This paper was presented at the Association for Education in Journalism
and Mass Communication in Toronto, Canada, August 2004.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
[log in to unmask] For an explanation of the subject line, send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Oct 2004)
Thank you.
Elliott Parker
************************************************************************
Cross-cultural Content Analysis: A Study of the Top Frequently Visited Web
Sites
in the United States, China, and Korea
Corie Forrest, Gennadi Gevorgyan, Cong Li, and Youjeong Kim
Kansas State University
(785) 532-3965
[log in to unmask], [log in to unmask], [log in to unmask], [log in to unmask]
Submitted to International Communication Division
Markham Competition (Student Papers)
Abstract
The current study identifies the relationship between cultural variability
and online communication. Hofstede's model of cultural variability is used
as the theoretical framework to compare the Web site design in the United
States, China, and Korea. The results of the study suggest that the four
cultural dimensions examined (collectivism, power distance, masculinity,
and uncertainty avoidance) in the context of Web site design demonstrate
different levels across cultures. The implications of cultural variability
are discussed.
CROSS-CULTURAL CONTENT ANALYSIS
Cross-cultural Content Analysis: A Study of the Top Frequently Visited Web
Sites
in the United States, China, and Korea
Submitted to International Communication Division
Markham Competition (Student Papers)
Abstract
The current study identifies the relationship between cultural variability
and online communication. Hofstede's model of cultural variability is used
as the theoretical framework to compare the Web site design in the United
States, China, and Korea. The results of the study suggest that the four
cultural dimensions examined (collectivism, power distance, masculinity,
and uncertainty avoidance) in the context of Web site design demonstrate
different levels across cultures. The implications of cultural variability
are discussed.
Cross-Cultural Content Analysis: A Study of the Top Frequently Visited Web
Sites
in the United States, China, and Korea
The rapid growth of advanced communication technologies in the
21st century has diminished geographic boundaries while increasing the
ability of global communication. An essential component and determinant of
international communication is culture in its broadest sense. As a result,
numerous studies have been devoted to the fundamental question: How does
culture influence various aspects of life, particularly communication?
The Internet has brought together various markets from different
geographic locations, thus widening the scope of market segmentation on a
global level. Although the Internet enlarges potential audiences, it is
hardly possible to appeal to populations of different countries without
incorporating the numerous aspects of national cultures. Thus, it is not
surprising that more transnational businesses are adjusting their marketing
and promotional strategies to the cultural distinctions of their target
countries.
In light of the latter, the goal of Web developers is to create Web sites
that appeal to broad audiences, specifically, audiences of different
cultures. To accomplish this, national cultures need to be investigated and
integrated into online communication. Some earlier studies have been
focused on the cultural determinants of communication regarding cultural
similarities, differences, and expectations among different nations
(Hofstede, 1980; Simon, 2001; Tsikriktsis, 2002).
The purpose of this study is to investigate the reflection of culture and
its dimensions on the most frequently visited Web sites of three
economically, socially, and politically diverse countries (United States,
China, and Korea) through a cross-cultural content analysis. The importance
of this study is viable for global marketers, advertisers, and public
relations practitioners since the ultimate intention of global business is
to increase commercial profits through improving global appeal and
increasing global market shares. Since the number of Internet users is
expected to increase to 765 million by 2005, incorporating culture into Web
design is becoming especially crucial (Paul, 2001, p.739). Overall, the
basic research question for this study: Does national culture impact Web
site design?
This study uses the accepted cultural variability framework established by
Hofstede's (1980) five cultural dimensions. His model evaluated 50
countries and three regions in which he identified four cultural dimensions
through survey questionnaires and interviews of the IBM Corporation (Pi,
1996, p.29). Hofstede concluded that four cultural dimension existed
between the numerous countries including individualism, power distance,
masculinity, and uncertainty avoidance (Pi, 1996, p.30-31). Subsequent
research conducted by Hofstede and Bond (Hofstede, 1994, para. 11) found a
fifth dimension, long-term orientation.
The organization of this paper first conceptualizes culture through
various perspectives. Then, Hofstede's model of cultural variability will
be described, as well as each of the five dimensions of culture.
Subsequently, the next section will describe the concept of Web design that
is used in this study. Finally, the summary will incorporate Web site
design expectations with the four dimensions of culture emphasized in this
study.
Literature Review
Culture
Humans create, study, and apply culture to respond to the environment,
which in turn allows them to control and modify their surroundings (Naylor,
1996, p. 14-34). According to Naylor, recognizing this difference between
cultures is an essential step in understanding the influence that culture
has on the daily lives of people.
Based on the most frequently used definitions, culture is a set of learned
knowledge, beliefs, and behaviors which are shared by a group (Naylor,
1996, p. 37; Taylor, 1970, p. 14). Most importantly, culture is learned;
therefore, it is acquired by people to become members of a society.
Secondly, culture is defined as being shared. Hence, cultural phenomena are
not unique to a particular individual.
Cultural Dimensions
Hofstede (1980; 1994, para. 2) defines culture in the sense of "collective
programming of the mind," in which people are categorized and
differentiated. Moreover, Hofstede further categorizes culture in terms of
"symbols, heroes, rituals, and values" (Hofstede, para. 3). According to
Hofstede (p. 12), values are reflect the "deepest level of a culture,"
while nationality is considered to be "an involuntary attribute." Most
people encounter a variety of different cultures daily. For instance, one's
work environment can be categorized as a culture, as well as the culture of
a university. However, a national culture is differentiated by country.
Hofstede (p. 12) argues that, "national cultures differ mostly at the level
of basic values, while occupational and, even more, organizational cultures
differ more superficially (in their symbols, heroes, and
rituals)." Therefore, the basic value systems established within different
countries is pertinent to cross-cultural research, specifically cultural
variability.
The utility of Hofstede's (1980) research on international culture
provides a basic framework for further research in many disciplines,
especially communication. The cultural dimensions established by Hofstede
(1980) have been the theoretical framework for many studies that
investigate international culture and communication (Furrer & Liu, 2000;
Hofstede, 1998; Kersten & Koeszegi, 2003; Tsikriktsis, 2002), as well as
military intelligence research (Hoppa & Gray-Briggs, 1999, p. 32).
Originally, Hofstede (1980) categorized cultural variability within the
parameters of four dimensions including: individualism, power distance,
masculinity, and uncertainty avoidance. However, a fifth dimension,
long-term versus short-term orientation, was found through subsequent
research by professor Michael H. Bond of the University of Hong Kong
(Hofstede, 1994, para. 11). Although five dimensions of cultural
variability have been established, the current study emphasizes the
original four dimensions and excludes orientation. The premise being that
the accepted instrument chosen to measure cultural variability does not
include orientation
Individualism
Individualism, in comparison to collectivism, is one cultural dimension
established by Hofstede (1980, 1994). According to Hofstede, this dimension
represents "the degree to which people in a country have learned to act as
individuals rather than as members of cohesive groups: from collectivist to
individualist" (Hofstede, 1994, para. 9). Subsequently, individualistic
cultures value initiative; personal goals are warranted, as well as
attainable, and achievements are awarded to the individual (Hoppa &
Gray-Briggs, 1999, para. 16). However, cultures representing collectivism
value group rewards (Hofstede, 1994, para. 10). Therefore, "collectivist
cultures subordinate those to group membership and the goals of the group,"
rather than to the goals of the individual (Hoppa & Gray-Briggs, para. 16).
A collectivist culture emphasizes tightly knit groups that do not accept
outsiders easily. Furthermore, promotion is emphasized more in
individualistic cultures, rather than in collectivist cultures.
For the most part, individualistic nations are usually economically
developed and democratic. Thus, Western nations such as the United States
are considered to be individualistic (Kim & Kasser, 2003, p. 277). Other
individualistic nations, reported in Hofstede (1991) include Australia,
Canada, Denmark, France, Germany, Great Britain, Ireland, Israel, Italy,
Netherlands, Norway, and Sweden (as cited in Simon, 2001, para. 24). Many
collectivistic nations, such as China and Korea, are not as economically
developed as individualistic nations. However, China has recently seen
substantial economic growth and proceeds to rapidly develop (Kim & Kasser,
p. 277). Additional collectivist nations reported by Hofstede (1991)
include Brazil, Columbia, Egypt, Greece, India, Japan, Mexico, Pakistan,
Panama, Peru, Venezuela, and Thailand (as cited in Simon, para. 24).
Power Distance
Hofstede (1980; 1994) reports that power distance indicates the level of
inequality that the people of a nation are willing to accept. Accordingly,
Hofstede defines power distance by "the degree of inequality among people
which the population of a country considers as normal: from relatively
equal to extremely unequal" (Hofstede, 1994, para. 9). This "power" can be
established within institutions and organizations. Basically, cultures
reflecting a high-power distance value obedience to authority. Countries
exhibiting high-power distance, such as China and Korea, rely on rigid
rules and norms of socialization that are established at birth by family
and close relationships (Hofstede, 1980, p. 99). These cultures represent
an unequal power distance in which the people "accept power and authority
as part of life" (Hoppa & Gray-Briggs, 1999, para. 7). Hofstede (1991)
reports that some of the other countries that reflect a high-power distance
include Egypt, India, Malaysia, Nigeria, Saudi Arabia, and Venezuela (as
cited in Simon, 2001, para. 21).
On the other hand, cultures with low-power distance tend to value equality
spherically. Individual equality is the focus of these cultures, therefore
people "feel more free to express their own opinions while stressing expert
and legitimate power" (Simon, 2001, para. 21). Hofstede (1991) reports that
nations with low-power distance include Austria, Canada, Denmark, Germany,
Ireland, Israel, Sweden, and the United States (as cited in Simon, 2001,
para. 21). Moreover, populations within low-power cultures are more
inclined to question authority rather than passively obeying rules and
regulations (Hoppa & Gray-Briggs, 1999, para.7).
Masculinity
Masculinity is another cultural dimension established by Hofstede
(1980;1994). Hofstede defines masculinity in terms of the "degree to which
'masculine' values like assertiveness, performance, success and competition
prevail over 'feminine' values like the quality of life, maintaining warm
personal relationships, service, caring, and solidarity: from tender to
tough" (Hofstede, 1994, para. 9). For the most part, this dimension
emphasizes the social roles of the sexes in a given culture. While a
culture representing high masculinity "tends to maximize the social
differentiation between the sexes," a culture reflecting high femininity
"tends to minimize social differentiation between the sexes" (Hoppa &
Gray-Briggs, 1999, para. 10). Although masculine cultures value power, and
emphasize assertiveness and heroism, these cultures also favor competition
and Darwinian survival (Hofstede, 1994, para. 11; Hoppa & Gray-Briggs,
para. 10). Some examples of nations reflecting masculinity reported by
Hofstede (1991) include Austria, Italy, Japan, Mexico, Switzerland, Brazil,
and Venezuela (as cited in Simon, 2001, para. 23).
Conversely, feminine cultures value unity, as well as "sympathy for the
weak" (Hofstede, 1994, para. 11). Unlike the tendency towards performance
and recognition as found within masculine cultures, feministic cultures
emphasize nurturance and "tend to value people and modesty" (Hoppa &
Gray-Briggs, para. 10). Countries associated with feminity include Korea,
China, Chile, Denmark, Finland, Netherlands, Norway, and Sweden (Hofstede,
1991; as cited in Simon, 2001, para. 23). The United States falls in the
middle of the masculine versus feminine variability scale (Hofstede, 1991;
as cited in Simon, para. 23). However, a cross-cultural comparison of
ethical attitudes found that the United States is more masculine than Korea
(Christie, Kwon, Stoeberl, & Baumhart, 2003, p. 270).
Uncertainty Avoidance
Uncertainty avoidance is the final dimension in Hofstede's (1980) original
cultural variability theory. According to Hofstede (1994, para. 9),
uncertainty avoidance is the "degree to which people in a country prefer
structured over unstructured situations: from relatively flexible to
extremely rigid." Essentially, uncertainty avoidance is the level of
ambiguity a culture is willing to accept (Hoppa & Gray-Briggs, 1999, para.
6). A culture representing high-uncertainty avoidance is extremely rigid
and values conformity (Hofstede, 1994, para. 9; Hoppa & Gray-Briggs, para.
6). Moreover, these cultures also abide by strict beliefs, rules, and
principles (Hofstede, 1994, para. 11). Hofstede (1991) reports that Korea
represents a culture with high-uncertainty avoidance, as well as Argentina,
Belgium, Chile, Egypt, France, Greece, Japan, Mexico, Peru, and Spain (as
cited in Simon, 2001, para. 22).
However, cultures reflecting low-uncertainty avoidance, such as the United
States and China, are more flexible, especially when it comes to rules and
principles. Other countries exemplifying low-uncertainty avoidance are
Canada, Denmark, England, India, and Sweden (Hofstede, 1991; as cited in
Simon, 2001, para. 22). These cultures tend to favor opportunism (Hofstede,
1994, para. 11). Basically, cultures with low-uncertainty avoidance will
tolerate deviant behavior from its members, while cultures with
high-uncertainty avoidance will not tolerate any form of deviation from
their strict societal rules or principles (Hofstede, 1994, para. 11).
Web Site Design
The conceptualization of Web site design is pertinent to this
cross-cultural study. The perception of a Web site includes three
aspects: communication effectiveness, social presence, and communication
interface. Communication effectiveness refers to the ability of a Web site
to accommodate the user. This aspect of Web design allows the user to
fulfill his/her needs in respect to the medium. The concept of social
presence relates to the ability of a Web site to relate to a specific
culture. Accordingly, "sites whose attributes closely match the social
perceptions of users are more likely to engage users" (Simon, 2001,
Perception, para. 2). Therefore, social presence is the ability of a Web
site to relate to a nation's culture, which in turn emphasizes cultural
variability. Social presence includes the availability of graphics, colors,
or tools (Simon, Perception, para. 2). Finally, communication interface is
defined in terms of "the actions required by the user to activate and use
the channel" (Simon, Perception, para. 2). Hence, communication interface
includes navigation options, hyperlinks, and the availability of increased
levels of interactivity. For this study, the concept of interactivity
refers to how closely a Web site resembles the reality.
Web sites are designed to attract people. The growth of e-commerce and
online transactions on a global level bring together various geographically
dispersed markets into a whole. Culture plays an important role in Web site
development. An empirical study conducted by Tsikriktsis (2002) found a
relationship between culture and Web site quality expectations according to
Hofstede's (1980) cultural dimensions. Tsikriktsis (2002) categorized
elements of Web site quality expectations accordingly. Thus, these
differentiations will be used to classify Web sites regarding the
dimensions of culture established by Hofstede (1980).
Summary
The study of relevant literature gives a general understanding of various
issues surrounding the conceptualization of culture and its impact on
communication. The studies have shown relative differences between national
cultures (Hofstede, 1980; 1994) that lead to differences in how various
geographically dispersed populations interact within the computer-mediated
environment. The cultural dimensions established in Hofstede's (1980; 1994)
model of cultural variability include power distance, individualism,
masculinity, uncertainty avoidance, and orientation. This study focuses on
the United States, China, and Korea to investigate the similarities and
differences between culture variability and its relationship to Web design.
All three of the countries selected for this study are on the top 10 list
of Internet usage in the world (InternetWorldStats, 2004). Obviously, the
expectation of increased Internet usage will likely occur in these
countries. Thus, cultural determinants will be especially reflected in the
context of online communication of the countries chosen.
Since it is nearly impossible to document each of the cultural dimensions
in the context of Web design, this study focuses on collectivism, power
distance, masculinity, and uncertainty avoidance as proponents of cultural
variability represented in the nation's most frequently visited Web sites.
The parameters of the dimensions are based on previous research that
investigated the impact of culture and gender on sites (Simon, 2001) and
the influence of culture on Web site quality expectations (Tsikriktsis, 2002).
RQ1: Does the cultural dimension of individualism versus collectivism
impact Web site design?
H1: The top 10 frequently visited Web sites in China and Korea will show a
higher collectivistic orientation compared to the top 10 frequently visited
Web sites in the United States.
H2: The top 10 frequently visited Web sites in China and Korea will be
similar in the representation of collectivism.
This cross-cultural content analysis expects to find a substantial
difference in the design elements of the most frequented Web sites in the
United States, China, and Korea as a result of cultural variability. For
that reason, Internet users in individualistic cultures such as the United
States are more likely to use the Internet to find specific facts and
detailed information according to personal needs. Thus, the United States
will probably use more search engines and visit content-based Web sites
that reflect personal interests. Moreover, since the United States is an
individualistic culture, Web sites might provide additional links to
sources with the same information.
On the other hand, collectivistic cultures such as China and Korea may
frequent Web sites that are more group oriented. These sites are likely to
be visually appealing with advanced graphics and symbolic images may be
used as links. Unlike the United States, Chinese and Korean Web site may
not provide a number of additional sources for similar information. Plus,
Chinese and Korean sites might display more e-mail links, as well as chat
rooms or message boards.
RQ2: Does the cultural dimension of power distance impact Web site design?
H3: The top 10 frequently visited Web sites in China and Korea will show a
higher power distance orientation compared to the top 10 frequently visited
Web sites in the United States.
H4: The top 10 frequently visited Web sites in Korea and China will be
similar in the representation of high power distance.
While the United States reflects low-power distance, Korea and China
demonstrate high-power distance. Cultures with low-power distance value
individual equality and question authority. However, cultures with
high-power distance accept inequality and levels of power and authority as
normality. Thus, we expect to find more company hierarchy information,
proper titles, and vision statements provided on Korean and Chinese Web
sites, rather than on American sites.
RQ3: Does the cultural dimension of masculinity impact Web site design?
H5: The top 10 frequently visited Web sites in the United States will show
a higher masculinity orientation compared to the top 10 frequently visited
Web sites in China and Korea.
H6: The top 10 frequently visited Web sites in China and Korea will be
similar in the representation of femininity.
Korean and Chinese cultures are categorized as feminine, while the United
States falls in the middle of masculinity versus femininity scale.
Masculine cultures distinguish clear gender roles and favor competition.
Feminine cultures favor modesty emotional appeal. Although the United
States is not categorized as entirely masculine, we expect to find more
masculine representations such as quizzes and games, realistic themes, and
product effectiveness on American Web sites, rather than on Korean or
Chinese Web sites
RQ4: Does the cultural dimension of uncertainty avoidance impact Web site
design?
H7: The top 10 frequently visited Web sites in China and the United States
will show lower uncertainty avoidance orientation compared to the top 10
frequently visited Web sites in Korea.
H8: The top 10 frequently visited Web sites in the United States and China
will be similar in the representation of low uncertainty avoidance.
The cultural orientation in both the United States and China are
categorized as representing low-uncertainty avoidance, while Korean culture
reflects high-uncertainty avoidance. Cultures with low-uncertainty
avoidance accept ambiguity and are more flexible—traditional themes are not
prevalent. High-uncertainty avoidant cultures abide by strict rules and
object to ambiguity. Therefore, we expect to find more customer
testimonials, guided navigation, and traditional themes in Korean Web sites
rather then in American or Chinese Web sites.
Methods
Sampling
Since the United States, China, and Korea are among the top 10 countries
with the highest number of Internet users (InternetWorldStats, 2004), this
study hypothesizes that the cultural differences among these three
countries will be reflected through their domestic Web sites. The vastness
of the Internet hinders systematic sampling procedures and the availability
of accepted sampling measures for Internet research is finite. Since this
study emphasizes culture, the researchers felt that the popularity of Web
sites in each of the countries would reflect generalizable cultural
variability. Therefore, the researchers chose the top 10 frequently visited
Web sites in each country as the sample for this study
Based on the literature review, Web site ranking is not reported by
official sources. The industry Web statistics reported by Alexa (one
amazon.com company), which was founded in 1996, was used to determine the
most frequently visited domestic Web sites in the sample countries. The top
Web sites were chosen according to the English, Chinese, and Korean
language. To avoid potential bias, the exact ranking of each Web site was
not emphasized. Basically, the top 10 Web sites in each country were
treated equally for the purpose of this study. The top 10 Websites (Alexa,
2004) used in this study are presented in Table 1.
Instrument
Singh, Zhao and Hu (2003) examined cultural adaptations reflected on U.S.-
based, international companies' domestic Web sites and their Chinese Web
sites. Through extensive review, Singh et al. created an instrument with 35
items that operationalized Hofstede's (1980) four cultural dimensions
(collectivism, uncertainty avoidance, power distance and masculinity) and
Hall's (1976) contextual dimensions on the Web. In their work, Singh et al.
offered thorough justifications for including certain Web elements under
certain cultural dimensions, which showed a high validity for their
instrument.
This study applied the same instrument with a slight modification. Since
this paper does not emphasize the high and low context of culture developed
by Hall (1976), the eight items in the original instrument that assessed
high versus low context of culture were excluded. The remaining 27 items
that measured Hofstede's four cultural dimensions were used to determine
the cultural variability of the Web sites and the dimensions were defined
accordingly. The reliability of the instrument was tested twice by Singh et
al. and was proved to be 85% and 88% respectively. The four cultural
dimensions and their correspondent items are presented in Table 2.
Procedure
The use of content analysis to investigate information on the Web is still
in its infancy (Singh et al., 2003, p68). Kerlinger (2000) argued that the
method of content analysis must encompass three distinctive
characteristics: systematic, objective, and quantitative. This study can be
qualified as a content analysis at the broadest sense of the meaning. Each
Web site is analyzed according to the cultural variability instrument
chosen by a selection of coders and the results are quantitative.
Since this study sampled Web sites in three countries with three diverse
languages, the coders had to be selected accordingly. Two American
students, two Chinese students, and two Korean students attending a large
Midwestern university were selected to be the coders of this study. The
researchers decided to use two coders for each country to increase
reliability. All of the coders were trained for the coding procedure prior
to evaluating the sample Web sites. The operational definitions for each of
the 27 items used to measure cultural variability were offered to the
coders (Singh, N., personal communication, March 11, 2004).
Two of the 10 Web sites for each country were randomly selected for a pilot
study to calculate the intercoder reliability using Holsti's percentage of
agreement. Overall reliability for the U.S. Web sites was 0.87, for Chinese
Web sites was 0.85, and for Korean Web sites was 0.80. All of the coders
evaluated the Web sites to examine the occurrence or non-occurrence of all
the 27 items that represent the four cultural dimensions. They finished
coding within one week.
Web sites are operationally defined as the index page, or homepage, and all
of the available links provided on the index page. Additional links
provided on secondary pages are excluded for the purpose of this study.
Statistical Analysis
After the collection of data from the coders, Chi-square tests and t-tests
were performed. All of the 27 items were counted for occurrence in each of
the sample countries. The percentages of occurrence were put into a
cross-tab table for a Chi-square test to evaluate the differences between
each of three countries on each of the 27 items. To test the hypotheses, an
aggregate score for each of the four cultural dimensions was calculated.
For example, the instrument utilized in this study has eight items
to operationalize collectivism, and uncertainty avoidance includes nine
items. Thus, the aggregate score for each dimension was calculated
according to the number of items within each dimension. By adding all of
the frequencies of occurrence of the items, and then dividing by the number
of items according to each dimension, the resulting overall score
represents the Web sites' cultural variability according to the four
dimensions.
Results
To test our hypotheses, all the 27 items of Hofsted's four dimensions
(Collectivism, Power Distance, Masculinity, and Uncertainty Avoidance) were
compared through cross-tabulations to test for the significant cultural
differences between the United States, Chinese, and Korean Web sites. The
researchers also performed t-tests to analyze the raw data of each of the
27 items that determine cultural variability in each country. An alpha
level of 0.05 was used for all statistical tests in this study and the
interval confidence level was set at 95%.
Collectivism
According to the t-tests results, Chinese Web sites (M = 6.00) are shown to
have more collectivistic features than US Web sites (M = 3.00) (t =
-3.873, p = .001). Korean Web sites (M = 4.70) also reflect more
collectivism than US Web sites (M = 3.00) (t = -2.047, p = .028). However,
Chinese Web sites (M = 6.00) show more collectivism than Korean Web sites
(M = 4.70) (t = 2.512, p = .011). Therefore, hypothesis 1 which stated that
Chinese and Korean Web sites show more collectivism features than U.S. Web
sites is supported. Hypothesis 2 which stated that Chinese and Korean Web
sites have similar collectivistic orientations is not supported.
Power Distance
Hypothesis 3 predicts that China and Korea, being of higher-power
distance societies, will show more high-power distance features in Web
sites design than the United States. Accordingly, hypothesis 4 predicts
that Chinese and Korean Web sites will be similar regarding power distance
orientation. The t-test results show that Chinese Web sites (M = 4.40)
reflect higher-power distance when compared to U.S. Web sites (M = 3.10) (t
= -1.793, p = .0475). Korean Web sites (M = 4.50) also received a higher
score on power distance than U.S. Web sites (M = 3.10) (t = -1.971, p =
.035). There is no significant difference between Chinese Web sites (M =
4.40) and Korean Web sites (M = 4.50) regarding this dimension (t = -.218,
p = .415). Therefore, both hypothesis 3 and hypothesis 4 are supported.
Masculinity
Hypothesis 5 proposes that U.S. Web sites will show more masculinity
features than Chinese and Korean Web sites. Contrary to the prediction,
t-tests results showed that Chinese Web sites (M = 2.7) received a higher
score on masculinity than U.S. Web sites (M = 1.6) (t = -2.305, p = .0165).
Korean Web sites (M = 2.4) also have more masculinity features than U.S.
Web sites (M = 1.6) (t = -1.852, p = .0405). Therefore, hypothesis 5 is not
supported.
Hypothesis 6 predicts that Chinese and Korean Web sites have similar
orientation on masculinity. The t-test results showed that there is no
significant difference between Chinese Web sites (M = 2.7) and Korean Web
sites (M = 2.4) (t = .701, p < .246) on this dimension. Subsequently,
hypothesis 6 is supported. The results of the cross-tabulations and
Chi-square analysis are shown in Table 3, and the t-tests results are shown
in Table 4
Uncertainty Avoidance
For the uncertainty avoidance dimension, it was hypothesized that Chinese
and U.S. Web sites show lower uncertainty avoidance than Korean Web sites,
while Chinese and US Web sites have similar features. The t-test results
revealed that there is no significant difference between Chinese Web sites
(M = 5.60) and Korean Web sites (M = 5.00) (t = .970, p = .1725) on this
dimension. No significant difference is found between U.S. Web sites (M =
4.10) and Korean Web sites (M = 5.00) (t = -1.274, p = .1095) as well.
However, Chinese Web sites (M = 5.60) demonstrate higher levels of
uncertainty avoidance in comparison to US Web sites (M = 4.10) (t = -2.544,
p = .01). Therefore, neither hypothesis 7 or 8 is supported.
Discussion
Overall, the results of the current study suggest that the four cultural
dimensions examined are evident in the Web sites of the countries sampled.
Although some of the hypotheses could not be supported since the results
did not render acceptable statistical significance, the Chi-square tables
show that the countries do share some similarities and differences in the
content of Web sites.
Hypothesis 1 is supported—Korean and Chinese Web sites reflect more
collectivistic features than the United States. Hypothesis 2 was not
supported since the results did not show statistical significance between
the similarities of Korean and Chinese Web sites regarding collectivism.
However, Table 3 reveals that community relations and clubs and chat rooms
are prominent in both Korea and China. Surprisingly, family themes are
prevalent in China and even evident in the United States, but Korean Web
sites were reported to have zero family themes. Therefore, although Korean
and Chinese Web sites reflect collectivism, Korean sites emphasize
community relations while Chinese sites are more family oriented.
Hypothesis 3 is supported since Korean and Chinese Web sites reflect
higher power distance in comparison to the United States. Accordingly,
hypothesis 4 is supported due to the fact that the results did not show any
statistical difference in power distance between Korean and Chinese Web
sites. However, it seems that the use of proper titles is very important in
the United States and somewhat important in China. But, Korean Web sites
emphasize pride of ownership appeal and company hierarchy information.
Thus, Web sites tailored to the Korean culture will be the most effective
when companies supply vision statements and demonstrate through imagery how
owning a product is appealing to others. Moreover, quality information and
awards should be supplied on Web sites seeking to reach Korean and Chinese
populations, rather than American populations.
Hypothesis 5 is not supported since Chinese Web sites showed more
masculinity than the United States. Accordingly, hypothesis 6 is supported
since Chinese and Korean Web sites did not show any statistical difference
regarding masculinity. However, it seems that none of the countries
identified clear gender roles in their sites. Web sites tailored to Korea
and China should include product effectiveness to be the most persuasive.
But, this is not the case in the United States. Furthermore, quizzes and
games should be considered very effective marketing strategies when
employed in Korea, somewhat effective in China, and useless in the United
States.
Hypothesis 7 is not supported since the United States and Chinese sites
did not show lower uncertainty avoidance than Korean sites. There was not a
statistical difference in the level of uncertainty avoidance between the
United States and Korea. However, Chinese Web sites have a higher level of
uncertainty avoidance in comparison to the United States; therefore,
hypothesis 8 is not supported either. The implications of the results
suggest that traditional themes are prevalent in China and would be a
highly effective marketing strategy in that country, but not in the United
States. Moreover, local terminology can be added to enhance traditional
themes. It seems that free downloads and trials are the most effective in
China, somewhat effective in Korea, and marginally effective in the US.
Overall, customer service is emphasized in all of the countries.
Limitations
One of the limitations of this study is the restricted number of Web sites
that were analyzed. Some of the cultural dimensions might not be reflected
in the content of Web sites that are less visited than those included in
this study. Thus, including more Web sites in the study might, to some
degree, change the results of the research. Moreover, the implications of
this study may differ since the sample consisted of the top frequently
visited Web sites in each country.
Another limitation is the time period given for analyzing the content of
the 10 most frequently visited Web sites. Throughout time, the analysis of
Web sites may change--taking new characteristics and dropping old ones.
Thus, the results of the study could be different if the study sampled a
longer period of time and if the Web sites were randomly selected.
Implications for Future Research
The primary recommendation for future studies would be increasing the
sample size, which would increase the representation of Web sites while
also including more Web sites in the study. Also, it would be interesting
to add the fifth cultural dimension, long-term versus short-term
orientation, to the list of independent variables. Future studies might
need to be done to check the validity and timeliness of Hofstede's cultural
dimensions that were established a long time ago.
Although the instrument used in the current study is accepted in the mass
communication research arena, the rapidly changing online environment will
need new instruments to evaluate Web sites. For instance, the actual
construct of a Web site tailored to specified cultures needs to be
measured. An instrument needs to be developed to evaluate the most
effective type of actual Web design according to culture. This would entail
levels of interactivity, levels of telepresence, sophistication of
graphics, color schemes, animation, and digital video. This study does not
correlate levels of sophistication to the persuasive effectiveness of a Web
site and the latter could greatly impact global marketing.
Conclusion
The results of the current study prove that cultural variability is
evident in computer-mediated communication. Thus, Web sites should be
tailored to the cultures they are intended to reach. Although countries
have been evaluated according to the five cultural dimensions, it seems
that cultural adaptation is different on the Internet. If global marketers
and advertisers intend to increase their market shares, then cultural
variability must be accounted for in Web site design. It would be a Web
developer's dream to construct a Web site that could be tailored to all
cultures. Unfortunately, this notion is impossible due to the many
differentiating aspects of national cultures. The ultimate goal is to
effectively reach the most people with the least amount of design
modifications. Hopefully, this study provides an insight on how to design
Web sites tailored to American, Chinese, and Korean cultures.
Table 1: Top 10 Web sites in each country
Country
U.S.
China
Korea
Top 10 Websites
www.yahoo.com
www.msn.com
www.google.com
www.ebay.com
www.microsoft.com
www.passport.net
www.amazon.com
www.go.com
www.doubleclick.com
www.cnn.com
www.sina.com.cn
www.163.com
www.sohu.com
www.3721.com
www.baidu.com
www.tom.com
www.tencent.com
www.21cn.com
www.chinaren.com
www.eachnet.com
www.daum.com
www.naver.com
www.sayclub.com
www.bugs.co.kr
www.nate.com
www.netmarble.net
www.damoim.net
www.auction.co.kr
www.empas.com
www.freechal.com
Table 2: Cultural dimensions and correspondent items
Dimensions
Collectivism
Power Distance
Masculinity
Uncertainty Avoidance
Items
*Community relations
* Clubs or chat rooms
* Newsletter
* Family theme
* Country specific news
* Symbols and pictures of national identity
* Loyalty programs
* Links to local Web sites
* Company hierarchy information
* Pictures of CEOs
* Quality information and awards
* Vision statement
* Pride of ownership appeal
* Proper titles
* Quizzes and games
* Realism theme
* Product effectiveness
* Clear gender roles
* Customer
service
* Secure payment
* Guided
navigation
* Tradition theme
* Local stores
* Local terminology
* Free trials or downloads
* Customer testimonials
* Toll-free numbers
Table 3: Chi-square tests results for 27 items
US China Korea c 2 p
Collectivism
Community Relations 50 90 100 8.75 .013*
Clubs or Chat Rooms 30 100 100 18.261 .000*
Newsletter 40 100 80 9.545 .008*
Family Theme 40 80 0 13.333 .001*
Country Specific News 30 80 70 5.833 .054
Symbols and Pictures of 10 0 0 2.069 .355
National Identity
Loyalty Program 30 70 40 3.482 .175
Links to Local Websites 70 80 80 .373 .830
Power Distance
Company Hierarchy 40 70 90 5.700 .058
Information
Pictures of CEO's 30 60 30 2.500 .287
Quality Information 50 100 100 12.000 .002*
And Awards
Vision Statement 50 70 90 3.810 .149
Pride of Ownership 50 70 90 3.810 .149
Appeal
Proper Titles 90 70 50 3.810 .149
Masculinity
Quizzes and Games 20 70 90 10.833 .004*
Realism Theme 50 70 30 3.200 .202
Product Effectiveness 60 80 80 1.364 .506
Clear Gender Roles 30 50 40 .833 .659
Uncertainty Avoidance
Customer Service 80 100 100 4.286 .117
Secure Payment 60 70 100 4.845 .089
Guided Navigation 100 100 70 6.667 .036*
Tradition Themes 10 70 0 14.659 .001*
Local Stores 30 60 10 5.700 .058
Local Terminology 10 70 60 8.304 .016*
Free Trials and 40 80 60 3.333 .189
Downloads
Customer Testimonials 50 0 70 10.833 .004*
9.Toll Free Numbers 30 10 30 1.491 .475
*p<.05
Table 4: The t-test results for the four dimensions
U.S. China t p
M M
1.Collectivism 3 6 -3.873 .0005*
2.Uncertainty Avoidance 4.1 5.6 -2.544 .01*
3.Power Distance 3.1 4.4 -1.793 .0475*
4.Masculinity 1.6 2.7 -2.305 .0165*
U.S. Korea t p
M M
1.Collectivism 3 4.7 -2.047 .028*
2.Uncertainty Avoidance 4.1 5 -1.274 .1095
3.Power Distance 3.1 4.5 -1.971 .035*
4.Masculinity 1.6 2.4 -1.852 .0405*
China Korea t p
M M
1.Collectivism 6 4.7 2.512 .011*
2.Uncertainty Avoidance 5.6 5 .970 .1725
3.Power Distance 4.4 4.5 -.218 .415
4.Masculinity 2.7 2.4 -1.852 .246
______________________________________________________________________________________
*p<.05
References
Alexa. (2004). Top sites. Retrieved March 3, 2004, from
http://www.alexa.com/site/ds/top_500?p=DestTrLearn_W_g_40_T1
Furrer, O., & Liu, B. (2000). The relationships between culture and service
quality
perceptions: Basis for cross-cultural market segmentation and resource
allocation. Journal of Service Research, 2, 4, 355-371. Retrieved February
14, 2004, from ABI_INFORM database.
Hall, E.T. (1976). Beyond Culture. New York: Anchor Press.
Hofstede, G. (1994). Business Cultures. UNESCO Courier, 4, 12-16. Retrieved
February 12, 2004, from the PerAbs database.
(1980). Culture's Consequences: International Differences in Work-Related
Values. Beverly Hills, CA: Sage Publications.
(1983). National Cultures in Four Dimensions: A Research-Based Theory of
Cultural Differences Among Nations. International Studies of Management
& Organization, 13, 1/2, 46-74.
Hofstede, G., & Bond, M. (1993). Individual perceptions of organizational
cultures:
A methodological treatise on levels of analysis. Organization Studies, 14,
4, 483-
503. Retrieved February 14, 2004 from ABI_INFORM database.
Hoppa, J., & Gray-Briggs, A. (1999). Investigative agents: Putting theory
into practice.
Military Intelligence Professional Bulletin, 25, 4, 32-36. Retrieved
February 12,
2004, from PerAbs database.
InternetWorldStats, (2004). Top 10 Internet countries with highest number
of users.
Retrieved March 4, 2004, from http://www.internetworldstats.com/stats.htm.
Kerlinger, F. N. (2000). Foundations of behavioral research (4th ed.). New
York: Holt,
Rinehart & Winston.
Kersten, G., & Koeszegi, S. (2003). The effects of culture in
computer-mediated
negotiations. Journal of Information Technology Theory & Application, 5, 2,
1-28. Retrieved February 14, 2004, from ABI_INFORM database.
Kim, Y., & Kasser, T. (2003). Self-concept, aspirations, and well-being in
South Korea
and the United States. Journal of Social Psychology, 143, 3, 277-290.
Retrieved
February 2, 2004, from the PerAbs database.
Naylor, L. (1996). Culture and change. USA: Bergin and Carvey
Pi, Chung-Ron. (1996). Expression of culture in economic development. USA:
Aldershot Brookfield.
Simon, S. (2001). The impact of culture and gender on Web sites: An
empirical study.
Databases for Advances in Information Systems, 32, 1, 18-37. Retrieved
February 14, 2004, from the ABI_INFORM database.
Singh, N., Zhao, H. & Hu, X. (2003). Cultural adaptation on the Web: a
Study of
American companies' domestic and Chinese Websites. Journal of Global
Information Management,11, 1, 63-81.
Tai, S., & Chan, R. (2001). Cross-cultural studies on the information
content of service
advertising. Journal of Services Marketing, 15, 6/7, 547-564. Retrieved
February
12, 2004, from the PerAbs database.
Taylor, R. (1970). Cultural ways. USA: Allyn and Bacon, Inc.
Tsikriktsis, N. (2002). Does culture influence Web site quality
expectations? An
empirical study. Journal of Service Research, 5, 2, 101-112. Retrieved
February 14, 2004, from ABI_INFORM database.
|
