Consumer Ethnocentrism:

A Test of Antecedents
and Moderators

Subhash Sharma
Terence A. Shimp
University of South Carolina

Jeongshin Shin
Chunnam National University, Korea

This article identifies theoretical antecedents of consumer triotic prejudices against imports--and generally have
ethnocentricity and the effect ethnocentricity has on evalu- found that consumers tend to evaluate domestic products
ations toward importing products. Hypotheses pertaining unreasonably favorably vis-a-vis imported products
to the relationship between the identified antecedents and (cf. Bannister and Saunders 1978; Bilkey and Nes 1982;
consumer ethnocentricity are developed based on an ex- Cattin, Jolibert, and Lohnes 1982; Chasin, Holzmuller, and
tensive review of the ethnocentrism and country-of-origin Jaffe 1988; Darling and Kraft 1977; Han 1988; Hung 1989;
literatures. Also identified are factors moderating the ef- Johansson, Douglas, and Nonaka 1985; Morello 1984;
fect of ethnocentric tendencies on consumers' attitudes Nagashima 1970; Narayana 1981; Reierson 1966;
toward importing products. The hypotheses are subjected Schooler 1965; Wall and Heslop 1986; Wang 1978; White
to an empirical test using data collected in Korea. 1979)--but these studies have not examined why and
under what conditions these effects exist.
The present research examines the process underlying
consumers' attitudes toward products being imported into
Patriotic, ethnocentric, nationalistic, and even xeno- their domestic economies. With consumer ethnocentrism
(Shimp and Sharma 1987) as the focal construct, this
phobic sentiments wax and wane among Americans and
article seeks to accomplish three objectives: first, it expli-
people of other countries. Periodic claims of unfair com-
cates the nature and role of ethnocentrism and ties this
petition, uneven playing fields, closed markets, and exces- construct into a framework of related constructs. Next, it
sive foreign influence cross international boundaries and specifies the factors that moderate the effect of ethnocen-
inflame business people and consumers alike. Consumers tric tendencies on consumers' attitudes toward importing
are made to feel guilty for purchasing imported products, products. Finally, hypotheses are subjected to an empirical
and patriotic advertisements remind us of our duty to test using data collected in Korea.
choose domestic-made goods over imports. Consequently,
a study of the factors affecting consumers' choice of do-
mestic products vis-a-vis imports would be an important ETHNOCENTRISM AND
step toward better understanding international exchange CONSUMER ETHNOCENTRIClTY
relations and domestic marketplace behavior. Numerous
country-of-origin studies have touched on consumers' pa- The phenomenon of consumer preference for domestic
products, or prejudice against imports, has been termed
Journal of the Academy of Marketing Science. economic nationalism, cultural bias against imports, or
Volume 23, No. 1, pages 26-37. consumer ethnocentrism. All these terms find their origin
Copyright 9 1995 by Academy of Marketing Science. in the general concept of ethnocentrism.
 Sharmaet al. / CONSUMERETHNOCENTRISM27

Ethnocentrism appropriate, desirable, and patriotic; buying them is bad,

inappropriate, undesirable, and irresponsible. Third, it re-
Ethnocentrism was introduced to the sociological lit- fers to a personal level of prejudice against imports, al-
erature nearly a century ago. Sumner (1906) defined eth- though it may be assumed that the overall level of
nocentrism as consumer ethnocentricity in a social system is the aggre-
gation of individual tendencies.
the view of things in which one's own group is the The consequences of consumer ethnocentricity include
center of everything, and all others are scaled and overestimation of domestic products or underestimation of
rated with reference to it . . . . Each group nourishes imports, a moral obligation to buy domestic products, and
its own pride and vanity, boasts itself superior, exalts preference for domestic products. For example, due to
its own divinities and looks with contempt on out- World War II hostilities, some older Americans may refuse
siders. (p. 13) to buy Japanese- or German-made products and some
Japanese and Germans probably eschew American-made
Ethnocentrism is a universal phenomenon and is rooted goods for the same reason. Even after eliminating extreme
deeply in most areas of intergroup relations (Lewis 1976). cases, ordinary consumers may still want to pay more to
Murdock (1931) observed that ethnocentrism is confined buy domestic products that are undifferentiated from im-
not only to tribes and nations but reveals itself in all kinds ports (Wall and Heslop 1986).
of social groups, developing into family pride, sectional- It is important to emphasize that consumer ethnocen-
ism, religious prejudice, racial discrimination, and patriot- trism was conceptualized by Shimp and Sharma (1987) as
ism. Some authors even argue that ethnocentrism is a part a trait-like property of individuals' personalities. As such,
of human nature (Lynn 1976; Mihalyi 1984; Rushton one's consumer-ethnocentric tendencies may be expected
1989). under certain circumstances to influence attitudes and be-
Ethnocentrism functions by helping to secure the sur- haviors toward foreign products and, reciprocally, toward
vival of groups and their cultures and by increasing a competitive domestic products. Hence, in this view, ethno-
group's solidarity, conformity, cooperation, loyalty, and centric tendencies represent an antecedent to attitudes but
effectiveness (Catton 1960; Lynn 1976; Mihalyi 1984; certainly are not tantamount to attitudes. This is demon-
Murdock 1931; Rosenblatt 1964; Sumner 1906). More strated empirically in the Results section.
specific properties of ethnocentrism include the tendency
(1) to distinguish various groups; (2) to perceive events in
terms of the group's own interests (economical, political,
CONCEPTUAL FRAMEWORK
and social); (3) to see one's own group as the center of the
AND HYPOTHESES
universe and to regard its way of life as superior to all
others; (4) to be suspicious of and disdain other groups;
(5) to view one's own group as superior, strong, and hon- Ethnocentric tendencies in consumers do not develop
est; (6) and to see other groups as inferior, weak, and in isolation but rather are part of a constellation of social-
dishonest troublemakers (LeVine and Campbell 1972). psychological and demographic influences. Figure 1 pre-
sents consumer ethnocentricity as the focal construct and
relates it to demographic variables and, more interestingly,
Consumer Ethnocentricity
to other social-psychological constructs: openness to for-
eign cultures, patriotism, collectivism-individualism, and
Consumer ethnocentricity, a uniquely economic form
of ethnocentrism, was formulated as a domain-specific conservatism. Consumer ethnocentricity is shown, in turn,
concept for the study of consumer behavior with marketing to be causally determinant of consumers' attitudes toward
implications (Shimp and Sharma 1987). This article adopts importing foreign products into their domestic economies.
Shimp and Sharma's (1987) definition of consumer ethno- This key theoretical linkage is not expected to be uniformly
centricity: "the beliefs held by consumers about the appro- strong, however, and must be qualified by pertinent mod-
priateness, indeed morality, of purchasing foreign-made erating circumstances.
products." (p. 280) Specifically, consumer ethnocentricity Two moderating factors are postulated. First, it is ex-
has the following characteristics: first, it results from the pected that consumer ethnocentric tendencies will have an
love and concern for one's own country and the fear of especially strong effect on attitudes toward importing
losing control of one's economic interests as the result of products that are perceived as relatively nonnecessary
the harmful effects that imports may bring to oneself and (versus bare necessities). Second, the strength of the rela-
countrymen. Second, it contains the intention or willing- tionship should vary as a function of perceived economic
ness not to purchase foreign products. For highly ethno- threats: Consumer ethnocentricity is expected to have a
centric consumers, buying foreign products is not only an strong impact on attitudes toward importing products
economic issue but also a moral problem. This involve- when consumers perceive that such importation poses
ment of morality causes consumers to purchase domestic threats to their personal welfare and/or the welfare of the
products even though, in extreme cases, the quality is domestic economy. The following sections develop the
below that of imports. Not buying foreign imports is good, constructs in Figure 1 and present hypotheses.
28 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE WINTER 1995

Figure 1
CONCEPTUAL MODEL

Antecedents Consumer
Attitudes Toward
- Openness to Ethnocentric
Importing Products
Foreign Cultures Tendencies
- Patriotism
- Conservatism
- Collectivism/
Individualism

Demographics Moderating Factors

- Age - Perceived Product Necessity
- Gender - Economic Threat
- Education -Personal Economic Threat
- income -Domestic Economic Threat

Social-Psychological Factors expected that patriotic individuals will show more con-
sumer-ethnocentfic tendencies than individuals who are
Cultural openness. Individuals differ in terms of their less patriotic.
experience with and openness toward the people, values,
and artifacts of other cultures. The opportunity to interact H2: Apositive correlation is expected between patriotism
with other cultures can have the effect of reducing cultural and consumer ethnocentric tendencies.
prejudice. Cultural openness may explain why Howard
(1989) found that U.S. residents on the West Coast--a
region where numerous cultures interact--tend to rate Conservatism. It is generally understood that conserva-
imported products more favorably than domestic products, tive persons show a tendency to cherish traditions and
whereas residents of the Midwest, which is less heteroge- social institutions that have survived the test of time, and
neous in its cultural composition, tend to rate imported to introduce changes only occasionally, reluctantly, and
products much lower in quality. Shimp and Sharma (1987) gradually. The extreme conservative has the following
detected that respondents from Los Angeles were less characteristics: religious f u n d a m e n t a l i s m , pro-
ethnocentric than were those from Denver, Detroit, and the establishment orientation, insistence on strict rules and
Carolinas. punishments, preference for the conventional, and an
antihedonic outlook (Wilson and Patterson 1968). Coun-
try-of-origin researchers (Anderson and Cunningham 1972;
HI: A negative correlation is expected between cultural
Wang 1978) have found an inverse relationship between
openess and consumer ethnocentrism. That is, con-
conservatism and attitudes toward foreign products; that
sumers who are more familiar with and open to
is, conservative people tend to evaluate imports negatively
foreign cultures should show less consumer ethno-
and domestic products positively.
centric tendencies than those who are less open.
H3: Apositive correlation is expected between conserva-
Patriotism. Patriotism represents love for or devotion tism and consumer ethnocentric tendencies.
to one's country. Patriotism is related to ethnocentrism
(Adorno et al. 1950; Catton 1960; Sumner 1906) and
works as a kind of defense mechanism for the group Collectivism-Individualism. This cross-culturally vali-
(Mihalyi 1984). Moore (1989) observed that some non- dated construct is regarded as one of the most promising
tariffbarriers are more reflective of patriotism than protec- dimensions of cultural variation (Triandis, Bontempo, and
tionism. Country-of-origin studies have implicated Villarel 1988). Essential attributes of collectivist cultures
patriotic emotions in the purchase of imported products are (1) that individuals have the tendency to subordinate
and country stereotyping (Crawford and Lamb 1981; Han their personal goals to the goals of the group, (2) that the
1988; Hooley, Shipley, and Krieger 1988; Howard 1989; self is subordinated to the group, and (3) that the group
Morello 1984; Wang 1978). Han (1988) showed that con- works as a source of one's identity. In contrast, people in
sumer patriotism has a significant effect on intentions to individualistic cultures tend to behave as if society is only
purchase domestic versus foreign products. Hence it is a means to their ends (Hui and Triandis 1986).
 Sharmaet al. / CONSUMERETHNOCENTRISM29

Collectivistic persons are likely to evince strong H6: Women are expected to exhibit greater consumer-
consumer-ethnocentric tendencies because they tend to ethnocentric tendencies than men.
consider the effect of their behavior on society, feel respon-
Education. More educated people tend to be less con-
sible for others, and are more susceptible to social influ-
servative (Ray 1990). They are less likely to have ethnic
ence against imports (cf. Triandis, Brislin, and Hui 1988);
prejudices (Watson and Johnson 1972) and to have pride
individualistic persons will have less consumer-
in their country (Rose 1985), and are more likely to favor-
ethnocentric tendencies because they act primarily for their
ably evaluate imports and to unfavorably evaluate domes-
own benefit. tic products (Anderson and Cunningham 1972; Wall and
Heslop 1986; Wang 1978).
H4:The correlation between collectitivism and consumer
ethnocentrism is expected to be positive; that is, H7: The correlation between the level of educational
people with collectivistic goals will reveal more achievement and consumer ethnocentrism is ex-
intensive ethnocentric tendencies than those with pected to be negative.
individualistic goals.
Income. Country-of-origin studies have found that
high-income consumers tend to evaluate foreign products
Demographic Variables
more favorably (Wall and Heslop 1986; Wang I978). In
Consumer ethnocentricity is expected to covary with general, as one's income increases, one may tend to travel
age, gender, educational level, and income based on the abroad and try more products, which may result in more
rationales that follow. It is noteworthy that these demo- cosmopolitan views and greater openness to foreign
graphic characteristics are not conceptually independent of products.
the social-psychological constructs just discussed. For ex-
ample, older individuals are, on average, more conserva- 1-18:The correlation between income level and consumer-
tive. It n o n e t h e l e s s is i m p o r t a n t to study these ethnocentric tendencies is expected to be negative.
demographic variables due to their greater implications for
actionable marketing practice. Moderating Factors
Age. Older people are generally more conservative, Perceivedproduct necessity. Consumers perceive prod-
more patriotic (Han 1988), and more likely to have expe- ucts differently in terms of whether those products are
rienced conflicts with foreign countries. Some researchers necessary to them personally or to the domestic economy.
have found that attitudes toward domestic products gener- When a product is perceived as a necessity, consumer
ally become more favorable with increasing age (Bannister ethnocentricity should play a relatively minor role in af-
and Sannders 1978; Schooler 1971; Tongberg 1972) and fecting attitudes toward importing that product. On the
have implied that the younger generation may be more other hand, for items perceived as dispensable, consumer
cosmopolitan in their preferences and attitudes and conse- ethnocentricity should have a more substantial impact on
quently more favorably inclined toward imports (Bannister attitudes. On logical grounds we expect that personal pref-
and Saunders 1978). Rose (1985) suggested that elderly erences and desires (or, as constituted here, product neces-
people who have known the ravages of military defeat and sity) counteract the more altruistic and non-self-centered
occupation, such as those in Japan, France, and Germany, motives contained in the ethnocentric ideal.
may feel more pride toward their country. However, some
research has reported the conflicting finding that older H9: The impact that consumer-ethnocentric tendencies
people tend to evaluate foreign products more favorably have on attitudes toward importing products is
than domestic goods (Han 1988; Schooler 1971; Tongberg expected to be moderated by the perceived neces-
1972). But overall, there seems to be more support for the sity of the imported item. Specifically, the effect of
idea that older people show more consumer-ethnocentric consumer ethnocentricity on attitudes should be
tendencies than younger people. relatively stronger for products perceived as un-
necessary.
1-15: Age and consumer-ethnocentric tendencies are ex- Perceived economic threat. Another moderating factor
pected to be positively correlated. relates to consumers' concerns about the threat that foreign
competitors pose to them personally and/or to the domestic
Gender. Females are more conservative, more conform- economy. When any country considers itself under attack
ist (Eagly 1978), more patriotic (Han 1988), more con- or threatened by competition from outsiders, "foreign-
cerned about preserving social harmony and promoting ness" takes on negative meanings (Polhemus 1988) and
positive feelings among group members, and less indi- nationalism and ethnocentrism increase (Rosenblatt
vidualistic (Triandis et al. 1985). Wall and Heslop (1986) 1964). The fear of losing jobs (either one's own or a related
reported that Canadian women are more positive than men person's) may influence consumers' reactions to imports.
toward the quality of Canadian-made products, whereas Sherif et al. (1961) found that when two groups com-
Howard (1989) observed that American women rated do- pete against each other, between-group hostility increased.
mestic products more favorably than did men. Thus, However, ingroup solidarity and cooperativeness also in-
30 JOURNALOF THE ACADEMYOF MARKETINGSCIENCE WINTER 1995

creased when the two groups were competing, and were Measures of Social-
highest when intergroup hostility was at its peak. In re- Psychological Constructs
search more directly related to the marketplace, it has been
detected that people who are involved in threatened indus- Previously used scales for measuring collectivism-
tries, such as automobiles or textiles, show higher con- individualism, conservatism, consumer-ethnocentric ten-
sumer-ethnocentric tendencies (Daser and Meriz 1987; dencies, and patriotism were identified. Items for each
Shimp and Sharma 1987). scale first were analyzed for cultural compatibility, and,
Based on this evidence, we propose that consumer where appropriate, were deleted or modified. All items
ethnocentricity will have a particularly strong impact on were measured on 7-point Likert-type scales.
attitudes toward importing products when consumers feel The scale for collectivism-individualism was proposed
either personally threatened or that the domestic economy and validated both in America and in Hong Kong by Hui
is threatened even though they may not be threatened (1988). The original scale contains six subscales that indi-
personally. vidually measure collectivism toward one's spouse, par-
ents, relatives, neighbors, friends, and coworkers. For this
H10:The impact that consumer-ethnocentric tendencies research, two subscales were adopted that have been
have on attitudes toward importing products is shown to be rather distant on a social-closeness contin-
moderated by the perceived threat of foreign com- uum: collectivism toward parents (6 items) and toward
petition. Consumer ethnocentricity should coworkers (5 items). Illustrative items measuring parental-
strongly influence attitudes primarily for products based collectivism/individualism include, "Young people
perceived to personally threaten the individual or should take into consideration their parents' advice when
the domestic economy. making education/career plans" and "It is reasonable for a
son to continue his father's business." Similar illustrative
items for the coworker measure include, "One needs to
METHOD return a favor if a colleague lends a helping hand" and
"There is everything to gain and nothing to lose for class-
The framework in Figure 1 and the corresponding hy- mates to group themselves for study and discussion."
potheses were investigated by studying Korean consum- Ray's (1983) conservatism measure was used in assess-
ers. Two data collection procedures were employed: one ing American public opinion. Illustrative items from the
involved mailing questionnaires to randomly selected 7-item scale include "Erotic and obscene literature should
households, and the other entailed distributing question- be prohibited from public sale" and "The government
naires through schoolchildren to their parents (specifics should make sure that our armed forces are stronger than
follow). Respondents were asked to reveal their attitudes those of North Korea at all times."
concerning whether a variety of different products should The patriotism scale was taken partially from Adorno
or should not be imported into Korea, to respond to the et al.'s (1950) scale, which was adapted to be compatible
multiple items that operationalized the constructs contained with Korean society. Two illustrative items from the 8-item
in Figure 1, and to provide demographic information. scale are "Devoting oneself for one's country is worth-
while" and "Patriotism and loyalty are the first and most
Products Investigated important requirements of a good citizen."
The c o n s u m e r - e t h n o c e n t r i c t e n d e n c y scale
It was essential to select product categories that varied (CETSCALE), proposed and validated in the United States
in perceived necessity to the lives of Korean consumers. by Shimp and Sharma (1987), operationalized this con-
Toward this end, in-depth discussions were undertaken in struct. The scale has been further validated in West Ger-
the United States with 42 Korean students and their many, France, and Japan by Netemeyer, Durvasula, and
spouses. The objective was to identify various products Lichtenstein (1991). The CETSCALE consists of 17 items,
consumed in Korea that differ in perceived necessity for including "Korean people should always buy Korean-
day-to-day living. Ten products differing in their level of made products instead of imports"; "Only those products
necessity were identified. Four of these were considered that are unavailable in Korea should be imported"; and
most necessary (medicine, beef, personal computers, and "Korean products, first, last, and foremost."
kitchenware), whereas six were considered relatively un- The unavailability of a scale to measure cultural open-
necessary (golf clubs, liquor, bananas, insurance, large ness necessitated developing an ad hoc, 7-item Likert-type
refrigerators, and jewelry). In the questionnaire distributed scale. Illustrative items include "I would like to have
in Korea, the 10 products identified as varying with respect opportunities to meet people from different countries" and
to daily-living necessity were measured on 5-point scales "I am very interested in trying food from different coun-
(1 = definitely necessary, 5 = definitely unnecessary). tries."
Consumers' attitudes toward importing each of the 10 An original scale also had to be constructed to measure
products, the dependent variable, were measured on 5- the moderating factor, perceived economic threat. This
point scales (1 = definitely should import, 5 = definitely 7-item ad hoc measure consisted of the following illustra-
should not import). tive items: "The present recession is due to the extensive
amount of foreign competition"; "The security of my
job/business is heavily influenced by foreign competi-
 Sharma et al. / CONSUMERETHNOCENTRISM 31

tors"; and "Economic problems are mainly due to exces- TABLE 1

sive foreign competitors." Sample Characteristics
Variable Total sample
Questionnaire
Demographics
Age M = 42.49
Translation and back translation. The questionnaire Gendera
was translated into Korean by a Korean doctoral student Male 56.9%
fluent in English, and back translated into English by Female 43.1%
another Korean doctoral student who was majoring in Income M = $35,194
English. The back-translated questionpaire was then Education (years) M = 13.24
judged by four Ph.D. students (three Americans and a Constructs
Korean) for meaning compatibility. This process of trans- Collectivism
Coworkerb M = 16.15
lation, back translation and judging for meaning compati-
Parentse M = 14.90
bility was repeated until all judges considered the Korean
Economicthreat
questionnaire to adequately represent the English version Personald M = 11.60
on which it was based. Domestice M = 13.12
Opennesse M = 35.47
Pretest. The final questionnaire, in Korean, was pre- Patriofismg M = 36.90
tested by administering it in the United States to a sample Conservatismh M = 22.57
of Korean students and their spouses. Respondents were CETSCALEi M = 85.07
asked to designate any ambiguities or awkwardness in
a. Percent of males and females in the sample.
question wording. Fifty-three questionnaires were col- b. Range = 3-21; higher scores ~ greater coworker-basedcollectivism.
lected. Based on the pretest results, several items were c. Range = 3-21; higher scores ~ greater parental-based collectivism.
refined or deleted due to ambiguities and/or possible cul- d. Range = 3-21; higher scores ~ higher personal threat.
tural incompatibility. e. Range = 3-21; higher scores ~ higher domestic economicthreat.
f. Range = 7-49; higher scores ~ greater cultural openess.
Sampling Methods in Korea g. Range = 7-49; higher scores ~ greater patriotism.
h. Range = 4-28; higher scores ~ greater conservatism.
i. Range = 17-119;higher scores ~ greater ethnocentrictendencies.
As noted, two sampling methods were used in order to
obtain representative coverage of Korean consumers. One
method (the mail sample) was a stratified sample of 1,500
significantly different with respect to demographic vari-
names and addresses purchased from a mailing list com-
ables or in their response to the dependent variables. Ac-
pany's database of 5 million economically diverse house-
cordingly, the two samples have been pooled for testing
holds. Questionnaires with an introduction letter and a
proposed relationships.
postage-paid return envelope (bearing the return address
of a local contact person) were mailed by the mailing list
company to the 1,500 members of the sample.
The second method (the school sample) involved dis- RESULTS
tributing 700 questionnaires to students at two elementary
schools, three middle schools, and a high school in the Table 1 presents demographic characteristics and mean
Seoul metropolitan area and in a southern city. Students responses to the key measures for the pooled sample of 667
with an odd-ending class number were instructed to ask Korean respondents. Two of the averages are particularly
their father to complete the questionnaire, whereas those noteworthy. First, the mean C E T S C A L E score of 85.07 is
with an even-ending number were to ask their mother to considerably higher (about one scale point per item) than
complete the questionnaire. This procedure ensured about scores obtained by Shimp and Sharma (1987) in the United
an equal representation of male and female respondents. States, which ranged from a high of 68.58 in Detroit to a
For the mail sample, 125 questionnaires were com- low of 56.62 in Los Angeles. Korean consumers either are
pleted and 127 questionnaires were returned as undeliver- more ethnocentric than their American counterparts or
able. This gives an effective response rate of slightly more respond to scales differently. Second, the average income
than 9 percent. For the school sample, 542 questionnaires of the sample is considerably higher than the average
were completed, giving a response rate of about 77 percent. income of Korean households. This occurred because
The low response rate for the mail sample was expected about 10 percent of the mail sample had extremely high
due to the nature of the study, length of the questionnaire, incomes. To assess the effect of including high-income
and the absence of a follow-up mailing due to budget respondents in the sample, all the analyses were repeated
constraints. The response-rate differential between the by excluding the high-income respondents. Because there
mail and school samples is not problematic, however, were no appreciable differences in the results, findings
because analyses revealed that the two samples are not reported hereafter are based on all 667 respondents.
32 JOURNALOF THE ACADEMYOF MARKETINGSCIENCE WINTER 1995

Measurement Properties TABLE 2

Confirmatory Factor Analysis Results
The sample was randomly split into analysis (n = 334) Numberof
and holdout (n = 333) samples. An exploratory factor Construct (Scale) Indicators Reliability
analysis on the analysis sample first assessed dimension-
ality of the constructs and identified unreliable items. Collectivism
Next, confirmatory factor analysis on the holdout sample Coworkers 3 .39
assessed measurement properties of the various scales. Parents 3 .56
EconomicThreat
Personal 3 .67
Exploratory factor analysis. Based on the eigenvalue- Domestic 3 .53
greater-than-one rule and the scree plot, ,the exploratory Openness 7 .74
factor analyses resulted, as expected, in one factor for the PATCON 11 .7l
CETSCALE, conservatism, patriotism, and openness CETSCALE 17 .91
scales, and two factors for the collectivism scale (i.e.,
collectivism toward parents and coworkers) and two fac- NOTE: Chi-square = 2271.03, df= 1013, p = .000; GFI = .834, NCP =
2.378, RNI = .817, TLI = .805.
tors for the economic threat scale (i.e., personal and do-
mestic economic threat). Several items were deleted as
either having low loadings or weak correlations with other NCP, RNI, and TLI because, according to the results of
statements. Deleted items included two for collectivism simulation studies, only these indices are unbiased in finite
toward coworkers, three for collectivism toward parents, samples. Although there are no established cutoff values
three for conservatism, one for patriotism, and one for for these indices, the values of the fit indices reported in
domestic economic threat. Table 2 are in line with those obtained by other researchers
for models with large numbers of indicators (e.g., Kohli,
Confirmatory factor analysis. Using the holdout sam- Jaworski, and Kumar 1993). Also, an examination of the
ple, a confirmatory factor analysis was performed with all
histogram for the residuals and the Q plot indicate that the
the constructs included in the model and correlated among
distribution of the residuals is approximately normal.
themselves. This approach is one of the many approaches
These goodness-of-fit statistics suggest a reasonable fit of
suggested for assessing scale measurement properties
the model to the data, thereby demonstrating reliable and
(cf. Anderson and Gerbing 1988; Kumar, Stern, and
unidimensional measures.
Achrol 1992). The underlying rationale is that measure-
The convergent validity of the scales was assessed by
ment properties should be assessed in the context of the
examining the loadings' t values--all of which are signifi-
theoretical model in which concepts are embedded. The
cant at an alpha of 0.05 or lower--thereby establishing
analysis on the holdout sample was repeated for the total
measurement convergence (Bagozzi 1981). Discriminant
sample. Because there are no substantial differences in the
validity was assessed by constraining the correlation of
results between the holdout and total samples, subsequent
each pair of constructs to one and examining the decrease
results are for the total sample only.
The conservatism and patriotism scales are highly cor- in fit via chi-square difference tests. The difference tests
related (phi = .98). A chi-square difference test indicates for all pairwise analyses are statistically significant, which
that this correlation is not significantly different from one indicates that the pairwise correlations of constructs are
suggesting that, although the two constructs are theoreti- statistically different from one; furthermore, the shared
cally different, they cannot be empirically separated. For variance between any two constructs is less than 0.50, thus
example, the following statement from the conservatism establishing discriminant validity for all constructs.
scale--"People who show disrespect for their country's The reliabilities of the scales range from .39 to .91. The
flag should be punished for it"---clearly shows why the CETSCALE's reliability of.91 (see Table 2) is comparable
two constructs are highly correlated at the empirical level. to the reliabilities reported in Shimp and Sharma's (1987)
Consequently, the conservatism and patriotism scales are study of Americans and Netemeyer, Durvasula, and
combined into one scale and labeled PATCON. Lichtenstein's (1991) study of French, Germans, and Japa-
Table 2 gives the confirmatory factor analysis results nese. The remaining reliabilities for the various scales are:
along with the goodness-of-fit indices and the reliability cultural openness (.74), PATCON (.71), personal eco-
coefficients computed by employing the formula sug- nomic threat (.67), collectivism toward parents (.56), do-
gested by J6reskog (1971). The model does not statistically mestic national economic threat (.53), and collectivism
fit the data as indicated by the chi-square statistic. How- toward coworkers (.39). Some of these reliabilities are
ever, because the chi-square statistic is sensitive to sample obviously lower than desirable; however, the effect of
size, researchers have traditionally discounted this test as these measurement deficiencies renders conservative the
the sole indicator of model fit and have resorted to other subsequent hypothesis tests, inasmuch as unreliable mea-
goodness-of-fit statistics such as J6reskog and S6rbom's surement attenuates observed relations between constructs
(1986) GFI, the rescaled noncentrality parameter (NCP) (Cote and Buckley 1987, 1988).
and the rescaled noncentrality index (RNI) of McDonald Also assessed was the conceptual difference between
and Marsh (1990), and the Tucker-Lewis Index (TLI). the measure of consumer ethnocentrism (CETSCALE)
McDonald and Marsh (1990) recommend the use of the and attitudes toward importing specific products. An ex-
 Sharma et al. / CONSUMER ETHNOCENTRISM 33

TABLE 3 TABLE 4
Correlations with CETSCALE Means and Standard Deviations for
Product Necessity Scores
Factors r p level
Product Mean a SD
Social-psychologicalfactors
Openness -.205 .000 Medicine 1.94 0.99
PATCON .526 .000 Kitchenware 2.71 1.34
Collectivism Beef 2.83 1.10
Parents .182 .000 PC 2.87 1.23
Coworkers .227 .000 Jewelry 3.39 1.22
Demographic factors Large fridge 3.46 1.34
Age -.033 .209 Liquor 3.46 1.21
Gender .172 .000 Bananas 3.88 1.10
Education -.246 .000 Insurance 4.12 1.04
Income -. 146 .000 Golf clubs 4.31 1.06
a. Smaller numbers indicate greater perceived product necessity.
ploratory factor analysis (with varimax rotation) on the 17
C E T S C A L E items and the 10 attitude items revealed that TABLE 5
when the solution was constrained to just two factors, the Moderator Regression Results
17 C E T S C A L E items loaded on the first factor and 9 of for Product Necessity a
the 10 product-specific attitude items loaded on the second
factor. (The tenth attitude item loaded on neither factor.) A Product Necessity (N) CET (C) N C R2
confirmatory factor analysis suggests that the correlation Medicine .202b .098 .247 .234
between the two constructs (r = .568) is significantly .000c .002 .000
different from one, thereby establishing discriminant va- Kitchenware .057 .291 .208 .225
lidity between consumer ethnocentricity and attitudes to- .025 .000 .000
ward importing products. Beef .052 .151 .134 .073
.041 .000 .000
PC .225 .085 .224 .220
Hypothesis Tests
.000 .007 .000
Jewelry .028 .215 .126 .088
Social-psychological factors. Table 3 gives the correla- .233 .000 .000
tions b e t w e e n the C E T S C A L E and the four social- Large fridge .277 .044 .246 .269
psychological antecedents postulated in Figure 1. In sup- .000 .165 .000
port of hypothesis H I , the correlation between cultural Liquor .073 .283 .200 .217
.003 .000 .000
openness and consumer ethnocentricity is predictably
Bananas .035 .299 .206 .210
negative (r = -.205, p = .000), which indicates that indi- 91 5 2 .000 .000
viduals who are open to other cultures are less ethnocen- Insurance .193 .131 .246 .224
tric. T h e c o r r e l a t i o n b e t w e e n the c o m b i n e d p a t r i - .000 . .000 .000
otism/conservatism scale (PATCON) and C E T S C A L E is Golf clubs -.012 .364 .143 .188
significantly positive (r = .526, p = .000), which supports .575 .000 .000
the expectations presented in hypotheses H2 and H3. Fi-
a. The dependentvariable is attitudes toward importing each product into
nally, the positive coefficients for the two collectivism Korea.
subscales (i.e., parents: r = . 182, p = .000 and coworkers: b. Standardizedregression parameter.
r = .227, p = .000) provide support for hypothesis H4. c. p level.

D e m o g r a p h i c factors. Table 3 also gives the correla-

tions between the C E T S C A L E and the various demo- Moderating factors. Moderated regression analysis
graphic factors. As can be seen, all hypotheses, except the (Sharma, Durand, and Gur-Arie 1981) was used to test
one pertaining to age (i.e., H5) are supported. Specifically, hypotheses H9 and H 10. These hypotheses posited that the
the correlation between gender and consumer ethnocen- impact of consumer-ethnocentric tendencies on attitudes
trism is significantly positive (r = . 172,p = .000) in support toward importing products into Korea would be moderated
of hypothesis H6, which predicted that women (coded 1) by the perceived necessity of the imported item (H9) and
would exhibit greater consumer-ethnocentric tendencies the perceived degree of personal and economic threat
than men (coded 0). Hypotheses H7 and H8 are supported imposed by the imported item (H10). Latent root regres-
by the negative coefficients between consumer ethnocen- sion rather than ordinary least squares regression was used
trism and education and income, respectively; as postu- for identifying moderator variables because the inclusion
lated, consumer-ethnocentric tendencies decrease with of an interaction term creates multicollinearity, which
greater levels of educational achievement (r = -.246, p = could suppress the identification of moderator variables
.000) and with increasing income (r = -.146, p = .000). (Morris, Sherman, and Mansfield 1986).
34 JOURNALOF THE ACADEMYOF MARKETINGSCIENCE WINTER 1995

TABLE 6 positively correlated with collectivist tendencies and with

Moderator Regression Results for patriotic/conservative attitudes but negatively correlated
Personal and Domestic Economic Threat a with cultural openness, education, and income. The rela-
Threat(T) CET(C) T R2 tionship between ethnocentric tendencies and attitudes
toward imports is moderated by product necessity and the
Personal economicthreat -.149b .491 .140 .314 extent of threat these products are perceived as having on
.000c .000 .000 the consumer personally and on the domestic economy
Domesticeconomicthreat -.207 .543 .145 .325 more generally.
.000 .000 .000
a. The dependentvariableis a combinedwillingnessto import10 different Implications for Marketing Theory
products into Korea.
b. Standardizedregressionparameter. This research and the theoretical base on which it is
c. p level. conducted hold valuable implications for the growing lit-
erature on consumer-ethnocentric tendencies, which itself
Table 4 provides means and standard deviations for the is an important individual-level construct for the better
10 products that respondents evaluated in terms of whether understanding of country-of-origin dynamics. Testing hy-
each should or should not be imported into Korea. The 10 potheses developed from an American vantage point in a
products had been selected from a pretest as products that
culturally dissimilar country (i.e., Korea) is a positive
varied considerably with respect to how necessary each is
response to the recommendation that theory development
for Korean daily life. The table orders these products from
in marketing be universalized beyond a parochial Ameri-
most to least necessary for maintaining Korean daily life.
can perspective (Dholakia, Firat, and Bagozzi 1980).
Table 5 gives the regression results testing the moder-
The research contributes substantively by fleshing out
ating effect of product necessity. Separate regression
the specific theoretical meaning and role that consumer
analyses were conducted for all 10 products by including
ethnocentrism plays. Prior research has established the
three terms in the moderated latent root regressions: prod-
psychometric properties of the CETSCALE in the United
uct necessity, consumer-ethnocentric tendencies, and the
States (Shimp and Sharma 1987) and in France, Germany,
interaction between these variables. Attitude toward im-
and Japan (Netemeyer, Durvasula, and Lichtenstein 1991);
porting a given product into Korea is the dependent vari-
however, studies have not examined in detail the antece-
able in all analyses. Moderation is indicated when the
dents of consumer ethnocentricity or the moderating fac-
interaction term between consumer ethnocentrism and per-
tors that determine the effect of this construct on attitudes
ceived necessity is statistically significant. As can be seen,
toward importing particular products. Our research estab-
the standardized regression parameters for all 10 interac-
tion terms are large and statistically significant, thereby lishes that consumer-ethnocentric tendencies play an in-
creasingly influential role under two circumstances:
supporting hypothesis H9: the more a product is perceived
as unnecessary, the greater the impact ethnocentric tenden- (1) when products are perceived as relatively unnecessary,
cies have on attitudes toward importing the product into and (2) when consumers perceive themselves and/or the
Korea) domestic economy to be threatened by the importation of
To test the moderating effect of perceived threat, an a particular product
overall measure of perceived willingness to import prod- These findings support the contextualist argument
ucts was formed by summing the willingness-to-import (McGuire 1983), which holds that theoretical relationships
responses for the 10 individual products. 2 Table 6 presents generally are not true under all circumstances but depend
moderated-regression results for the separate threat mea- on the specifics of the situation. In this case, we have
sures: personal economic threat and domestic economic shown that the impact of consumer ethnocentricity on
threat. Both interaction terms are significant, which indi- attitudes toward importing products varies in amplitude
cates that as consumers perceive imports to be more threat- depending on perceived product necessity and threat.
ening to their personal or domestic economic welfare, Hence there are instances in which consumers' attitudes
ethnocentric tendencies have increased impact on consum- and purchase behaviors can be expected to be influenced
ers' resistance to importing products into Korea. That is, by their ethnocentric tendencies, but there are perhaps even
ethnocentric consumers are more willing to import if the more instances where they will not be affected.
imported item is perceived as less threatening to their
personal welfare or to the domestic economy. These results Managerial Implications
offer strong support for hypothesis H10.
This study holds several implications for marketing
practice. First, consumer ethnocentrism provides market-
DISCUSSION ing managers with a useful concept for understanding
consumers' reasons for buying domestic versus imported
This article examines antecedents of consumer- products, and especially why certain segments of consum-
ethnocentric tendencies and the impact these tendencies ers prefer domestic goods whereas others do not care about
have on attitudes toward importing products. Regarding the distinction between domestic and imported products.
antecedent relationships, the CETSCALE is shown to be Second, it provides useful information for both importers
 Sharma et al. / CONSUMER ETHNOCENTRISM 35

and exporters in selecting target markets and formulating tween ingroup and outgroup countries, they may evalute
appropriate marketing strategies. The prejudice against ingroup countries' products highly and prefer them, show-
imported products is a barrier for multinational marketers ing wider variation in their evaluation of the imported
to overcome, but it is also a favorable edge for domestic products by country.
manufacturers to take advantage of. Finally, the reliability of several of the scales was less
The segmenting variable, consumer-ethnocentric ten- than desirable. It should be noted, however, that all the
dencies, can be easily measured by the well-developed and constructs were conceptualized mostly from U.S. studies,
cross-culturally validated CETSCALE (Lichtenstein, and measures were adapted from scales developed in the
Netemeyer, and Durvasula 1991; Shimp and Sharma United States with appropriate modifications to make them
1987). The segments of interest are identifiable because relevant for Korea. Indeed, it is somewhat remarkable that
they possess distinctive and durable traits 6f conservatism, scale reliability in such a culturally disparate country is as
patriotism, cultural openness, and collectivism, and such high as detected. Furthermore, the major purpose of this
differences in demographic profiles as gender, income, and article was not to develop psychometrically rigorous cross-
education. Although the size of the consumer-ethnocentric cultural scales but to test the proposed hypotheses. In the
segment is unknown, there are hints of a substantial seg- final analysis, any problems with scale reliability merely
ment in the United States. For example, an opinion survey render conservative the strength of the observed effects.
by Purchasing Worm ("Reader Survey Shows" 1990) re-
vealed that 17 percent of respondents personally avoid
imports and another 33 percent prefer domestic products. ACKNOWLEDGMENTS
Seventeen percent indicated that their company has a"Buy
American" policy, whereas 82 percent favor buying do- The authors express appreciation to the Center for In-
mestic products, all things being equal. These results do ternational Business and Economic Research, College of
not directly bear on the population at large, but they do Business Administration, the University of South Carolina
suggest that the consumer-ethnocentric segment for at least for funding the project and Kwangyoon Nah and Kwang-
some products (cf. Herche 1992) may be far too big to be shin Choi for their help in collecting the data in Korea.
disregarded.
Third, international trade policymakers may gain some
insights from this study regarding consumer prejudices NOTES
against foreign goods. One problem encountered by policy- 1. Most of the moderation effects across the 10 products are quasi-,
makers in Congress with regard to trade with Japan is that, rather than pure-moderator effects (cf. Sharma et al. 1981).
in spite of few explicit trade barriers and government 2. Although no construct relating to overall willingness to import was
efforts to promote foreign products, Japan imports much hypothesized, the reliability of the 10 items is .788.
3. We conceptualized perceived product importance as another mod-
less than it should. One explanation for this phenomenon erating variable, but improper measurement prevents us from reporting
is that Japanese consumers simply prefer Japanese prod- the results of this analysis.
ucts ("Fear of Foreigners," 1989).

Limitations REFERENCES

The antecedent constructs included in this study are not Adomo, T. W., Else Frenkel-Brunswik, Daniel J. Levinson, and R. Nevitt
exhaustive; other variables such as allocentrism and dog- Sanford. 1950. The Authoritarian Personality. New York: Harper &
matism may be related to consumer ethnocentrism. Ac- Row.
cording to Tongberg (1972), products made in countries Anderson, James C. and David W. Gerbing. 1988. "Structural Equation
Modeling in Practice: A Review and Recommended Two-step Ap-
perceived to be culturally similar to the United States are
proach." Psychological Bulletin 103 (3): 411-423.
more acceptable to the highly dogmatic person and thus Anderson, W. T. and W. H. Cunningham. 1972. "Gauging Foreign Prod-
she or he is likely to be more willing to purchase products uct Promotion." Journal of Advertising Research (February): 29-34.
from such countries. Bagozzi, Richard P. 1981. "Evaluating Structural Equation Models with
Also, there may be other factors besides product neces- Unobservable Variables and Measurement Error: A Comment." Jour-
sity and perceived threat that moderate the effect of nal of Marketing Research 18 (August): 375-381.
consumer-ethnocentric tendencies on attitudes toward for- Bannister, J. P. and J. A. Saunders. 1978. "UK Consumers' Attitudes
eign products. 3 For instance, cultural similarity, historical Towards Imports: The Measurement of National Stereotype Image."
associations between countries, and present political- European Journal of Marketing 12: 562-570.
Bilkey, Warren and J. Erik Nes. 1982. "Country-of-origin Effects on
economic relations between countries may moderate the
Product Evaluations." Journal of International Business Studies
effect of consumer-ethnocentric tendencies. This differen-
(Spring/Summer): 89-99.
tial sensitiveness suggests that citizens of a country may Cattin, Philippe, Alain Jolibert, and Colleen Lohnes. 1982. "A Cross-
regard certain countries as ingroup (e.g., England and cultural study of 'Made In' Concepts." Journal of International Busi-
Canada for the United States) while regarding other coun- ness Studies 13 (Winter): 131-141.
tries as outgroup (e.g., Japan for the United States). If Catton, William R. 1960. 'q?he Functions and Dysfunctions of Ethnocen-
highly ethnocentric consumers do indeed distinguish be- trism: A Theory." Social Problems 8:201-211.
36 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE WINTER 1995

Chasin, Joseph B., Hartmut H. Holzmuller, and Eugene D. Jaffe. 1988. Lewis, I. M. 1976. Social Anthropology.in Perspective. Harmondsworth,
Stereotyping, Buyer Familiarity and Ethnocentrism: A Cross-cultural UK: Penguin9
Analysis" Journal of lnternational Consumer Marketing 1 (2): 9-29. Lynn, Richard. 1976. "The Sociobiology of Nationalism." New Society
Cote, Joseph A. and M. Ronald Buckley. 1987. "Estimating Trait, Method (July): 11-14.
and Error Variance: Generalizing across Seventy Construct Validation McDonald, R. P. and H. W. Marsh. 1990. "Choosing a Multivariate
Studies." Journal of Marketing Research 24 (August): 315-318. Model: Noncentrality and Goodness of Fit." Psychological Bulletin
Cote, Joseph A. and M. Ronald Buckley. 1988. "Measurement Error and 107 (2): 247-2559
Theory Testing in Consumer Research: An Illustration of the Impor- McGuire, William J~ 1983. "A Contextualist Theory of Knowledge: Its
tance of Construct Validation." Journal of Consumer Research 14 Implications for Innovation and Reform in Psychological Research"
(March): 579-582. Advances in Experimental Social Psychology 16: 1-47.
Crawford, John C. and Charles W. Lamb, Jr. 1981. "Source Preferences Mihalyi, Louis J. 1984. "Ethnocentrism vs. Nationalism: Origin and
for Imported Products." Journal of Purchasing and Materials Man- Fundamental Aspects of a Major Problem for the Future." Homboldt
agement 17 (4): 28-339 Journal of Social Relations 12 (1): 95-1139
Darling, John R. and E B. Kraft. 1977. "A Competitive Profile of Moore, David E. 1989. "Creating a Global Constituency for Free Trade."
Products and Associated Marketing Practices of Selected European Vital Speeches 56 (December): 108-110.
and Non-European Countries." European Journal of Marketing 11 (7): Morello, Gabriele. 1984. "The 'Made In' Issue: A Comparative Research
519-5309 on the Image of Domestic and Foreign Products." European Research
Daser, Sayeste and Havva J. Meriz. 1987. "Does Patriotism Have Any 12 (January): 5-21.
Marketing Value--Exploratory Findings for the 'Crafted with Pride Morris, James H., J. Daniel Sherman, and Edward R. Mansfield. 1986.
in U.S.A.' Campaign" In Advances in Consumer Research 14: "Failures to Detect Moderating Effects with Ordinary Least Squares-
536-537. moderated Regression: Some Reasons and a Remedy." Psychological
Dholakia, Nikhilesh, A. Fuat Firat, and Richard P. Bagozzi. 1980. "The Bulletin 99 (2): 282-288.
De-Americanization of Marketing Thought." In Theoretical Develop- Murdock, George P. 1931. "Ethnocentrism." In Encyclopaedia of the
ments in Marketing. Eds. Charles W. Lamb, Jr. and Patrick M. Dunne. Social Sciences9 Ed. Edwin R. A. Seligman. New York: Macmillan,
Chicago: American Marketing Association, 25-29. 613-614.
Eagly, Alice H. 1978. "Sex Differences in Influenceability." Psychologi- Nagashima, Akira. 1970. "A Comparison of Japanese and U.S. Attitudes
calBulletin 85 (1): 86-116. toward Foreign Products9 Journal of Marketing 34 (January): 68-74.
"Fear of Foreigners." 1991. Economist (August 10): 15-16. Narayana, C. L. 1981. "Aggregate Image of American and Japanese
Han, C. Min. 1988. "The Role of Consumer Patriotism in the Choice of Products: Implications on International Marketing." Columbia Jour-
Domestic versus Foreign Products." Journal of Advertising Research nal of World Business 16 (2): 31-35.
(June/July): 25-32. Netemeyer, Richard G., Srinivas Durvasula, and Donald R. Lichtenstein.
Herche, Joel. 1992. "A Note on the Predictive Validity of the 1991. "A Cross-national Assessment of the Reliability and Validity of
CETSCALE." Journal of the Academy of Marketing Science 20 (Sum- the CETSCALE" Journal of Marketing Research 28 (August): 320-
mer): 261-264. 327.
Hooley, Graham J., David Shipley, and Nathalie Krieger. 1988. "A Polhemus, Ted. 1988. "Pride and Prejudice." Marketing (UK). (July 14):
Method for Modeling Consumer Perceptions of Country of Origin?' 28-32.
International Marketing Review 5 (Autumn): 67-769 Ray, J. J. 1983. "A Scale to Measure Conservatism of American Public
Howard, Donald G. 1989. "Understanding How American Consumers Opinion." Journal of Social Psychology 119: 293-294.
Formulate Their Attitudes about Foreign Products." Journal of Inter- 9 1990. "Racism, Conservatism and Social Class in Australia:
national Consumer Marketing 2 (2): 7-24. With German, Califomian and South African Comparisons." Person-
Hui, C. H. 19889"Measurement of Individualism-Collectivism" Journal ality and Individual Differences 11 (2): 187-189.
of Research in Personality 22: 17-36. "Reader Survey Shows: Imports Still Stir Emotions." 1990. Purchasing
Hui, C. H., and H. C. Triandis. 1986. "Individualism-Collectivism: A World 34 (January): 25-26.
Study of Cross-cultural Researchers." Journal of Cross-cultural Psy- Reierson, C. 1966. "Are Foreign Products Seen as National Stereotype?"
chology 17 (June): 225-2489 Journal of Retailing 42 (Fall): 33-40.
Hung, C. L. 1989. "A Country-of-Origin Product Image Study: The Rose, Richard. 1985. "National Pride in Cross-national Perspective."
Canadian Perception and Nationality Biases." Journal of International International Social Science Journal 37: 85-96.
Consumer Marketing 1 (3): 5-26. Rosenblatt, Paul C. 1964. "Origins and Effects of Group Ethnocentrism
Johansson, Johny K9 Susan P. Douglas, and Ikujiro Nonaka. 1985. and Nationalism." Journal of Conflict Resolution 8 (2): 131- 1469
"Assessing the Impact of Country of Origin on Product Evaluations: Rushton, J. Philippe. 1989. "Genetic Similarity, Human Altruism, and
A New Methodological Perspective." Journal of Marketing Research Group Selection." Behavioral and.Brain Sciences 12: 503-559.
22 (November): 388-396. Sharma, Subhash, Richard M. Durand, and Oded Gur-Arie. 1981. "Iden-
J0reskog, Karl G. 1971. "Statistical Analysis of Sets of Congeneric tification and Analysis of Moderator Variables." Journal of Marketing
Tests9 Psychometrika 36: 109-1339 Research 17 (August): 291-300.
Jrreskog, Karl G. and Dag Stirbom. 1986. LISREL VI: User's Guide. Schooler, R. D. 1965. "Product Bias in the Central American Common
Mooresville, IN: Scientific Software, Inc. Market" Journal of Marketing Research 4 (November): 394-3979
Kohli, A. K., B. J. Jaworski and A. Kumar. 1993. "MARKOR: A Measure 9 1971. "Bias Phenomena Attendant to the Marketing of Foreign
of Market Orientation" Journal of Marketing Research 30 (Novem- Goods in the U.S." Journal of International Business Studies 2
ber): 467-4779 (Spring): 71-80.
Kumar, N., L. W. Stern, and R. S. Achrol. 1992. "Assessing Reseller Sherif, M., O. J. Harvey, B. J. White, W. R. Hood, and Carolyn W. Sherif.
Performance from the Perspective of the Supplier9 Journal of Mar- 1961. Intergroup Conflict and Cooperation: The Robber's Cave Ex-
keting Research 29 (May): 238-2539 periment. Norman, OK: University Book Exchange.
LeVine, Robert A. and Donald T. Campbell. 1972. Ethnocentrism: Theo- Shimp, Terence A. and Subhash Sharma. 1987. "Consumer Ethnocen-
ries of Conflict, Ethnic Attitudes, and Group Behavior. New York: trism: Construction and Validation of the CETSCALE" Journal of
Wiley. Marketing Research 27 (August): 280-2899
 Sharma et al. / CONSUMER ETHNOCENTRISM 37

Sumner, William Graham. 1906. Folkways:The SociologicalImportance tions." Journalof InternationalBusinessStudies 10 (Spring/Summer):
of Usages, Manners, Customs,Mores, and Morals. New York: Ginn & 81-90.
Co. Wilson, G. D., and J. R. Patterson. 1968. "A New Measurement of
Tongberg, R. C. 1972. "An Empirical Study of Relationships between Conservatism." British Journal of Social and Clinical Psychology 7:
Dogmatism and Consumer Attitudes toward Foreign Products." Dis- 264-269.
sertation. Pennsylvania State University.
Triandis, Harry C., Robert Bontempo, and Marcelo J. Villarel. 1988.
"Individualism and Collectivism: Cross-cultural Perspectives on Self-
Ingroup Relationships." Journalof Personality and Social Psychology
54 (2): 323-338.
Triandis, Harry C., Richard Brislin, and C. Harry Hui. 1988. "Cross-cul-
tural Training across the Individualism-Collectivism Divide." Interna- ABOUT THE AUTHORS
tional Journal of Intercultural Relations 12: 269-289.
Triandis, Harry C., Kwok Leung, Marcelo J. Villareal, and Felicia L. S u b h a s h S h a r m a is a professor of marketing and Distinguished
Clack. 1985. "Allocentric versus Idiocentric Tendencies: Convergent Foundation Fellow at the University of South Carolina. He has
and Discriminant Validation." Journal of Research in Personality 19: published in Journal of Marketing, Journal of Marketing Re-
395-415. search, Marketing Science, Management Science, and elsewhere.
Wall, Marjorie and Louise A. Heslop. 1986. "Consumer Attitudes toward
Canadian-made versus Imported Products." Journal of the Academy of Terence A. S h i m p is a professor of marketing and Distinguished
Marketing Science 14 (Summer): 27-36.
Foundation Fellow at the University of South Carolina. He has
Wang, Chih-Kang. 1978. "The Effect of Foreign Economic, Political and
published in Journal of Consumer Research, Journal of Market-
Cultural Environments on Consumers' Willingness to Buy Foreign
Products." Dissertation. Texas A & M University. ing Research, Journal of Marketing, and elsewhere.
Watson, Goodwin and David Johnson. 1972. SocialPsychology: Issues
and Insights. New York: J. B. Lippincott. J e o n g s h i n Shin is a lecturer of marketing at Chunnam National
White, Phillip D. 1979. "Attitudes of U.S. Purchasing Managers toward University, Korea. He received his Ph.D. in marketing from the
Industrial Products Manufactured in Selected Western European Na- University of South Carolina.
 -

www.xuebalib.com

24 IP