation with advertising responses, there are some issues related to cultural
variables that should be considered carefully. For example, in the current study, the only
indicator of congruency effect used in the hypothesis testing was perceived model
similarity. While this indicator appeared to be signiﬁcant in the process of Asian
American consumers’ advertising responses, future studies should also seek for other
congruency indicators, such as “the ability to identify with the model (Whittler and
DiMeo 1991),” and familiarity with the model, and should investigate how these
indicators serve as the predictors of the model congruency effect. Such an attempt will
provide an opportunity to deepen the knowledge of advertising responses and consumer
behaviors of Asian American consumers.

Also, it is advised that future studies examine the potential effects of temporary
ethnicity-related constructs such as ethnic self-awareness (Forehand et al. 2002) and
situational ethnicity (Stayman and Deshpandé 1989) on Asian American consumers’
responses to ethnic advertising. Situational ethnicity assumes that one’s ethnic identity is
likely to be inﬂuenced by the immediate environment (Stayman and Deshpandé 1989),
and ethnic self-awareness is likely to depend on many other factors including social
situations and cultural cues presented in advertisements (Forehand et al. 2002). Although

these temporary constructs have been discussed previously, there have not been

99

well-established measures for them. For example, the open-ended question developed by
McGuire et al. (1978) has been used as a measure of ethnic self-awareness. However, this
is a single-item measure with an item that asks respondents to freely write about who
they are. It is measured by the total numbers of ethnicity-related words is used as the
ethnic self-awareness score. Since it is a single item measure, a question about the scale’s
reliability still remains as a concern. Other scholars have used the scale of ethnic
identiﬁcation as an indicator of ethnic self-awareness. However, conceptually, these
temporary constructs have been distinguished from ethnic identiﬁcation which is often
regarded as an enduring construct. As these examples indicate, scholars have been
experiencing difficulties to construct a valid and reliable scale for these constructs. For
better understanding of the mechanisms of Asian Americans’ advertising responses, it is
essential to examine the impact of ethnicity with different approaches. To do so, more
efforts are necessary to establish reliable scales of these temporary ethnicity-related
constructs to better understand the effectiveness of cultural cues including model’s race
presented in advertisements.

Lastly, as indicated in the previous discussion section, there may be other
cultural antecedents of perceived model credibility that explain the process more clearly
than the strength of ethnic identiﬁcation, such as the degree of acculturation and in—group
bias. While the replication of the study is certainly preferable and rewarding to learn
more about the effect of ethnic identiﬁcation on perceived model credibility and
advertising outcomes of Asian Americans, future studies should also explore other

potential cultural antecedents to disclose the mechanism of the inﬂuence of cultural

100

variables on cognitive processing of Asian American consumers. Despite these
difﬁculties and limitations, the current study has certainly made a signiﬁcant contribution
to adding the knowledge on Asian American consumers, and it is hoped that this study
will help bring to attention of researchers and practitioners this long-marginalized ethnic
consumer group in the future.
Implications

Practical Implications

The ﬁndings from the current study clearly showed the signiﬁcant effect of
model congruency through perceived model similarity on perceived model credibility and
the advertising responses that followed. From the practical perspective, these results
conﬁrm for practitioners that when targeting Asian American consumers, like other ethnic
minority consumer groups, model congruency is a very important key for maximizing the
effectiveness of advertising campaigns, regardless of the product type advertised in the
campaign. Also, the strong relationship between perceived model credibility and
advertising responses including purchase intensions should be regarded as the key to the
successful practice targeting Asian American consumers. Speciﬁcally, among three
components of perceived model credibility, trustworthiness appears to be the most
inﬂuential factor on the advertising outcomes. Based on these ﬁndings, it is
recommended that when advertising to Asian Americans, marketers should select not
only racially-congruent but also endorsers or models that are perceived as trustworthy in

their advertising campaigns to appeal to this consumer group.

101

Academic Implications

From the academic viewpoint, in addition to the ﬁnding that conﬁrms the strong
inﬂuence of model congruency and perceived credibility, this study has raised several
important issues that help scholars to advance the knowledge of the cognitive processing
of Asian American consumers and their reactions to ads featuring cultural cues. Unlike
African Americans, the strength of ethnic identiﬁcation did not present itself as a
signiﬁcant predictor of perceived model credibility or advertising outcomes for Asian
Americans. This ﬁnding suggests that compared with other ethnic minority groups,
cultural antecedents of model credibility and the resulting advertising evaluation process
of Asian Americans may involve different cultural variables other than ethnic
identiﬁcation. As stated earlier, this different should be explored in mode detail and may
stem from Asian Americans’ immigration situation or their perceived status in society.
Potential cultural variables that could be further explored include the degree of
acculturation and in-group bias. Certainly, more studies are necessary to explore the
cognitive processing of Asian American advertising evaluations, and in a sense, it is
believed that the current study has contributed to make a path for future research in this
area.

Moreover, the current study has revealed that for Asian American consumers, the
factor of model trustworthiness was the most signiﬁcant indicator of perceived model
credibility, and the effect of product type in relation to perceived model expertise was not
a signiﬁcant indicator of model credibility. These results suggest that there may be other

potential underlying cultural factors that inﬂuence perceived model credibility and its

102

components. Hence, it can be said that this study has provided more food for thought to
advance our knowledge about source credibility and ethnic consumers. At the same time,
the current study has conﬁrmed previous ﬁndings on source credibility in advertising
research by replicating the signiﬁcant relationship between perceived model credibility
and advertising evaluations. Therefore, the efforts to discover the underlying mechanism
of Asian American consumers’ advertising processing made in this study have greatly
contributed to the knowledge of cross-cultural advertising and source credibility.
Conclusion

For Asian American consumers, model congruency, perceived model similarity,
and trustworthiness seem to be the inﬂuential factors on their cognitive processing of
advertisements. In particular, as a cultural antecedent of perceived model credibility,
perceived model similarity emerged as a distinctive predictor. Several potential
explanations to this phenomenon were discussed in the previous chapter, such as Asian
Americans’ cultural characteristics represented by collectivism and Confucianism that
hold high value in group memberships and harmony.

While more research is necessary to determine what aspects of perceived
similarity plays a role in the development of perceived model credibility and the process
of Asian Americans’ advertising evaluations, the results from the current study suggest
that in addition to physical appearances such as race, lifestyle may be a key indicator of
perceived similarity with models presented in advertisements. Particularly for younger
generations who tend to be more accustomed to mainstream American values and

lifestyle, perceived similarity beyond racial congruency may be the driving factor of

103

perceived model credibility and resulting advertising evaluations of Asian Americans.
Future studies can certainly examine the use of the same ethnic model in different
lifestyle images, and investigate whether it is race of the model or the lifestyle portrayed
that is more important for perceived model credibility and advertising evaluations of
Asian American consumers.

It appears that the growth in Asian immigrants in the US. will continue in this
century, and the ﬁrture of this market is promising for advertisers. Hence, more research
in both academia and the industry is essential in order to achieve better marketing
communication strategies targeting this ethnic minority group. It is believed that this
study has made a notable contribution to bring to light this understudied consumer group,
and also, it is hOped that more studies will follow for the advancement of knowledge on

Asian Americans and cross-cultural advertising.

104

 

APPENDIX

Sample Advertisements

 

105

 

106

 

107

 

108

REFERENCES

Aaker, Jennifer L., Anne M. Brumbaugh, and Sonya A Grier (2000), “Nontarget Markets
and Viewer Distinctiveness: The hnpact of Target Marketing on Advertising
Attitudes,” Journal of Consumer Psychology, 9 (3), 127-140.

Abreu, Jose' M., Estella Ramirez, Bryan S. K. Kim and Chris Haddy (2003), “Automatic
Activation of Yellow Peril Asian American Stereotypes: Effects on Social
Impression Formation,” Journal of Social Psychology, 143 (6), 691-706.

Agbayani-Siewert, Pauline (2004), “Assumptions of Asian American Similarity: The
Case of Filipino and Chinese American Students,” Social Work, 49 (1), 39-51.

Ang, Swee Hoon, Peng Sim Cheng, Elison A. C. Lim and Siok Kuan Tambyah (2001),
“Spot the Difference: Consumer Responses towards Counterfeits,” Journal of
Consumer Marketing, 18 (3),. 219-235.

Appaih, Osei (2001), “Black, White, Hispanic, and Asian American
Adolescents’ Responses to Culturally Embedded Ads,” Howard Journal of
Communications, 12, 29-48.

Appaih, Osei (2000), “Ethnic Identiﬁcation on Adolescents’ Evaluations of
Advertisements,” Journal of Advertising Research, 41 (5), 7-22.

Baker, Michael J. and Gilbert A. Churchill, Jr. (1977), “The Impact of Physically
Attractive Models on Advertising Evaluations,” Journal of Consumer Research,
14 (November), 538-555.

Baron, Rueben M. and David A. Kenny (1986), “The Moderator-Mediator Variable
Distinction in Social Psychological Research: Conceptual, Strategic, and
Statistical Considerations,” Journal of Personality and Social Psychology, 51
(6), 1173-1182.

Basil, Michael D. (1996), “Identiﬁcation as a Mediator of Celebrity Effects,” Journal of
Broadcasting & Electronic Media, 40 (4), 478-495.

Bimbaum, Michael H. and Steven E. Stegner (1979), “Source Credibility in Social
Judgment: Bias, Expertise, and the Judge’s Point of View,” Journal of
Personality and Social Psychology, 37 (1), 48-74.

Batra, Rageev and Michael Ray (1986), “Affective Responses Mediating Acceptance of
Advertising,” Journal of Consumer Research, 13 (2), 234-249.

109

Brewer, Marilyn B. (197 9), “In-Group Bias in the Minimal Intergroup Situation:
A Cognitive-Motivational Analysis,” Psychological Bulletin, 86 (2), 307-324.

Byme, Barbara M. (2001), Structural Equation Modeling with AMOS: Basic Concepts,
Applications, and Programming. Mahwah, NJ: Lawrence Erlbaurn Associates.

Kara, Ali and Natasha R. Kara (1996), “Ethnicity and Consumer Choice: A Study of
Hispanic Decision Processes across Different Acculturation Levels,” Journal of
Applied Business Research, 12 (2), 22-34.

Khairrullah, Durriya 2., Frances Gaither Tucker, and Clint B. Tankersley (1996),
“Acculturation and Immigrant Consumers’ Perceptions of Advertisements: A
Study Involving Asian-Indians,” International Journal of Commerce &
Management, 6 (3/4), 81-104.

Carpenter, Sandra (2000), “Effects of Cultural Tightness and Collectivism on
Self-Concept and Causal Attribution,” Cross-Cultural Research, 34 (1), 38-56.

Chawla, Sudhir K. and Rajan Nataraajan (1995), “Endorsing an Ethnic Food Product:
The Roles of Race and Gender,” Journal of Food Products Marketing, 2 (3),
3-16.

Choi, Sejung Marina and Nora J. Rifon (2002), “Antecedents and Consequences of Web
Advertising Credibility: A Study of Consumer Responses to Banner Ads,”
Journal of Interactive Advertising, 3 (1), http://jiadorgl (Visited: July 5, 2004).

Chung, Ed and Eileen Fischer (1999), “It’s Who You Know: Intracultural Differences in
Ethnic Product Consumption,” Journal of Consumer Marketing, 16 (5), 482-501.

Cohen, Judy (1992), “White Consumer Response to Asian Models in Advertising,”
Journal of Consumer Marketing, 9 (2), 17-27.

de Mooij, Marieke K. (1998), Global Marketing and Advertising: Understanding
Cultural Paradoxes, Thousand Oaks, CA: Sage Publications, Inc.

Deshpande, Rohit, Wayne D., Hoyer, and Naveen Donthu (1986), “The Intensity of
Ethnic Affiliation: A Study of the Sociology of Hispanic Consumption,” Journal
of Consumer Research, 13 (2), 214-220.

Deshpande, Rohit and Douglas M. Stayman (1994), “A Tale of Two Cities:

Distinctiveness Theory and Advertising Effectiveness,” Journal of Marketing
Research, 31 (1), 57-64.

110

Dimofte, Caludiu V., Mark R. Forehand and Rohit Deshpande (2004), “Ad Schema
Incongruity as Elicitor of Ethnic Self-Awareness and Differential Advertising
Response,” Journal of Advertising, 32 (4), 7-17.

Dontshu, Naveen and Joseph Cherian (1992), “Hispanic Coupon Usage: The Impact of
Strong and Weak Ethnic Identiﬁcation,” Psychology & Marketing, 9 (6),
501-510.

Edwards, Steven M. and Carrie La Ferle (2003), “Role-Taking: Enhancing the Online
Experience,” Journal of Current Issues and Research in Advertising, 25 (2),
45-56.

Faber, Ronald J ., Thomas C. O’Guinn and John A. McCarty (1987), “Ethnicity,
Acculturation, and the Importance of Product Attributes,” Psychology &
Marketing, 4, 121-134.

Feldman, Robert H. (1984), “The Inﬂuence of Communicator Characteristics on the
Nutrition Attitudes and Behavior of High School Students,” Journal of School
Health, 54(4), 149-151.

Forehand, Mark R., Rohit Deshpande, and Americus Reed II (2002), “Identify Salience
and the Inﬂuence of Differential Activation of the Social Self-Schema on
Advertising Response,” Journal of Applied Psychology, 87 (6), 1086-1099.

Forehand, Mark R. and Rohit Deshpande (2001), “What We See Makes Us Who We Are:
Priming Ethnic Self-Awareness and Advertising Response,” Journal of
Marketing Research, 38 (3), 336-348.

Freiden, Jon B. (1984), “Advertising Spoesperson Effects: An Examination of Endorser
Type and Gender on Two Audiences,” Journal of Advertising Research, 24
(October/November), 33-41.

Friedman, Hershey H. and Linda Friedman ( 1979), “Endorser Effectiveness by Product
Type,” Journal of Advertising Research, 19 (October/November), 63-71.

Gardyn, Rebecca and John Fetto (2003), “Race, Ethnicity, and the Way We Shop,”
American Demographics, 25 (1), 30-33.

Gergen, Kenneth J. (1971), The Concept of Self New York: Holt, Rinehard & Winston.
Gerritsen, Marine], Hubert Korzilius, Frank van Meurs, and Inge Gijsbers (2000),

“English in Dutch Commercials: Not Understood and Not Appreciated,” Journal
of Advertising Research, 40 (4), 17-31.

111

Goldsmith, Ronald E., Barbara A. Lafferty and Stephen J. Newell (2000), “The Impact of
Corporate Credibility and Celebrity Credibility on Consumer Reaction to
Advertisements and Brands,” Journal of Advertising, 26, (3), 43-54.

Gotlieb, Jerry B. and Dan Sarel (1991), “Comparative Advertising Effectiveness: The
Role of Involvement and Source Credibility,” Journal of Advertising, 20 (1),
38-45.

Green, Corliss, L. (1999), “Ethnic Evaluations of Advertising: Interaction Effects of
Strength of Ethnic Identiﬁcation, Media Placement, and Degree of Racial
Composition,” Journal of Advertising, 28 (1), 49-64.

Gregory, Gary D. and James M. Munch (1997), “Cultural Values in International
Advertising: An Examination of Familial Norms and Roles in Mexico,”
Psychology & Marketing, 14 (2), 99-119.

Hofstede, Geert (1997), Cultures and Organizations: Software of the Mind, New York:
McGraw-Hill International.

Hogg, Margaret K., Alastair J. Cox and Kathy Keeling (2000), “The Impact of
Self-Monitoring on Image Congruence and Product/Brand Evaluation,”
European Journal of Marketing, 34 (5/6), 641-665.

Homer, Pamela M. and Lynn R. Kahle (1990), “Source Expertise, Time of Source
Identiﬁcation, and Involvement,” Journal of Advertising, 19(1), 30-40.

Hong, Jae W. and George M. Zinkhan (1995), “Self-Concept and Advertising
Effectiveness: The Inﬂuence of Congruency, Conspicuousness, and Response
Mode,” Psychology & Marketing, 12 (1), 53-77.

Hovland, Carl 1. and Walter Weiss (1951), “The Inﬂuence of Source Credibility on
Communication Effectiveness,” Public Opinion Quarterly, 15 (4), 63 5-650.

Hovland, Carl 1., Irving L. Janis, and Harold H. Kelley (1953), Communication and
Persuasion. New Haven, CT: Yale University Press.

Hurh, Won Mon and Kwang Chung Kim (1989), “The ‘Success’ Image of Asian
Americans: Its Validity and its Practical and Theoretical Implications,” Ethnic and
Racial Studies, 12 (October), 512-538.

Jones, Eli, Jesse N. Moore, Andrea J. S. Stanaland, and Rosalind A. J. Wyatt (1998),
“Salesperson Race and Gender and the Access and Legitimacy Paradigm: Does
Difference Make a Differences?” Journal of Personal Selling & Sales
Management, 18 (4), 71-88.

112

 

Joseph, W. Benoy (1982), “The Credibility of Physically Attractive Communicators: A
Review,” Journal of Advertising, 11 (3), 15-24.

Kahle, Lynn B. and Pamela M. Homer (1985), “Physical Attractiveness of the Celebrity
Endorser: A Social Adaptation Perspective,” Journal of Consumer Research, 11
(March), 954-961.

Kahnna, Nikki (2004), “The Role of Reﬂected Appraisals in Racial Identity: The Case of
Multiracial Asians,” Social Psychology Quarterly, 67 (2), 115-131.

Kamins, Michael A. (1990), “An Investigation into the ‘Match-Up’ Hypothesis in
Celebrity Advertising: When Beauty May Be Only Skin Deep,” Journal of
Advertising, 19 (1), 4-13.

Kamins, Michael A., Meribeth J. Brand, Stuart A. Hoeke, and John C. Moe (1989),
“Two-Sided Versus One-Sided celebrity Endorsements: The Impact on
Advertising Effectiveness and Credibility,” Journal of Advertising, 18 (2), 4-10.

Kamins, Michael A. and Kama] Gupta (1994), “Congruence between Spokesperson and
Product Type: A Matchup Hypothesis Perspective,” Psychology & Marketing, 11
(6), 569-586.

Kaufman-Scarborough, Carol (2000), “Asian-American Consumers as a Unique Market
Segment: Fact or Fallacy?” Journal of Consumer Marketing, 17 (3), 249-262.

Keller, Gray F., and Creig R. Kronstedt (2005), “Connecting Confucianism, Communism,
and the Chinese Culture of Commerce,” Journal of Language for International
Business, 16(1), 60-75.

Kelman, Herbert C. (1961), “Processes of Opinion Change,” Public Opinion Quarterly,
25 (1), 57-78.

Kibra, Nazli (1998), “The Contested Meanings of ‘Asian American’: Racial Dilemma in
the Contemporary US,” Ethnic and Racial Studies, 21 (5), 939-985.

Lafferty, Barbara A. and Ronald E. Goldsmith (1999), “Corporate Credibility’s Role in
Consumers’ Attitudes and Purchase Intentions When A High versus a Low
Credibility Endorser is Used in the Ad,” Journal of Business Research, 44,
109-116.

113

Lee, Cheolhan, Yung Jung Choi, and Jong Hyuk Lee (2003), “Asian American’s
Evaluations of Racially Integrated Advertising: Effects of Endorser’s Race on
Ads,” in 2003 Asia-PaciﬁcConference Proceedings of the American Academy of
Advertising, 51-55.

Lee, Eun-Ju, Ann F iarhurst, and Susan Dillard (2002), “Usefulness of Ethnicity in
International Consumer Marketing,” Journal of International Consumer
Marketing, 14 (4), 25-48.

Lee, Wei-Na (1993), “Acculturation and Advertising Communication Strategies:
A Cross-Cultural Study of Chinese and Americans,” Psychology & Marketing,
10(5), 381-397.

Lee, Wei-Na and David K. Tse (1994), “Changing Media Consumption in a New Home:

Acculturation Patterns among Hong Kong Immigrants to Canada,” Journal of
Advertising, 23 (1), 57-70.

Lehman, Darrin R., Chi-yue Chiu and Mark Schaller (2004), “Psychology and Culture,”
Annual Reviews, 55, 689-714.

Liu, Annie H. and Mark P. Leach (2001), “Developing Loyal Customers with a
Value-adding Sales Force: Examining Customer Satisfaction and the Perceived

Credibility of Consultative Salespeople,” Journal of Personal Selling & Sales
Management, 21 (2), 147-156.

Lwin, May 0. and Chow-Hou Wee (2000), “The Inﬂuence of Spokesperson’s Race in
Relation to Accents Used in Television Commercials,” Journal of Current Issues
and Research in Advertising, 22 (1), 67-83.

MacKenzie, Scott B. and Richard J. Lutz (1989), “An Empirical Examination of the
Structural Antecedents of Attitudes toward the Ad in an Advertising Pretesting
Context,” Journal of Marketing, 53 (2), 48-65.

MacKenzie, Scott B., Richard J. Lutz, and George E. Belch (1986), “The Role of Attitude
toward the Ad as a Mediator of Advertising Effectiveness: A Test of Competing
Explanations,” Journal of Marketing Research, 23, 130-143.

Mandler, George P. (1982), “The Structure of Value: Accounting for Taste,” in Aﬂect and
Cognition: The 17’” Annual Carnegie Symposium, Margaret S. Clarke and Susan
T Fiske eds., Hillsdale, NJ: Erlbaum, 3-36.

Mazursky, David and Yaacov Shul (1992), “Learning from the Ad or Relying on Related

Attitudes: The Moderating Role of Involvement,” Journal of Business Research,
25, 81-92.

114

McGuire, William. J. (1985), “Attitudes and attitude change,” in Handbook of Social
Psychology, G Lindzey and E. Aronson, eds. Vol. 2, New York: Random House,
233-346.

McGuire, William (1984), “Search for the Self: Going Beyond Self-Esteem and the
Reactive Self,” in Personality and the Prediction of Behavior, R. A. Zucker, J.
Amoff, and A. I. Rabin, eds. New York: Academic Press, 73-120.

McGuire, William, J ., Claire V. McGuire, Pamela Child, and Terry F ujioka (197 8),
“Salience of Ethnicity in the Spontaneous Self-Concept as a Function of One’s
Ethnic Distinctiveness in the Social Environment,” Journal of Personality and
Social Psychology, 36 (5), 511-520.

Meyer-Levy, Joan, Therese A. Louie and Mary T. Owen (1994), “How Does the
Congruity of Brand Names Affect Evaluations of Brand Name Extensions,”
Journal of Applied Psychology, 79 (1), 46-53.

Meyers-Levy, Joan and Alice M., Tybout (1989), “Shema Congruity as a Basis for
Product Evaluation,” Journal of Consumer Research, 16 (June), 39-54.

Misra, Shekhar and Sharon E. Beatty (1990), “Celebrity Spokesperson and Brand

Congruence: An Assessment of Recall and Affect,” Journal of Business
Research, 21, 159-173.

Morimoto, Mariko and Carrie La F erle (2005, April), “Advertising Evaluations of Asian
American Consumers,” Paper presented at the 2005 Conference of the American
Academy of Advertising, Houston, TX.

Morimoto, Mariko and Carrie La Ferle (2002), “Representations of Asian Americans in
Magazine Advertisements,” in 2002 Conference Proceedings of the American
Academy of Advertising, 138-147.

Newell, Stephen J. and Ronald E. Goldsmith (2001), “The Development of a Scale to
Measure Perceived Corporate Credibility,” Journal of Business Research, 52,
23 5-2471

Ogden, Denise T., James R. Ogden and Hope Jensen Schau (2004), “Exploring the
Impact of Culture and Acculturation on Consumer Purchase Decisions: Toward a
Microcultural Perspective,” Academy of Marketing Science Review, 2004, 1.

Ohanian, Roobina (1990), “Construction and Validation of a Scale to Measure Celebrity

Endorsers’ Perceived Expertise, Trustworthiness, and Attractiveness,” Journal of
Advertising, 19 (3), 39-52.

115

Ohanian, Roobina (1991), “The Impact of celebrity Spokespersons’ Perceived Image on
Consumers’ Intention to Purchase,” Journal of Advertising Research, 31 (1),
46-54.

Okamura, Jonathan Y. (1981), “Situational Ethnicity,” Ethnic and Racial Studies, 4 (4),
452-465.

O’Mahony, Sheila and Tony Meenaghan (1998), “The Impact of Celebrity Endorsements
on Consumers,” Irish Marketing Review, 10 (2), 15- 24.

Osgood, Charles E. and Percy H. Tannenbaum (1955), “The Principle of Congruity in
The Prediction of Attitude Change,” Psychological Review, 62 (1), 42-55.

Ownbey, Shiretta F. and Patricia E. Horridge (1997), “Acculturation Levels and Shopping

Orientations of Asian-American Consumers,” Psychology & Marketing, 14 (1),
1-18.

Oyserman, Daphna and Izumi Sakamoto (1997), “Being Asian American: Identify,
Cultural Constructs, and Stereotype Perception,” Journal of Applied Behavioral
Science, 33 (4), 435-453.

Parker, Vanessa, Steve Sussman, David Crippens, Pam Elder, and Donna Scholl (1998),
“The Relations of Ethnic Identiﬁcation with Cigarette Smoking among US.
Urban African American and Latino Youth: A Pilot Study,” Ethnicity and Health,
3 (1-2), 135-143.

Patzer, Gordon L. “Source Credibility as a Function of Communicator Physical
Attractiveness,” Journal of Business Research, 11, 229-241.

Peracchio, Laura A. and Alice M. Tybout (1996), “The Moderating Role of Prior
Knowledge in Schema-Based Product Evaluation,” Journal of Consumer
Research, 23 (December), 177-192.

Pompitakpan, Chanthika (2004), “The Persuasiveness of Source Credibility: A Critical
Review of Five Decades’ Evidence,” Journal of Applied Social Psychology, 34
(2), 243-281.

Putreve, Sanjay and Kenneth R. Lord (1994), “Comparative and Noncomparative
Advertising: Attitude Effects Under Cognitive and Affective Involvement
Conditions,” Journal of Advertising, 23 (2), 77-91.

Raymond, J. (2001), “The Multicultural Report,” American Demographics, 23 (11),
S3-S6.

116

Reed, Americus II (2002), “Social Identity as 3 Useful Perspective for
Self-Concept-Based Consumer Research,” Psychology & Marketing, 19 (3),
23 5-266.

Rosenberg, Morris (1979), Conceiving the Self, New York: Basic Books.

Schank, Roger C. and Robert P. Abelson (1977), Scripts, Plans, Goals and
Understanding, Hillsdale, NJ: Erlbaurn.

Shavitt, Sharon (1990), “The Role of Attitude Objects in Attitude Functions,” Journal of
Experimental Social Psychology, 26 (2), 124-148.

Sirgy, M. Joseph (1982), “Self-Concept in Consumer Behavior: A Critical Review,”
Journal of Consumer Research, 9 (3), 287-300.

Stayman, Douglas M., Dana L. Alden and Karen H. Smith (1992), “Some Effects of
Schematic Processing on Consumer Expectations and Disconﬁrrnation
Judgments,” Journal of Consumer Research, 19 (September), 240-255.

Stayman, Douglas M. and Rohit Deshpande (1989), “Situational Ethnicity and Consumer
Behavior,” Journal of Consumer Research, 16 (3), 361-348.

Stephan, Cookie White and Walter G. Stephan (1989), “After Interrnarriage: Ethnic
Identity among Mixed-Heritage Japanese-American and Hispanics,” Journal of
Marriage and the Family, 51 (5), 507-519.

Tan, Chin Tiong and John U. Farley (1987), “The Impact of Cultural Patterns on
Cognition and Intention in Singapore,” Journal of Consumer Research, 13
(March), 540-544.

Tajfel, H. (1979), Diﬂerentiation between Social Groups. London: Academic Press.

Taylor, Charles R. and Ju Yung Lee (1994), “Not in Vogue: Portrayals of Asian
Americans in Magazine Advertising,” Journal of Public Policy & Marketing, 13
(Fall), 239-245.

Taylor, Charles R., Lee, Ju Yung, and Barbara B. Stern (1995), “Portrayals of African,
Hispanic, and Asian Americans in Magazine Advertising,” American Behavioral
Scientist, 38 (Feb.), 608-621.

Taylor, Charles R. and Barbara B. Stern (1997), “Asian-Americans: Television

Advertising and the “Model Minority” Stereotype,” Journal of Advertising, 26
(2), 47-61.

117

Till, Brian D. and Michael Busler (1998), “Matching Products with Endorsers:
Attractiveness versus Expertise,” Journal of Consumer Marketing, 15 (6),
576-586.

Till, Brian D. and Michael Busler (2000), “The Match-Up Hypothesis: Physical
Attractiveness, Expertise, and the Role of Fit on Brand Attitude, Purchase Intent
and Brand Belief,” Journal of Advertising, 29 (3), 1-13.

Triandis, Harry C. (1989), “The Self and Social Behavior in Differing Cultural Contexts,”
Psychological Review, 96 (3), 506-520.

Tripp, Carolyn, Thomas D. Jensen and Les Carlson (1994), “The Effects of Multiple
Product Endorsements by Celebrities on Consumers Attitudes and Intention,”
Journal of Consumer Research, 20 (4), 535-547.

Tversky, Amos and Itamar Gati (1978), “Studies of Similarity,” in Cognition and
Categorization, E. Rosch and B. B. Lloyd, eds., Hillsdale, NJ: Lawrence Erlbarun
Associates, Publishers, 79-98.

US. Census Bureau (2003), The Asian and Pacific Islander Population in the United
States: March 2002, Washington, DC: US. Government Publishing Ofﬁce.

US. Census Bureau (2000), Citizenship, Year of Entry, and Region of Birth,
http ://factﬁnder.census. gov/servlet/QTTable?_bm=y&-geo_id=0 1 000US&-qr_na
me=DEC_2000__SF3_U_QTP 1 4&-ds_name=DEC_2000_SF3_U&-_lang=en&-r
edoLog=false&-_sse=on (Visited: May 23, 2005).

US. Census Bureau (2001), The Population Proﬁle of the United States: 2000,
Washington, DC: US. Government Publishing Office.

Wallendorf, Melanie and Michael D. Reilly (1983), “Ethnic Migration, Assimilation, and
Consumption,” Journal of Consumer Research, 10 (December), 292-302.

Webster, Cynthia (1994), “Effects of Hispanic Ethnic Identiﬁcation on Marital Roles in
the Purchase Decision Process,” Journal of Consumer Research, 21 (2),
319-330.

Whittler, Tommy E. and Joan S. Spira (2002), “Model’s Race: A Peripheral Cue in
Advertising Messages?” Journal of Consumer Psychology, 12 (4), 291-301.

Whittler, Tommy E. (1989), “Viewers’ Processing of Source and Message Cues in
Advertising Stimuli,” Psychology & Marketing, 6 (4), 287-309.

118

 

Whittler, Tommy E. and Joan DiMeo (1991), “Viewers’ Reactions to Racial Cues in
Advertising Stimuli,” Journal of Advertising Research, 31 (6), 37-46.

Wilder, David D. (1981), “Perceiving Persons as Group: Categorization and Intergroup
Relations,” In D. L. Hamilton (Ed), Cognitive Processes in Stereotyping in
Intergroup Behavior, Hillsdale, NJ: L. Erlbaum Associates, 213-257.

Williams, Jerome D. and William J. Qualls (1989), “Middle-Class Black Consumers and
Intensity of Ethnic Identiﬁcation,” Psychology & Marketing, 6 (4), 263 -266.

Yzerbyt, Vincent, Charles M. Judd and Oliver Comeille (2004), “Perceived Variability,

Entitativity, and Essentialism,” in The Psychology of Group Perception:
Perceived Variability, Entitativity, and Essentialism, V. Yzerbyt, C. M. Judd and O.
Comeille, eds., New York: Psychology Press, 1-22.

Xu, J ing, Soyeon Shim, Sherry Lotz, and David Almeida (2004), “Ethnic Identity,
Socialization Factors, and Culture-Speciﬁc Consumption Behavior,” Psychology

& Marketing, 21 (2), 93-112.

Zaichokowsky, Judith Lynn (1985), “Measuring the Involvement Construct,” Journal of
Consumer Research, 12 (3), 341-352.

Zhang, Y. and Betsy D. Gelb (1996), “Matching Advertising Appeals to Culture: The
Inﬂuence of Products’ Use Conditions,” Journal of Advertising, 25 (3), 29-46.

119

   

Itjtijjjjtgjrjjttrill