Journal of Business Research 68 (2015) 561568

Contents lists available at ScienceDirect

Journal of Business Research

Consumers' reactions to global versus local advertising appeals:

A test of culturally incongruent images in China
Lianxi Zhou a,, Patrick Poon b,1, Haizhong Wang c,2
a
Department of Marketing, International Business, and Strategy, Goodman School of Business, Brock University, St. Catharines, Ontario L2S 3A1, Canada
b
Department of Marketing and International Business, Lingnan University, Tuen Mun, Hong Kong
c
Department of Marketing, Sun Yat-Sen Business School, Sun Yat-Sen University, Guangzhou, China

a r t i c l e i n f o a b s t r a c t

Available online 26 September 2014 This study experimentally examines consumers reactions to culturally incongruent images in advertising, which
are characterized by a mix of global and local cultural elements and symbols featured in product advertisements.
Keywords: Drawing on the persuasion knowledge perspective, the authors argue how advertising image incongruence in
Global consumer culture the globalized marketing communication context of China may lead to consumers' use of persuasion knowledge
Marketing communication (i.e., beliefs about the agent's persuasive tactics) and result in unfavorable attitudes toward the advertisements.
Persuasion knowledge
This research contributes to a better understanding of the interplay between global and local consumer cultures
China
in a globalized marketplace.
2014 Elsevier Inc. All rights reserved.

1. Introduction category. For instance, an incongruence of advertising image will appear

One of the most important challenges facing international marketing
and advertising managers today is the need to nd the right balance of
the conicting inuences of global and local consumer cultures (Arnett,
2002). In many parts of the developing world, Western/foreign brands
and global elements of consumption are well received. Behind these
global symbols is a corresponding set of traditional values and norms
that dene local identities. In China and other Asian countries, for exam-
ple, global and local cultural inuences have resulted in a mix of foreign
and local goods, images, advertising appeals and consumption
(Neelankavil, Venkatapparao, & Sessions, 1994; Zhou & Belk, 2004;
Zhou, Yang, & Hui, 2010). The conceptualization of local and global
identities by Arnett (2002) suggests that there is a complex interplay
between local context and global content, which makes it hard to
argue for the primacy of one over the other.
While previous research has explored the ways in which consumers
aspire, reject, or identify the cultural meanings of advertisements
among conicting values in a globalizing world (e.g., Hung, Li, & Belk,
2007; Zhou & Belk, 2004), this study investigates the effects of culturally
incongruent advertising images on consumers' evaluative responses to
advertisements. Specically, this paper focuses on one specic type of
culturally incongruent advertising image which is dened as the mis-
match between model ethnicity and the culture of origin of the product
Corresponding author. Tel.: +1 905 688 5550x5832.
E-mail addresses: lzhou@brocku.ca (L. Zhou), patpoon@ln.edu.hk (P. Poon),
wanghzh.05@alum.sem.tsinghua.edu.cn (H. Wang).
1
Tel.: +852 2616 8235.
2
Tel.: +86 20 222 3453.
when a Western model is used to promote an Eastern culture associated
product, and vice versa. The study reects a central issue in advertising
research with regard to the perceived t of the advertising image with
the viewer's contingent situation (Scott, 1994).
In the midst of conicting inuences of global and local cultural
images in modern China (Hung et al., 2007; Zhou et al., 2010), this
study intends to provide evidence on how the cultural incongruence be-
tween global and local elements affects consumers' cognitive and affec-
tive reactions to advertisements. Although the effect of incongruent
advertising element on consumers' attitudes toward advertisements or
brands has been well recognized in the related literature (e.g., Heckler
& Childers, 1992; Lee & Mason, 1999), the theoretical rationale and
general ndings may not necessarily apply to the issues arising from a
new cultural system consisting of a mixture of global and local elements.
This research involves two experimental studies to test the effects of
culturally-based incongruence of advertising image (hereafter referred
to as ad image) on Chinese consumers' responses. Model ethnicity is con-
sidered as one of the experimental factors largely due to the fact that the
use of foreign models is a distinctive feature of most magazine advertis-
ing in modern China today (Hung et al., 2007). Also, advertisements
using models with different cultural backgrounds reect marketing ef-
fort for adaptation to local culture in international advertising strategy
(Zhou & Belk, 2004). Besides this, model ethnicity has diverse effects
on consumers' responses toward the type of product being advertised
(Martin, Lee, & Yang, 2004).
Another experimental factor of interest is referred to as culture of
origin of product category (COPC). Unlike the concept of country of or-
igin (COO) for a product or a brand, COPC represents a productculture
association, which is dened as consumers' stereotypical beliefs about

http://dx.doi.org/10.1016/j.jbusres.2014.09.006
0148-2963/ 2014 Elsevier Inc. All rights reserved.
562 L. Zhou et al. / Journal of Business Research 68 (2015) 561568
products that are associated with a particular culture of origin (Laroche,
Papadopoulos, Heslop, & Mourali, 2005). For example, pizza is widely
perceived as a Western/foreign product among Indian consumers, re-
gardless of extensive localization efforts among marketers (Eckhardt,
2005). By the same token, lanterns and silk are typical products that
are associated with Chinese traditions and culture. Other examples
include Korean kimchi and Japanese sushi. By combining two types of
model ethnicity (Western vs. Chinese) and two types of COPC (local
vs. foreign) into an experimental design, this study is able to examine
the incongruence effects of advertising images pertaining to the
interplay between global and local cultural elements. Unlike previous
research, this study focuses particularly on the mismatch between
COPC and culturally based advertising images. As a result, this research
extends the conceptual domain of advertising incongruity to culturally
based elements or symbols, and provides novel insight about con-
sumers' responses to the interplay involving global and local advertising
appeals.
2. Conceptual background and hypotheses
2.1. Local and global advertising images
While there are increasing tensions regarding consumer preferences
for local vs. global identities (Zhang & Khare, 2009), consumers
nowadays tend to simultaneously demand both global and local
goods. According to Hung et al. (2007), even with the adoption of a
global popular culture, consumers tend to scrutinize foreign goods and
images through the lens of a local perspective. As a result, a new global
cultural system is increasingly expressed in advertising and marketing
communications as the ideology of contemporary consumption. This
phenomenon is particularly apparent in the context of globalization in
China (Hung et al., 2007). Similar to other developing countries, the
consumers in China today are inuenced by a mixture of foreign and
local goods, images, advertising appeals and consumption (Davis,
2000; Nelson & Paek, 2007; Tai & Pae, 2002). Although the trend using
Western advertising appeals seems still to be on the rise, foreign adver-
tisers nowadays tend to locally adapt certain aspects of their advertise-
ments to cater for the Chinese market (Cheong, Zheng, & Kim, 2011).
From a theoretical standpoint, the literature has reected on an in-
creasing number of studies that have used content analysis of advertise-
ments to examine the effects of global and local advertising appeals
(e.g., Tse, Belk, & Zhou, 1989; Zhang, 2009). The ndings from these
studies suggest that despite the fact that there has been an increase in
the use of Western images and appeals in China's mass media, Chinese
consumers embrace both traditional and contemporary foreign values.
They tend to make sense of foreign features in the advertisements
(Hung et al., 2007; Zhou & Belk, 2004). The primary function of this
evaluation process is based on self-referencing responses of their own
lifestyles and aspirations which are often constructed through the lens
of traditional values and norms on one hand, and the authenticity and
symbolism associated with global or Western images on the other
(Ozsomer, 2012; Steenkamp, Batra, & Alden, 2003).
2.2. Image incongruence in advertising
The roots of image incongruence in advertising stem from earlier
research relating to t or the match-up effect between a celebrity or
spokesperson and a particular product or brand characteristic
(e.g., Friedman & Friedman, 1979; Kahle & Homer, 1985; Kamins, 1990).
Prior research in this area suggests that in order for such advertising to
be effective, a good match-up between the spokesperson image or per-
sonality and the nature of the product being advertised should exist.
The basis for the effectiveness of endorserproduct congruence in adver-
tising can be linked to a variety of theoretical perspectives. Among them,
schema-based explanations have been proven to be a useful framework
(Lee & Thorson, 2008; Lynch & Schuler, 1994; Misra & Beatty, 1990).
Schema congruity occurs when the image of an endorser and the image
of a product are relevant, such that the relationship between them can
be readily accessible within the existing celebrity and/or product category
(Lee & Thorson, 2008).
While the schemacongruity framework provides a key theoretical
basis for a better understanding of advertising image congruence
through matching the image of an endorser with a product, the design
of previous research has not tended to fully address other aspects of
image congruence in advertising. In particular, the match-up effect has
not been examined in relation to the cultural elements of advertising
appeals. As a result, this paper attempts to extend the match-up hypoth-
esis by exploring the domain of globalized advertising. The purpose is to
determine the possibility that consumers may respond less (vs. more
favorably) to culturally incongruent advertising images.
2.3. Culturally incongruent advertising images and persuasion knowledge
theory
Within the context of this study, culturally incongruent advertising
images are characterized by a mismatch between the culture of origin
of product category (COPC) and model ethnicity. They refer to the
usage of a Western model/endorser on a typical local product (Eastern)
or the usage of a local model/endorser (Eastern) on a typical Western
product. Research drawing on schema-based explanation for the
match-up effect suggests that a high degree of incongruity cannot be re-
solved within activated schemas or can only be resolved if major chang-
es are made in the existing cognitive structure (Lee & Thorson, 2008). In
response, this paper draws on the persuasion knowledge model
(Freidstad & Wright, 1994, 1995) as an alternative theoretical perspec-
tive to explain consumer responses to culturally incongruent advertis-
ing images.
Freidstad and Wright (1994) was the rst to propose the theory of
persuasion knowledge. The theory asserts that consumers use their
persuasion knowledge to recognize, analyze, interpret, and evaluate per-
suasion attempts. Central to persuasion knowledge is consumer percep-
tions of persuasion tactics that marketers can use to inuence their
beliefs, attitudes, decisions or actions. Such perceptions are signicantly
related to how consumers respond to persuasion attempts. As presented
in the persuasion knowledge model (PKM), three knowledge structures
would interact to shape and determine the outcomes of the marketers'
persuasion attempts. These are: persuasion knowledge, agent knowledge,
and topic knowledge. In an advertising context, persuasion knowledge re-
fers to consumers' beliefs about the tactics or persuasion attempts used
in advertising. Agent knowledge refers to consumers' beliefs and feelings
about the advertiser while topic knowledge refers to consumers' beliefs
about the advertised product. These knowledge structures develop
over time as consumers are exposed to tactics and come to recognize
them as such.
Surrounding the PKM model, several studies have been conducted.
Campbell and Kirmani (2000) investigated conditions that inuence
consumers' use of persuasion knowledge in an interpersonal sales
context. They proposed that persuasion knowledge is used when con-
sumers draw an inference about a persuasion motive which may under-
lie a salesperson's behavior (such as attering). These motive inferences
then affect the consumers' perceptions about the salesperson. Campbell
and Kirmani (2000) conducted four experimental studies with the use
of role-playing scenarios to test their conceptual model, which
described accessibility and cognitive capacity as underlying factors
that inuenced consumers' use of persuasion knowledge. The study in-
dicates that persuasion knowledge can be activated fairly easy when
persuasion motives are made accessible.
In an advertising context, Campbell (1995) found some evidence
about consumers' use of persuasion knowledge in response to advertis-
ing tactics. By focusing on attention-getting tactics used in television ad-
vertising, she discovered how these tactics may sometimes lead
consumers to infer that the advertiser is attempting to manipulate the
 L. Zhou et al. / Journal of Business Research 68 (2015) 561568
audiences. Further, the inferences of manipulative intent may lower
advertising persuasion (e.g. advertising attitudes, brand attitudes and
purchase intentions). In particular, she emphasized the importance of
t which refers to the extent that the advertising appeal matches
the consumers' schemas or expectations for the advertised product.
Advertisers sometimes use appeals with a low t and hope these unex-
pected elements can attract consumers' attention in product informa-
tion processing. The research by Campbell (1995) suggests that by
raising the information processing level, these tactics may increase the
activation of persuasion knowledge from the viewers which potentially
make them less susceptible to the persuasion attempts.
Other advertising tactics have also been studied. For example, Cotte,
Coulter, and Moore (2005) examined a commonly employed emotional
tactic (i.e., guilt appeal) and found that when consumers infer manipu-
lative intent, they do not feel guilty, but have negative attitudes toward
the advertisement. Ahluwalia and Burnkrant (2004) proposed a
conceptual model of rhetorical question effects. They argued that
consumers high in persuasion knowledge tend to be more sensitive to
the meaning of symbols used in advertising. If consumers perceive a
persuasion tactic, they tend to draw inferences of some sort, disengage
somewhat from the advertisements, or discount the advertising claims
(Freidstad & Wright, 1995).
From the PKM perspective, consumers' use of persuasion knowledge
can explain various advertising effects. When exposed to an advertise-
ment, consumers may access their knowledge about persuasion
attempts that they feel the advertisers may use (Wright, Friestad, &
Boush, 2005). The use of persuasion knowledge seems to be more prev-
alent when the advertising appeal is incongruent with the consumers'
schemas or expectations for the advertised product (Freidstad &
Wright, 1994). For instance, when a Western model is used to endorse
a typical local product, most consumers may feel that the advertiser at-
tempts to promote the popularity of the local product by emphasizing
its global reach potential and authenticity. Owing to the poor t of the
advertising appeal with the consumers' prior expectations for the ad-
vertised product, the authors expect that the consumers are generally
suspicious of the advertiser's motivation and tend to produce substan-
tial counterarguments or negative feelings about the advertiser's
persuasive attempts. In contrast, when exposed to congruent advertise-
ments (featuring a t between model ethnicity and COPC), the
consumers may consider the persuasive intent to be less salient and
subsequently their inferences to persuasion knowledge tend to be
lower. The consumers are more inclined to form beliefs consistent
with the advertising claims and generate less counterarguments and
unfavorable emotions evoked by the advertisement (Batra & Stephens,
1994; Obermiller & Spangerberg, 1998). Therefore:
Hypothesis 1. Consumers' use of persuasion knowledge is higher for
culturally incongruent advertising images than the culturally congruent
ones.
In line with the prediction of consumers' use of persuasion knowledge,
the authors further expect that consumers tend to have less favorable at-
titudes toward the advertisements that are characterized by culturally in-
congruent images than those featured by culturally congruent ones. For a
product associated with a local culture of origin, local consumer identity is
readily activated and accessible (Zhang & Khare, 2009). This may then
make the use of a Western model in the advertising become less appeal-
ing. However, the incongruent advertising image may increase the
consumers' processing attention to the exaggeration of advertising com-
munication, such that the consumers may draw some sorts of inferences
about their expectations of the COPC. Consumers' responses to the per-
suasiveness of the advertising images would be reduced. Similarly,
when a typical foreign product is featured in an advertisement, global
consumer identity is readily activated and accessible (Zhang & Khare,
2009). Using a local endorser for the product with a foreign culture of or-
igin may lead to a lower degree of trustworthiness associated with such
563
advertisement and also a weakened belief about the advertised product.
In contrast, when consumers are exposed to a culturally congruent adver-
tisement, their usage of persuasion knowledge is to a lesser degree. There-
fore, they tend to have less inconsistent beliefs derived from an accessible
local or global identity with the advertising appeals, which can make
them feel more positive about the advertisements and the advertised
products. Consequently, from the perspective of persuasion knowledge
combining with local and global consumer identities, consumers'
attitudes toward advertisements should be more positive for culturally
congruent advertising images.
Hypothesis 2. Consumers' attitudes toward advertisements are more
positive for culturally congruent advertising images than the culturally
incongruent ones.
3. Study 1
Two experimental factors were manipulated in the research design.
The rst one is model ethnicity. As noted earlier, the use of foreign/
Western models to signify global consumer cultures is commonly used
in developing countries and represents one of the most dominant global
approaches featured in advertising. The second factor is cultural origin of
product category (COPC), which refers to cultural stereotypes of the prod-
uct category or perceived productculture association (Laroche et al.,
2005). This factor represents another global versus local cultural ele-
ment. The study adopted a 2 (model ethnicity: Western vs. Chinese)
2 (COPC: foreign vs. local) between-subjects factorial design. The con-
gruent/incongruent advertising images were operationalized as the
match/mismatch between the advertised product category (foreign ver-
sus local) and the model ethnicity (Western versus Chinese).
3.1. Selection of local versus foreign product category
A pilot study was conducted to choose typical local and foreign prod-
ucts for the actual experiment. Following a similar procedure used by
Dahlen and Lange (2004) when determining brand associations, the
study involved ten undergraduate business students from a major uni-
versity in China and asked them to develop a list of typical Chinese
and Western product categories. The participants were given examples
like lanterns and pizza, but were emphasized to provide products which
were strongly associated with something either Chinese or Western.
The ve most commonly mentioned products were then chosen for
follow-up verication. The local product categories included dumplings,
steamed stuffed bun, steamed vermicelli roll, herbal tea and Chinese
liquors. The Western product categories included pizza, egg custard
tart, coffee, hamburger and perfume. In considering the most relevance
of the product categories to the target audiences, food items were nally
chosen as the experimental stimuli (i.e., typical Chinese food: steamed
vermicelli roll; typical Western food: egg custard tart which is actually
adapted from English custard tart and widely recognized as a foreign
food in China and other Asian countries).
3.2. Design of advertising stimulus
Four hypothesized advertisements were professionally developed by
a copywriter of a leading international advertising agency located in
Guangzhou. The advertising agency adopted real products and brands
steamed vermicelli roll for Real Kungfu and egg custard tart for KFC
(Kentucky Fried Chicken). Pictures of the products were collected from
commercial advertisements via an Internet search. To resemble the
advertising featured in the marketplace, real brand logos and slogans
from advertisements were included in the mock advertisements used in
this study.
A local Chinese model and a Western model were created to feature
alongside the advertised products respectively. Simulating real world
564 L. Zhou et al. / Journal of Business Research 68 (2015) 561568
print advertisements, the mock advertisements consisted of a big pic-
ture of the model with a small picture of the product and the logo/
slogan in the upper right corner (see Appendix 1). Except for the exper-
imental controlled factors, all the other advertising elements, including
the size of the picture, the background color and the composition of the
picture, were kept identical so that the effects of advertising image con-
gruence/incongruence could be adequately tested. The professional
copywriter conrmed that the four hypothesized advertisements all
appeared to look like real advertising copies.
3.3. Pretest of the mock advertisements
The advertisements were pretested twice prior to conducting the
main study. The objectives of the two pretests were to (1) verify the ma-
nipulation of congruence/incongruence between model ethnicity and
COPC and (2) make sure no confounding effects exist. Thirty subjects
were invited to participate in the pretest conducted in the same univer-
sity in China as before. The rst pretest was administered to ensure the
manipulation of advertising image incongruence. Subjects were
required to rate the t between the model ethnicity and the COPC on
a ten-point scale, for example, How well do the model and product
t together? (Dahlen & Lange, 2004). The mean values were 7.6
(steamed vermicelli roll) and 7.0 (egg custard tart) for the congruent
subset advertisements (p = .221) and 3.0 (steamed vermicelli roll)
and 2.5 (egg custard tart) for the incongruent subset advertisements
(p = .325). The results show a signicant difference between groups
(F = 61.727, p b 0.001) and suggest that manipulation of the advertis-
ing image incongruence in the advertisements would be satisfactory.
The second pretest was used to test the control variables and make
sure that no confounding effects were present. Two control variables
were measured on seven-point scales. The rst control variable was ad-
vertising attractiveness (copy style and visual elements). Subjects were
shown four advertisements (one at a time) and asked to rate their at-
tractiveness on a seven-point scale (1 = extremely not attractive,
7 = extremely attractive). The second control variable was likeability
of model used (Lee, 2000). Subjects were shown two model images
taken from the stimulus advertisements. They were asked how they
liked the model on a seven-point scale (1 = don't like at all, 7 = like
very much). Based on the ANOVA results, the advertisements did not
differ signicantly from each other in terms of their attractiveness
(F = .847, p = .472) and also the subjects did not show any signicant
difference in terms of the models' likeability (Western model = 5.4;
Chinese model = 5.2, p = .141). The results suggest that any confound-
ing effects due to advertising attractiveness or liking of the model type
would be marginal.
3.4. Experimental procedure
The two COPCs (i.e., egg custard tart and steamed vermicelli roll)
were matched with two model ethnicities (a Western model and a
Chinese model) to create four different advertisements. These four ad-
vertisements formed four experimental conditions (see Appendix 1):
two were congruent advertising images (steamed vermicelli roll with
the Chinese model; egg custard tart with the Western model) and the
other two were incongruent images (steamed vermicelli roll with the
Western model; egg custard tart with the Chinese model).
In the actual experiment, one hundred undergraduate business stu-
dents were recruited and randomly assigned to one of the four experi-
mental conditions (25 subjects for each condition). Each subject was
instructed to read the target advertisement along with three other ad-
vertisements (serving as lters). The sequence of the advertisements
was randomly rotated in each condition and the subjects were
instructed to provide their opinions about the advertisements by
completing a short questionnaire which contained measures about
manipulation checks and the dependent variables (i.e., consumers' use
of persuasion knowledge and attitudes toward the advertisement).
The subjects were told that a local advertising agency was interested
in their opinions about the newly designed advertisements in appealing
to their segment of the market. They were then instructed to read each
of the stimulus advertisements at their own pace and complete three
sections of the questionnaire. The rst section consisted of questions
about their attitudes toward the advertisements and manipulation
check variables. The second section included questions concerning
their use of persuasion knowledge and reactions to other aspects of
the advertisements. The third section contained questions relating to
demographic information. After completing the questionnaire, the sub-
jects were debriefed with a written description of the general purpose
of the study. No subject was aware of the main purpose of the experi-
ment before the debrief.
3.5. Measures
The study measured three sets of variables: advertising image incon-
gruence, individual's use of persuasion knowledge, and ad attitudes.
Most scales were adapted from previous studies. The perceived congru-
ence scale from Menon and Kahn (2003) was used to measure advertis-
ing image incongruence. Subjects were asked to rate on a seven-point
Likert scale ranging from 1 (denitely disagree) to 7 (denitely
agree). The items were stated as follows: 1) The ad image is logically
related to the advertised product; 2) The ad image is relevant to the
advertised product; 3) The ad image is compelling for the advertised
product; 4) It is strange to see the ad image in advertisements of this
kind of product; 5) The ad image is congruent with the advertised prod-
uct; and 6) Overall, the match between the product and the ad image is
appropriate. Except item 4, all the other items were reversed so that the
higher the score, the more incongruent the advertising image.
Individual's use of persuasion knowledge was adapted from three
different sources used in previous studies (Campbell, 1995; Campbell
& Kirmani, 2000; Moore & Rodgers, 2005). Based on a seven-point Likert
scale ranging from 1 (denitely disagree) to 7 (denitely agree), sub-
jects responded to the following items: 1) The way this advertisement
tries to persuade people seems acceptable to me (reverse scale);
2) The advertiser tries to manipulate the audience in ways that I don't
like; 3) I am annoyed by this advertisement because the advertiser
seemed to be trying to inappropriately manage or control the consumer
audience; 4) The advertisement was fair in what was said and shown
(reverse scale); 5) When I read the advertisement, I think it's pretty ob-
vious that it is trying to persuade me to buy the product; 6) I notice
tricks in this advertisement to promote the product; and 7) This adver-
tisement is meant to sell the product. As commonly used in previous
studies, attitudes toward advertisement (ad attitudes) were measured
on a 7-point semantic differential scale with three items: likeable/
unlikable, positive/negative, and favorable/unfavorable.
3.6. Analyses and results
3.6.1. Manipulation checks
A manipulation check is essential to determine whether the experi-
mental manipulation produced the desired controlled condition. In this
study, the t between model ethnicity and the COPC was measured to
check the experimental conditions of advertising image incongruence.
A 2 2 ANOVA test was conducted. As manipulated, there was no signif-
icant difference between the congruent advertisements the mean
values of advertising image congruence were 7.76 (steamed vermicelli
roll) and 7.48 (egg custard tart) (p = .309). Also, there was no signi-
cant difference between the incongruent advertisements the mean
values of advertising image incongruence were 2.84 (steamed vermicel-
li roll) and 2.92 (egg custard tart) (p = .771). As expected, there was a
signicant difference between groups (F = 200.089, p b 0.001). The
manipulation of the advertising image incongruence was deemed
satisfactory.
 L. Zhou et al. / Journal of Business Research 68 (2015) 561568
Furthermore, analysis was performed to examine other control
variables, as in the pretests. First, a 2 2 ANOVA test was conducted
to examine the attractiveness of each advertisement. As expected, the
results indicate that the subjects perceived no difference in the attrac-
tiveness of the advertisements (F = .868, p = .460). Second, the likabil-
ity of each model type was compared. The 2 2 ANOVA test reveals that
there was no signicant difference in the likability of the model types
(F = .487, p = .692). Thus, potential confounding effects due to other
aspects of the advertisements such as copy style and visual elements
were minimal.
3.6.2. Hypothesis testing
The hypotheses indicate that individuals' use of persuasion knowl-
edge is higher for culturally incongruent ads than culturally congruent
ads (Hypothesis 1), and attitudes toward culturally congruent ads
would be more positive than the attitudes toward culturally incongru-
ent ads (Hypothesis 2). Specically, in the congruent conditions,
subjects should be less likely to use persuasion knowledge, and thus
would respond more positively to the advertisements. A reverse pattern
could be identied among subjects in the incongruent conditions. As a
baseline information, the original experimental variables (COPC and
model ethnicity) were examined simultaneously through a MANOVA
test. Table 1 shows the results.
Based on the results, there are signicant interactions between COPC
and model ethnicity in terms of ad attitudes (F = 395.28, p b .001) and
individual's use of persuasion knowledge (F = 309.121, p b .001). The
MANOVA analysis also reveals that there is a signicant main effect of
model ethnicity on ad attitudes (F = 13.966, p b .01), but not on
individual's use of persuasion knowledge (F = .019, p = .892). There
is no signicant main effect of COPC on any dependent variables.
For the purpose of hypothesis testing, the experimental conditions
were re-organized by creating a dummy variable, and assigned 0 to
the congruent advertising image condition, and 1 to the incongruent
advertising image condition. Congruent condition represents the original
observations of foreign product category with a Western model, and
local product category with a Chinese model. In contrast, incongruent
condition represents the original observations of foreign product
category with a Chinese model, and local product category with a for-
eign model. Product category (COPC) was excluded in the subsequent
analysis since it had no effect on any dependent variables. Table 2
presents the mean values of the two dependent variables, which are
organized along with the conditions involving congruent/incongruent
advertising images by model ethnicity. The dependent variables were
examined with the two-factor ANOVA model.
In testing the effects of individual's use of persuasion knowledge,
there is a signicant main effect of advertising image incongruence.
Subjects in the congruence condition exhibit a mean value of 2.50 in
their use of persuasion knowledge, which is signicantly smaller than
the mean value of 5.08 as demonstrated by those in the incongruence
condition (F = 17.32, p b .001). Hypothesis 1 is therefore supported.
Besides, the effect of model ethnicity is not signicant on individual's
use of persuasion knowledge. The advertisements featuring the
Western model has a mean value of 3.80 which is not signicantly
higher than the mean value of the advertisements adopting the Chinese
Table 1
The results of ad model ethnicity and product category in Study 1.
Source Dependent variables F Sig.
Product category Ad attitudes .201
Individuals' use of persuasion knowledge .910
Model ethnicity Ad attitudes 13.96
Individuals' use of persuasion knowledge .0190
Product category Ad attitudes 395.28
model ethnicity Individuals' use of persuasion knowledge 309.12
565
Table 2
Mean values and hypotheses testing in Study 1.
Congruent ad image Incongruent ad image Main effect
Ad attitudes
Western ad model 5.52 2.80 4.16
Chinese ad model 4.96 2.36 3.66
Main effect 5.24 2.58
Individual's use of persuasion knowledge
Western ad model 2.44 5.16 3.80
Chinese ad model 2.56 5.00 3.78
Main effect 2.50 5.08
model (3.78) (F = .019, p = .892). Also, the mean values do not yield
a signicant interaction effect involving congruent/incongruent ad
images and model ethnicity (F = .910, p = .342).
There is a signicant main effect of advertising image incongruence
on ad attitudes. The congruent advertisements receive a mean value of
5.24, which is signicantly higher than the mean value of the incongru-
ent advertisements (2.58) (F = 18.15, p b .001). Hypothesis 2 is
supported. In addition, there is a signicant main effect of model ethnic-
ity. The advertisements with the Western model receive a mean value of
4.16 which is signicantly higher than the mean value of the advertise-
ments featuring the Chinese model (3.66) (F = 13.96, p b .001). Be-
sides, there is no signicant interaction effect between congruent/
incongruent ad images and model ethnicity (F = .201, p = .655).
In summary, the experimental study indicates that advertising
image incongruence tends to increase individual's use of persuasion
knowledge but decrease ad attitudes. Furthermore, the negative effect
of advertising image incongruence on ad attitudes is found to be highly
associated with individual's greater use of persuasion knowledge. As
theorized, individual's use of persuasion knowledge may account for
the inuence of advertising image incongruence on ad attitudes. Also,
not surprisingly, the results show that ad attitudes are generally more
positive for the advertisements with Western appeals than the ones
with local (Chinese) appeals.
4. Study 2
4.1. Overview
In order to further validate the theoretical explanation for con-
sumers' responses to culturally incongruent advertising images, this re-
search involved a follow-up study using real advertisements and
another product category. Consumer skepticism toward advertising
(or ad skepticism, as an individual difference variable) was introduced
to determine its potential impact on individuals' use of persuasion
knowledge in response to culturally incongruent advertisements.
Consumer skepticism refers to the consumers' general tendency toward
disbelief of advertising claims (Obermiller & Spangerberg, 1998). Draw-
ing on the persuasion knowledge model (PKM) in advertising effects,
consumers who are skeptical of advertising are apt to generate greater
persuasion knowledge to dislike the advertisements as well as the ad-
vertised products. In other words, by demonstrating the moderating in-
uence of ad skepticism on individuals' use of persuasion knowledge
related to culturally incongruent advertising images, this study intends
to provide further support for the research hypotheses and the underly-
ing theoretical rationale.
A further 2 (ad image: congruence vs. incongruence) 2 (ad
skepticism: low vs. high) experimental study with a between-subjects
.655
.342 factorial design was conducted. Since there was no signicant difference
.000 in individuals' use of persuasion knowledge in their responses to incon-
.892 gruent advertising images involving local and foreign product catego-
.000 ries in Study 1, a local product (i.e., Shun herbal tea) was chosen as an
.000
advertising stimulus in Study 2. To increase the validity of the results,
566 L. Zhou et al. / Journal of Business Research 68 (2015) 561568

a real advertisement for the local brand was used. The real advertise-
ment featured a Chinese model for the brand (congruent advertising
image). As in Study 1, a copywriter at a leading international advertising
agency in Guangzhou was appointed to develop a copy of the incongru-
ent advertisement. The new advertisement was the same as the original
one except that the Chinese model was replaced by a Western model
(see Appendix 2).
As in Study 1, this study involved 180 undergraduate business
students in a major Chinese university to participate in the follow-up
study. Subjects were randomly assigned to the congruent and incongru-
ent advertising image conditions (90 subjects for each experimental
condition). Ad skepticism was measured using a nine-item SKEP scale
developed by Obermiller and Spangerberg (1998). A median split ap-
proach was used to separate subjects into low versus high ad skepticism
conditions. As in Study 1, ad attitudes and individual's use of persuasion
knowledge were used as the dependent variables. Consumer attitudes
toward the advertised brand were also measured. In line with the
previous research, consumer attitudes toward the advertisements can
also be extended to their attitudes toward the advertised brand.
4.2. Manipulation checks
Following the same procedure used in Study 1, the cultural t be-
tween model ethnicity and the COPC was veried. As manipulated,
there was a signicant difference in the perception of t between
the congruent advertising image condition (mean value = 5.54) and
the incongruent advertising image condition (mean value = 2.33)
(p b 0.001). The manipulation of the ad image congruence/incongruence
was therefore considered satisfactory. The likability of the model type
and the attractiveness of the advertisements were also compared. No
signicant differences were found.
4.3. Analyses and results
Table 3 presents the main and interaction effects of the experimental
factors on the three dependent variables (i.e., ad attitudes, brand atti-
tudes and individual's use of persuasion knowledge). As expected, the
results of the MANOVA test reveal that advertising image incongruence
has a signicant main effect on ad attitudes (F = 856.020, p b .001) and
individual's use of persuasion knowledge (F = 925.788, p b .001).
No signicant effect is observed on brand attitudes (F = 854.909,
p = .239). In line with the predictions, there is a signicant main effect
of skepticism toward advertising on ad attitudes (F = 5.854, p b .01)
and also on individual's use of persuasion knowledge (F = 19.056,
p b .001). Brand attitudes are not affected (F = 1.690, p = .188).
More interestingly, the results also demonstrate a signicant interaction
effect of advertising image incongruence and skepticism toward
advertising on ad attitudes (F = 10.712, p b .001) as well as on
individual's use of persuasion knowledge (F = 9.113, p b .001). The in-
signicant results on brand attitudes may be due to the subjects' famil-
iarity or prior usage experience with the real brand.
The main and interaction effects were examined by using a 2-way
ANOVA test. Table 4 shows the results. Consistent with the research hy-
potheses, subjects in the ad image incongruence condition demonstrate
lower ad attitudes and greater use of persuasion knowledge than those
in the ad image congruence condition. Also as expected, high skeptical
individuals demonstrate a greater tendency to use persuasion knowl-
edge and dislike the advertisements more than those individuals with
low ad skepticism.
More importantly, the results show a clear pattern of the interaction
effects. For the ad image congruence condition, the group differences
between low and high skepticism toward advertising are not signicant
on ad attitudes (F = 1.039, p = .358) and individual's use of persuasion
knowledge (F = 1.703, p = .188). For the ad image incongruence con-
dition, consumers with high skepticism toward advertising demon-
strate a greater use of persuasion knowledge (F = 22.142, p b .001)
and less positive ad attitudes (F = 19.729, p b .001) than their counter-
parts. As a result, ad skepticism tends to moderate consumers'
responses to advertising communication pertaining to culturally incon-
gruent images or symbols. This provides an additional insight to the
ndings reported in Study 1.
5. Conclusion and discussion
The purpose of this research was to examine how culturally in-
congruent advertising images can affect consumers' responses to
global versus local advertising appeals. This study provides fresh in-
sights that can contribute further to the debate concerning the use of
global, foreign, or local cultural components in advertising commu-
nications parallel to the growth of the global marketplace (Alden,
Steenkamp, & Batra, 1999). Specically, the study demonstrates
that consumers' attitudes toward advertisements are less positive
when the cultural identities of advertising appeals such as model
ethnicity and culture of origin of product category (COPC) are per-
ceived incongruent. This nding can be explained by the fact that in-
dividuals are more likely to use persuasion knowledge when the
cultural incongruence of advertising images prevails. As a result,
the use of persuasion knowledge is more profound when consumers
exhibit a higher level of ad skepticism.
The authors recognize that previous studies have examined in-
congruity in advertising. However, there are different forms of in-
congruence related to advertising communications. This study
focuses particularly on the type of incongruence regarding the cul-
turally based advertising images and the cultural meaning of a

Table 4
Table 3 Mean values and hypotheses testing in Study 2.
Results of ad image incongruence and consumer skepticism toward advertising in Study 2. Congruent ad Incongruent ad Main effect
Source Dependent variables F Sig. Ad attitudes
Ad image incongruence Ad attitudes 856.020 .000 Low skepticism 5.59 2.87 4.23
Attitudes toward the brand 854.909 .239 High skepticism 5.33 1.95 3.64
Individuals' use of persuasion 925.788 .000 Main effect 5.46 2.41
knowledge Brand attitudes
Skepticism toward Ad attitudes 5.854 .003 Low skepticism 5.59 4.90 5.24
advertising Attitudes toward the brand 1.690 .188 High skepticism 5.44 4.71 5.07
Individuals' use of persuasion 19.056 .000 Main effect 5.55 4.80
knowledge
Ad image incongruence Ad attitudes 10.712 .000 Individual's use of persuasion knowledge
skepticism toward Attitudes toward the brand .474 .623 Low skepticism 2.55 4.81 3.68
advertising Individuals' use of persuasion 9.113 .000 High skepticism 2.73 5.82 4.27
knowledge Main effect 2.64 5.31
 L. Zhou et al. / Journal of Business Research 68 (2015) 561568 567

product. The theoretical development lies in the persuasion knowl- Appendix 1. Stimulus advertisements used in Study 1
edge perspective (Freidstad & Wright, 1994, 1995) and has gone be-
yond the previous information processing theories. Specically, this Local product features local ad model Western product features local ad model
research contributes to a better understanding of global and local ad-
vertising from the perspective of the persuasion knowledge model
(PKM). PKM is a general theoretical paradigm that helps to explain
how consumers respond to marketers' attempts in persuasion. The
ndings indicate that persuasion knowledge takes place when per-
suasive intent is salient (i.e., when the model ethnicity is incongru-
ent with the cultural meaning of a product). In general, this study
suggests that it is critical to nd a balance between global and local Local product features Western ad model Western product features Western ad model
advertising appeals, particularly when dealing with products with
strong country stereotypes.
In practice, although marketers and advertisers may prefer to use
global or foreign appeals as important components of their branding
and marketing communications in emerging markets, they should
pay special attention to potential conicts with national, ethnic and
local identity sentiments (Ozsomer, 2012; Zhang & Khare, 2009).
Incongruent cultural identities in advertising images, such as the use
of a Western model to endorse a typical Chinese product, can result
in unfavorable responses due to consumer sentiment or deeply rooted Appendix 2. Stimulus advertisements used in Study 2
cultural meanings associated with local iconness (Steenkamp et al.,
2003). A key contribution of this study is to highlight the impact of
incongruent advertising appeals on consumer attitudes as they are re-
lated to advertising and marketing communications in a globalizing
marketplace.
There are several limitations that should be addressed in future
research. First, the subjects were only exposed to the designed advertise-
ments once in the study. People may question whether the same results
would occur following multiple exposures. In the external marketplace,
consumers are frequently exposed to certain advertisements and
perhaps the perceived incongruence decreases when consumers have
become more familiar with the advertising appeals. Therefore,
researchers may want to explore how advertising image incongruence
is perceived when consumers are exposed to the advertisements several Congruent advertising image Incongruent advertising image
times. Second, future research may also examine the phenomenon by (Local ad model features local brand) (Foreign ad model features local brand)
using other kinds of cultural identities pertaining to advertising images,
such as global or foreign symbols and cultural elements (instead of the
model ethnicity used in this study). A better understanding of conicting
forces surrounding global and local consumer cultures requires a broader References
view of the cultural elements used in advertising communications. Third,
Ahluwalia, R., & Burnkrant, R. E. (2004). Answering questions about questions: A persua-
while this study carefully controlled for the potential confounding effects
sion knowledge perspective for understanding the effects of rhetorical questions.
such as attractiveness of the advertisements and likability of the model Journal of Consumer Research, 31(1), 2641.
types, the inuence of other factors might not be completely ruled out. Alden, D. L., Steenkamp, J. B. E. M., & Batra, R. (1999). Brand positioning through advertis-
ing in Asia, North America, and Europe: The role of global consumer culture. Journal of
For example, involvement was not considered, but it may relate to
Marketing, 63(1), 7587.
individuals' use of persuasion knowledge. Future studies might take Arnett, J. J. (2002). The psychology of globalization. American Psychologist, 57(10),
high involvement products into account to investigate whether product 774783.
involvement may moderate the effects of ad image incongruence on Batra, R., & Stephens, D. (1994). Attitudinal effects of ad-evoked moods and emotions:
The moderating role of motivations. Psychology & Marketing, 11(3), 199215.
individuals' use of persuasion knowledge. Campbell, M. C. (1995). When attention-getting advertising tactics elicit consumer infer-
Finally, the varying levels of image congruence were not included in ences of manipulative intent: The importance of balancing benets and investments.
the experimental designs. Future research may be directed toward an Journal of Consumer Psychology, 4(3), 225254.
Campbell, M. C., & Kirmani, A. (2000). Consumers' use of persuasion knowledge: The
understanding of how different levels of culturally based ad image effects of accessibility and cognitive capacity on perceptions of an inuence agent.
incongruity may affect consumers' information processing style of the Journal of Consumer Research, 27(1), 6982.
advertising content. This research direction may require more complex Cheong, Y., Zheng, L., & Kim, K. (2011). Product global reach, advertising standardization,
and cultural values: An analysis of 2008 Beijing Olympic TV commercials. Asian
experimental designs and manipulations of the variables of interest. Journal of Communication, 21(3), 279300.
Nevertheless, such effort could provide additional new insights to the Cotte, J., Coulter, R. A., & Moore, M. (2005). Enhancing or disrupting guilt: The role of ad
extant literature. credibility and perceived manipulative intent. Journal of Business Research, 58,
361368.
Dahlen, M., & Lange, F. (2004). To challenge or not to challenge: Ad-brand incongruency
and brand familiarity. Journal of Marketing Theory and Practice, 12(3), 2035.
Acknowledgment Davis, D. S. (2000). The consumer evolution in urban China. University of California Press.
Eckhardt, G. M. (2005). Local branding in a foreign product category in an emerging
market. Journal of International Marketing, 13(4), 5779.
This research was supported through a grant from the Social Freidstad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope
Sciences and Humanities Research Council (SSHRC) (410-2009-1385). with persuasion attempts. Journal of Consumer Research, 21, 131.
The authors are grateful to the editor and two anonymous reviewers for Freidstad, M., & Wright, P. (1995). Persuasion knowledge: Lay people's and researchers'
beliefs about the psychology of advertising. Journal of Consumer Research, 22, 6278.
insightful guidance. The authors thank Yan Wu for helping them with Friedman, H., & Friedman, L. (1979, October/November). Endorser effectiveness by prod-
the eldwork. uct type. Journal of Advertising Research, 19, 6371.
568 L. Zhou et al. / Journal of Business Research 68 (2015) 561568
Heckler, S. E., & Childers, T. L. (1992). The role of expectancy and relevancy in memory for
verbal and visual information: What is incongruency? Journal of Consumer Research,
18(4), 475492.
Hung, K. H., Li, S. Y., & Belk, R. W. (2007). Glocal understandings: Female readers' percep-
tions of the new woman in Chinese advertising. Journal of International Business
Studies, 38, 10341051.
Kahle, L. R., & Homer, P. M. (1985, March). Physical attractiveness of the celebrity endors-
er: A social adaptation perspective. Journal of Consumer Research, 11, 954961.
Kamins, M. A. (1990, March). An investigation into the match-up hypothesis in celebrity
advertising: When beauty may be only skin deep. Journal of Advertising, 19, 413.
Laroche, M., Papadopoulos, N., Heslop, L. A., & Mourali, M. (2005). The inuence of coun-
try image structure on consumer evaluations of foreign products. International
Marketing Review, 22(1), 96115.
Lee, Y. H. (2000). Manipulating ad message involvement through information expectan-
cy: Effects on attitude evaluation and condence. Journal of Advertising, 29(2), 1224.
Lee, Y. H., & Mason, C. (1999). Responses to information incongruency in advertising: The
role of expectancy, relevancy, and humor. Journal of Consumer Research, 26, 156167.
Lee, J. G., & Thorson, E. (2008). The impact of celebrityproduct incongruence on the ef-
fectiveness of product endorsement. Journal of Advertising Research, 48(3), 433449.
Lynch, J., & Schuler, D. (1994). The matchup effect of spokesperson and product congruency:
A schema theory interpretation. Psychology & Marketing, 11(5), 417445.
Martin, B. A., Lee, K. C., & Yang, C. (2004). The inuence of ad ethnicity model and self-
referencing on ad attitudes: Evidence from New Zealand. Journal of Advertising,
33(4), 2737.
Menon, S., & Kahn, B. E. (2003). Corporate sponsorships of philanthropic activities: When
do they impact perception of sponsor brand? Journal of Consumer Psychology, 13(3),
316327.
Misra, S., & Beatty, S. E. (1990). Celebrity spokesperson and brand congruence: An assess-
ment of recall and affect. Journal of Business Research, 21(2), 159173.
Moore, J. J., & Rodgers, S. L. (2005, January 1). An examination of advertising credibility
and scepticism in ve different media using the persuasion knowledge model.
American Academy of Advertising Conference Proceedings (pp. 10).
Neelankavil, J. P., Venkatapparao, M., & Sessions, D. N. (1994). Use of foreign language and
models in print advertisements in East Asian countries: A logit modelling approach.
European Journal of Marketing, 29(4), 2438.
Nelson, M. R., & Paek, H. J. (2007). A content analysis of advertising in a global magazine
across seven countries: Implications for global advertising strategies. International
Marketing Review, 24(1), 6486.
Obermiller, C., & Spangerberg, E. R. (1998). Development of a scale to measure consumer
skepticism toward advertising. Journal of Consumer Psychology, 7(2), 159186.
Ozsomer, A. (2012). The interplay between global and local brands: A closer look at per-
ceived brand globalness and local iconness. Journal of International Marketing, 20(2),
7295.
Scott, L. M. (1994). The bridge from text to mind: Adapting reader-response theory to
consumer research. Journal of Consumer Research, 21(3), 461480.
Steenkamp, J. -B. E. M., Batra, R., & Alden, D. L. (2003). How perceived brand globalness
creates brand value. Journal of International Business Studies, 34(1), 5365.
Tai, H. C., & Pae, J. H. (2002). Effects of TV advertising on Chinese consumers: Local versus
foreign-sourced commercials. Journal of Marketing Management, 18(12), 4972.
Tse, D. K., Belk, R. W., & Zhou, N. (1989). Becoming a consumer society: A longitudinal and
cross-cultural content analysis of print ads from Hong Kong, the People's Republic of
China, and Taiwan15(4), 457472.
Wright, P., Friestad, M., & Boush, D. M. (2005). The development of marketplace persua-
sion knowledge in children, adolescents, and young adults. Journal of Public Policy &
Marketing, 24(2), 222233.
Zhang, J. (2009). The effect of advertising appeals in activating self-construals: A case of
bicultural Chinese generation consumers. Journal of Advertising, 38(1), 6382.
Zhang, Y., & Khare, A. (2009). The impact of accessible identities on the evaluation of glob-
al versus local products. Journal of Consumer Research, 36(3), 524537.
Zhou, N., & Belk, R. W. (2004). Chinese consumer readings of global and local advertising
appeals. Journal of Advertising, 33(3), 6376.
Zhou, L., Yang, Z., & Hui, M. (2010). Non-local or local brands? A multi-level investigation
into condence in brand origin identication and its strategic implications. Journal of
the Academy of Marketing Science, 38(2), 202218.