History Of Measuring Consumer Ethnocentrism Marketing Essay

Published Date: 23 Mar 2015

For marketers it is essential to understand the importance of attitudes, motives, beliefs and attitude change in the study of Marketing and Consumer Behavior. Therefore, Consumer Behavior is about the perceptive of the consumers' needs and what affect their purchase intention. Consumer behavior can be distinct as the study of individuals, groups or organizations and the processes they use to select, secure, use and dispose of products, services, experiences or ideas to satisfy needs and the impacts that these processes have on the consumer and society (Hawkins, 2001). It covers the field from psychology, sociology and economics.

With the open-door-policy in 1979, China has improved its trade relations with the rest of the world. China benefited from foreign direct investments and its export expansion. Today, China is the second largest economy in the world. The rapid growth in China led to an increase in consumption. Chinese consumers nowadays have comprehensive choices in buying different kind of products. Due to the open-door-policy, many overseas companies have entered China and benefited from the increased consumer spending. Chinese consumers can choose from foreign goods that are imported from overseas or manufactured by foreign companies in China, or domestic goods. Due to the increased foreign investments, domestic companies face fierce competition in several markets. Markets are globalizing in China but Chinese consumers still keep buying local products. The globalization affects consumer behavior but China has a strong traditional culture which remains noticed in the country's choice of product purchase. According to Mooij (2011), globalization can also cause a strengthening of local identities. Studies have identified that consumers in developing countries have a different perception than consumers in developed countries towards local made products. The consumers in developing countries believe that foreign sourced goods are better than local made products. Belk (1996) identified that globalization can lead to motives to resist the globalization policies, such as localism, ethnogenesis and the neo-nationalism. Further research about this paradox can be an effective way for investors, entrepreneurs, companies and producers to develop and enhance their marketing strategy in China.

There have been several studies about consumer ethnocentrism and its effect on consumer behavior and purchase intention. Most of the studies have their focus on the Western consumers. There has been little emphasis on the research of Chinese consumer' purchasing behavior and evaluation towards consumer ethnocentrism, which leads to a scarce knowledge of the Chinese consumers. This study focuses at explaining the degree of consumer ethnocentrism and its assumptions on evaluation of several products. This research will focus on food products by Chinese consumers.

To explain the relationship there has to be a literature review to issue the concepts in the research and analyzing existing studies on consumer ethnocentrism to understand the theoretical background. Secondly, a research question and the hypotheses will be described and after the methodology will be explained. To test the hypotheses, data will be collected. At last, after evaluation of the hypotheses, a conclusion can be drawn from the present study.

1.1 Problem statement

Chinese consumers are shifting towards other consuming patterns as the economy grows faster than any other country in the World. The behavior of Chinese consumers has shown significant changes in the couple of years. In the 12th five-year plan, the Chinese government wants to boost its domestic consumption. The changes have led to a declining demand for foreign goods. An explanation can be that Chinese consumers are becoming more ethnocentric because of the favoring for domestic products. The central question in this research is therefore: Does Consumer Ethnocentrism (CE) apply for the Chinese consumer towards food products? Which moderators are influencing the relationship between CE and the outcomes of CE for Chinese consumers towards food products?

1.2 Research method

A literature review is needed in order to answer the central question. A literature review is needed to gain insight about a suitable method to measure Consumer Ethnocentrism. It is important to set a theoretical background to gain insight about the factors that are influencing CE, the effects of CE and the moderators that are influencing the relationship between CE and the outcomes of CE. A conceptual framework can be developed after a literature review. In this framework, the factors that can be found that are influencing CE and also the outcomes and moderators that are influencing the relationship between CE and the outcomes, will be presented. The final conceptual model will be empirical tested. This research will be a quantitative research. A questionnaire will be developed and a 'convenience sample' will be used to collect data. A total of 210 questionnaires will be delivered randomly to adult participants.

1.3 Thesis structure

In Chapter 2, there will be an explanation how Consumer Ethnocentrism can be measured and if this is applicable in this research. In Chapter 3 there will be a brief description of the Chinese consumer market. In Chapter 4, the conceptual framework of Consumers Ethnocentrism will be developed and thereby its hypotheses. This framework will be the basis for the empirical research that will be described in Chapter 5. At last, the conclusions, limitations and recommendations will be discussed in Chapter 6.

2. Theoretical background

2.1 Ethnocentrism

The concept ethnocentrism was first introduced by Sumner (1906). According to the author, ethnocentrism is the tendency to perceive that a particular ethnic or cultural group is being more significant than other groups. Individuals who tend to be more ethnocentric find their group better than others, who are seen as inferior. Ethnocentric individuals have developed an own perspective of how they see people or groups and they don't accept individuals who are different from them (Netemeyer et al., 1991; Shimp and Sharma, 1987).

In the field of sociology Lewis (1976) argued that individuals tend to give their group members a privileged treatment than non-group members. Thus, ethnocentrism established the development of "in-groups", which is the group of the individual as pride and on the other side a disfavor for ''out-groups'' (Levin & Campbell, 1972). According to Lynn (1976) ethnocentrism is part of the human nature. Thus, the phenomenon can be used for large diversity of societal groups including the local community, regions and nations. In the study of Chryssochoidis et al (2007, p. 1518) he issued: "ethnocentrism is based on the formation of "we-group" feelings, whereby the in-group is the focal point and all out-groups are judged in relation to it". Levinson (1950) also related ethnocentrism to cultural narrowness which explains behavioral tendencies of favoring the individuals with the same culture and excluding other cultures, who are not the same.

2.2 Consumer ethnocentrism

Consumer ethnocentrism is derived from the psychological concept of ethnocentrism and particularly points out to ethnocentric perceptions held by consumers in one country, the so called in-group, towards products from a different country, the out-group (Shimp and Sharma, 1987; Shankarmahesh, 2004). Ethnocentrism is adapted in order to suit Consumer behavior, which is suitable for marketers. Ethnocentrism has already been discussed to the study of consumer behavior in the 1970s (Markin, 1974; Berkman et al, 1978). Shimp and Sharma (1987) defined the concept in order to suit the field of marketing.

Sharma, Shimp & Shin (1995) argue that the ethnocentric tendencies of consumers are influenced by social-psychological and demographic factors. The basic conceptual model of Sharma, Shimp & Shin (1995) shows that Consumer Ethnocentric is the focal concept that is related to demographic variables (age, gender, education) and social physiological variables (openness to foreign cultures, patriotism, collectivism/individualism and conservatism).

The result of Consumer ethnocentrism is overestimating the domestic products and underestimating the import products. Thus, Consumer Ethnocentrism is a significant factor towards import product attitude. In their model, there is also a moderating factor. The model issues that when products are perceived as not necessary, consumers will have stronger ethnocentric tendencies towards the attitude of import products. Sharma, Shimp & Shin (1995) have tested this conceptual model in Korea. Results have shown that there is a negative correlation between openness to foreign cultures and Consumer ethnocentrism. This shows that people who are open to foreign cultures, are less ethnocentric. There is a positive correlation between patriotism and Consumer ethnocentrism. The demographic variable 'gender' is also significant related to CE: women show more ethnocentric tendencies than men. The study also identified that higher education leads to less ethnocentric tendencies. The research did not find a negative correlation between age and CE.

Consumer Ethnocentrism tendencies

Antecedents

Openness to foreign cultures

Patriotism

Conservatism

Collectivism/individualism

Attitude towards imports

Moderators

Perceived necessity

Economic threat

Demographics

Age

Gender

Education

Figure : Basic model Sharma, Shimp & Shin (1995)

Consumer ethnocentrism is about the perceived appropriateness and morality when buying foreign products and preferring domestic goods over foreign goods which can be seen as consumer loyalty (Shimp and Sharma, 1987). According to Shrimp and Sharma (1987) highly ethnocentric consumers tend to believe that purchasing foreign-produced goods will affect the country's economy negatively and will cause unemployment. Conversely, non-ethnocentric consumers, who do not consider ethnocentric beliefs, have no bias where a product is produced. These consumers will consider their purchase based on evaluation of the merits of the products. Shimp and Sharma (1987) state: "The family unit would be expected to be the primary socialization agent, but adult opinion leaders, peers and mass media would also influence a child's ethnocentric orientation during the prime period of early childhood socialization". This statement explains that an adult's perception is created by what he or she has learned and experienced as a child.

According to Lingquist (2001) the degree of ethnocentrism is influenced by the developed levels of different regions. In some studies there has been concluded that consumers in developing countries tend to perceive foreign sourced products as prestigious and of high quality while on the other side, developed countries tend to perceive domestic products as prestigious and of higher quality (Agbonifoh and Elimimian, 1999; Bow and Ford, 1993; Sklair, 1994; Wang et al., 2000).

Sharma et al. (1995) assume that the less significant product categories lead to greater ethnocentric tendencies and behavior by consumers. Orth & Firbasova (2003) and Dosen & Krupka (2007) have researched the food industry, where it has been identified that ethnocentric consumers have more willingness to purchase domestic food products instead of foreign food products.

To measure consumer ethnocentrism, Shimp and Sharma (1987) developed The Consumer Ethnocentric Tendencies Scale (CETSCALE). The CETSCALE is a 17-item predictor of consumer ethnocentrism (Lindquist et al. 2001). This measurement distinguishes consumers whether they are highly ethnocentric or not, thus consumers scoring high on this scale will tend to prefer domestic products over foreign ones. There are several researchers that use the CETSCALE with their own modification for aiming on employment impact, patriotism and economic impact in studies (Lindquist et al., 2001; Cleveland et al., 2009). The scale has been applied widely in different countries, such as Turkey, France and the Czech Republic (Klein et al., 2006). Herche (1992) identified that the CETSCALE is able to forecast consumers' choices to buy domestic or foreign products. He argues that this measurement is even better than demographic and marketing mix variables. According to Balabanis (2001), the Customer ethnocentric measure of buying intentions differs from countries. Good and Huddleston (1995) agreed on this. They identified that Polish consumers find it significant to purchase foreign goods, contrary to Russian consumers. In a research study of Wei (2008) he issues that the Consumer ethnocentrism will decrease when brand sensitivity and product cues are taken in account. These factors will influence the purchase intention of the consumer. The CETSCALE will be further explained in paragraph 2.3.

2.3 CETSCALE

The central question in this research is: Does Consumer Ethnocentrism apply for the Chinese consumer towards food products? This question can only be explored if there is a method to measure CE. There are different kinds of methods and tools to measure CE, but the Consumer Ethnocentric Tendencies Scale is the most common used. Shimp and Sharma (1987) have developed this scale to measure CE (see 3.1). Previous studies have tested this scale on its reliability and validity (see 3.2 and 3.3). At last there will be a conclusion if this CETSCALE is valid for measuring CE in this research.

2.3.1 Measuring Consumer-ethnocentrism

The first study to measure CE was done by Shimp (1984). He came to a conclusion that measuring CE was being hindered by a suitable tool. Before Shimp developed the CETSCALE, there were other tools, such as the famous F- and E-scales of Adorno et al. (1950). These scales were able to measure ethnocentric tendencies but were useless because of the commonality and dated nature of the scales. Shimp (1984) used an open question to test CE empirically. He used the following sentence: "Please describe your views of whether it is right and appropriate for American consumers to purchase products that are manufactured in foreign countries." The question is aimed on the perception of the respondent, questioning if buying foreign products is ethical. Shimp (1984) indicated that the choice for an open question is legitimate, given that the concept of CE is very complex and the relevant dimensions were not known at that moment.

Shimp & Sharma (1987) developed and validated a multi-item scale to measure the CE of individual consumers: the CETSCALE, which stands for Consumer Ethnocentric Tendencies Scale. The CETSCALE consist 17 items, with a 7-point Likert-scale ranged from 'Totally disagree ' (1) to 'Totally agree'' (7). The 17-item CETSCALE is validated and tested on its reliability in three studies, which are the 'four area study', Carolinas study' and 'crafted-with-pride study'. Shimp & Sharma (1987) also developed a shortened version of the CETSCALE for the 'national consumer good study'. This scale consists 10 items of the original 17-item scale. This is because the 17-item scale is often being used for commercial market research. For this subset scale Shimp & Sharma (1987) use a five-point Likert-scale. In table 1, the 17-item and the subset 10-item scale is shown.

17-item CETSCALE

10-item CETSCALE

1

American people should always buy American-made products instead of imports.

2

Only those products that are unavailable in the U.S. should be imported.

1

Only those products that are unavailable in the U.S. should be imported.

3

Buy American-made products. Keep America working.

4

American products, first, last, and foremost

2

American products, first, last, and foremost

5

Purchasing foreign-made products is un-American

3

Purchasing foreign-made products is un-American

6

It is not right to purchase foreign products, because it puts Americans out of jobs.

4

It is not right to purchase foreign products, because it puts Americans out of jobs.

7

A real American should always buy American-made products.

5

A real American should always buy American-made products.

8

We should purchase products manufactured in America instead of letting other countries get rich off us.

6

We should purchase products manufactured in America instead of letting other countries get rich off us.

9

It is always best to purchase American products.

10

There should be very little trading or purchasing of goods from other countries unless out of necessity.

11

Americans should not buy foreign products, because this hurts American business and causes unemployment.

7

Americans should not buy foreign products, because this hurts American business and causes unemployment.

12

Curbs should be put on all imports.

13

It may cost me in the long-run but I prefer to support American products.

8

It may cost me in the long-run but I prefer to support American products

14

Foreigners should not be allowed to put their products on our markets.

15

Foreign products should be taxed heavily to reduce their entry into the U.S.

16

We should buy from foreign countries only those products that we cannot obtain within our own country.

9

We should buy from foreign countries only those products that we cannot obtain within our own country.

17

American consumers who purchase products made in other countries are responsible for putting their fellow Americans out of work.

10

American consumers who purchase products made in other countries are responsible for putting their fellow Americans out of work.

Table 1: The CETSCALE of Shimp & Sharma (1987)

2.3.2 International reliability

The basic requirements of a scale, that can be used in several countries, is the reliability of a scale (Craig & Douglas 2000). In case a scale is being used in a different country than where the scale is developed, there can be a 'measurement invariance'. This refers to 'observing the construct in different conditions and studying constructs where the measurement does not always measure the same attribute'. Linguistically or conceptual inequality of measuring instruments can cause variations in the reliability. This is a threat to the validity of conclusions. The focus on the study of the reliability is to obtain the same results when using the existing measuring instrument in a different context, different way or different point of time. These tools do not have to be reliable in every context, despite the attempt to develop tools that can be used in every culture (Craig & Douglas 2000). Therefore, the CETSCALE does not have to be reliable everywhere. If the scale in the United states gives a good reflection of the items that are measuring the CE in the United States, however this does not mean that these are the right items for a different country (Douglas & Craig 2000, p.277). It is important to determine if the CETSCALE is international reliable and therefore suitable in other countries. The reliability of a scale can be identified in multiple ways. There are two parts of reliability of the CETSCALE that come up for discussion: The internal consistency and the dimensionality of the CETSCALE and the stability of the CETSCALE over time. The internal consistency is being measured with the Cronbach's Alpha. The Cronbach's alpha measures if the items of the CETSCALE truly measure one concept: Consumer Ethnocentrism. The internal consistency has to be large enough (alpha higher than 0.60) to measure CE with the CETSCALE. When this is the case, the items can be considered as reliable indicators of CE (Hair et al. 1998). When a low alpha occurs, the respondents do not have consistent image of the concept and in that case the measurement is not reliable. A requirement by measuring the internal consistency of a set of items, is that the items are one-dimensional (Hair et al. 1998). This means that the items have to be interdependent and combined they have to measure one concept (Hair et al. 1998). Factor analyses are important to measure the dimensionality of a set of items by determine several factors (Hair et al. 1988). For one-dimensionality all items of the CETSCALE have to load high on one single factor (Hair et al. 1998). The stability of the CETSCALE over time is being measured by a test/retest method. The consistency of the answers of a respondent is being measured at different moments. The objective of this method is to be sure that the answers do not vary over time, in order for the CETSCALE to be considered reliable at every moment (Hair et al. 1998).

According to Hair et al. (1998) Cronbach's alpha is being defined as: 'Cronbach's Alpha is most widely used objective measure of reliability. It is used to measure the internal consistency of a test or scale. The measurement is expressed as number of 0 and 1, whereby the nearer Cronbach's Alpha coefficient is to 1, the better the internal consistency of the set of variables'.

Internal consistency and dimensionality of the CETSCALE

Shimp & Sharma (1987) were the first researchers that executed four distinctive studies to determine the reliability of the 17-item CETSCALE. These studies have shown that the internal consistency of the CETSCALE is very high; the Cronbach's alpha of the four studies varies from 0.94 to 0.96. The internal consistency and the dimensionality of the CETSCALE is identified in several studies and countries (Appendix 1).

The alpha coefficients in the appendix show a consistency. That implies that the alpha in every study lies about 0.90 (with exception of the alpha for Hungary in the study of Lindquist et al. (2000)). In any other cases the internal consistency meets the minimum of 0.60.Orth & Fibrasova (2003) have studied the role of CE towards the evaluation of food. They have found a high Cronbach's Alpha value for the CETSCALE and imply that the CETSCALE is internal consistent. There are different opinions about the amount of items of the scale that must be used. Shimp & Sharma (1987) have used 10 items in the 'national consumer good study', because of the limited possibilities in the questionnaire.

Lindquist et al. (2001) have studied the dimensionality of the shortened 10-item CETSCALE in Hungary, Czech and Poland. According to this research, the CETSCALE does not have a universal ''good fit' in the central- and east-European countries. They assume that a scale that can be used in every country does not exist. Therefore, they suggest to use a part of the CETSCALE items and add some land specific or population specific items to effectively measure the CE. Another problem the authors have identified is the translation of the CETSCALE. The authors imply that the scale functions better in English then translated to their native languages in their research. The authors emphasize that the 'back-translation' technique must be used to set up an accurate translation of the scale.

Stability of the CETSCALE over time

Shimp & Sharma (1987) have used the test-/retest method by doing the research twice with a interim period of five weeks. These studies have found comparable alpha values. Shimp & Sharma (1987) conclude that the CETSCALE is stable over time. Nielsen & Spence (2001) have also researched the stability of the CETSCALE over time. They studied the stability in the United States over an eight week period during the summer of 1992. In this period they have found that the CETSCALE-scores are stable over time but by investigating different groups, the scores can vary over time. They conclude that a longer period of data collection, scores can vary more.

2.3.3 International validity

The validity of a scale refers whether a scale or set of measurements measures the focal concept accurate (Hair et al. 1998). The three most accepted types of validity are convergent, discriminant and nomological validity. Convergent validity examines the degree to which the operationalization converges with other known measuring instruments of the concept. Discriminant validity assures that the scale differs enough of other similar concepts and nomological validity measures if the scale shows the relationship based on previous studies or theories. These types of validity are empirical tested by defining the correlation between theoretical defined sets of variables (Hair et al. 1998).

Convergent validity

This type of validity examines the correlation between two measurements that measure the same concept (Hair et al. 1998, p. 118). A researcher can look for this and find an alternative tool to measure CE and correlate this method with the CETSCALE. If the correlations are high, then the CETSCALE truly measures CE. Shimp & Sharma (1987) have found a convergent validity in their study. The 17-item CETSCALE correlates with the open question of Shimp (1984). This correlation (r (=correlation coefficient) =0.54, n=388, p <0,001) implies that the CETSCALE is convergent valid.

Discriminant validity

Discriminant validity is the degree where two concepts are different. This type of validity examines the correlation between measurements. In this case, the CETSCALE is being correlated with a different conceptual measurement. The correlation has to be low, as it shows that the CETSCALE differs enough comparing to another measurement (Hair et al. 1998, p. 118).

Shimp & Sharma (1987) have found evidence for discriminant validity. Three constructs are being used in the studies (patriotism, politics, economic conservatism and dogmatism) that are related to Consumer Ethnocentrism. The researchers have used a test of Fornelll & Larcker (1981). They assume that the average variance of the separated constructs is larger than the variance where the constructs are being combined. They have found that this is the case for the CETSCALE and the politic-economic conservatism-scale. The variance distinctive is 71% for the CETSCALE and 61% for the conservatism-scale, as the combined variance of the two constructs is 34%. Thus, Shimp and Sharma (1986) indicate that there is a discriminant validity.

The study of Sharma, Shimp & Shin (1995) also proves that there is a discriminant validity between CE and the product attitude towards foreign goods. They have found this through a factor analyses on the 17-item CETSCALE and the 10 attitude items. Findings have shown that the CETSCALE items and the attitude-items load on two different factors. A factor analyses have shown that the correlation between the two constructs (r = 0.568) differs significantly . Thus, these two constructs are not similar.

Nomological validity

Nomological validity examines how far the CETSCALE can give an accurate prediction for other concepts in a model that is based on theories. In this case, a research must identify theoretical supporting relationships from earlier studies, then he has to determine if the scale has comparable correlations (Hair et al. 1998, p.118).

For the 17-item scale Shimp and Sharma (1987) have used the CETSCALES-scores of the respondents to determine the nomological validity. Shimp and Sharma (1987) applied different measuring instruments of Warshaw (1980), Fishbein & Ajzen (1975) and Ajzen & Fishbein (1980). These tools have been used two years before the development of the CETSCALE to investigate the purchase of a domestic or foreign car. The purchase of a foreign car is negative correlated with the CETSCALE score of the ethnocentric respondents of Shimp and Sharma (1987). Furthermore, the ethnocentric consumers have less favorable cognitive structures and attitudes regarding foreign cars than non-ethnocentric consumers. These results imply the nomological validity of the Consumer Ethnocentrism concept and measuring this concept with the CETSCALE.

The 10-item scale is also tested on its nomological validity by Shimp and Sharma (1987). They have tested this by looking to the country origin of the manufactures. According to this study, the country of origin of the manufacturer is an important purchase consideration when the scores on the CETSCALE increase. Consumers with a higher score on the CETSCALE are likely to choose for American manufactures and have less preference for European and Asian manufactures.

Predictive validity

Herche (1992) has some remarkable notes about the predictive validity of the CETSCALE. He assumes that the predictive validity of the CETSCALE is product specific (the predictive validity for cars is higher than computers). Herche (1992) gives several explanations for this phenomena. First, when buying a more expensive product, people tend to have more ethnocentric tendencies because these transactions have a bigger economic impact. Another possibility is that if there are no acceptable domestic goods available, even ethnocentric consumers are being forced to purchase foreign goods. Also, the degree of involvement can be responsible for the difference between product categories. Purchases with a higher involvement can evoke emotional reactions about the country origin. However, this does not have to be true because the involvement of a purchase of a car does not has to be bigger than the purchase of a computer.

Witkowski (1998) assumes that the predictive validity of the CETSCALE does not has to be product specific but also country specific. He has found this assumption in his study in Mexico and Hungary. Respondents in this study are asked to give nine durable products in the past three years. There was a significant negative relationship between the CETSCALE-scores and the purchase of a foreign car, television, video recorder and washer in Hungary. No significant relationship was found between CE and the purchases in Mexico.

2.3.4 CETSCALE in this research

The internal consistency of the scale is high in most of the studies and meets the minimum requirements of the Cronbach's alpha value of 0.60. Orth & Fibrasova (2003) have done studies of the role of CE regarding food products. To measure CE, they make use of the CETSCALE. In this research the shortened 10-item scale of Shimp & Sharma (1987) will be used. The main reason the use the 10-item scale is because the length of the questionnaire has to be limited.

In paragraph 3.3, sufficient evidence has found for the international validity of the CETSCALE. Thus, the CETSCALE is an useful scale to measure CE in China.

3. China's consumer market

Due to the globalization of markets, there is an abundance of foreign goods in China. Domestic companies had to face strong competition of foreign companies during the last ten years. According to Cui (1999), the Chinese consumer market has the largest growth opportunity in the world. The market of 1.3 billion people, with an emerging spending power of the middle income class, is an enormous opportunity for producers of consumer goods. Each year 10 million new Chinese consumers enter the market. In 2010, China's consumer market was approximately worth $1.7 trillion. According to Forbes (2011) the domestic consumer market in China could grow to about $15 trillion within ten years. A.T. Kearney (2007) assumes that the middle class will increase its consumptions especially on food, the branded food product market supposed to increase from $150 billion to $650 billion by 2017. The statistics show that the mediocre consumer has no income tax, free education and no mortgage payment, which allows the consumer to save up to 40% of their annual income. Their annual income is approximately $11.000. As a result of the increasing average income of the Chinese middle class, they will be able to purchase more foreign products who are known about their famous brand. Rein (2009) stated that Chinese consumers trusted foreign brands more than domestic ones because they tend to believe that foreign companies would put more effort in their production and quality processes.

Although the Western companies have made some profit of the consumption growth, domestic Chinese companies are improving the quality of their products and processes. The Western companies in China now have to face local competition. According to Ewing et al. (2002) Chinese brands are rapidly emerging, which includes increased market share and advertising expenditures. Chinese consumers seem to value the emerge of Chinese brands, who now have the same high quality as foreign products but with a cheaper price. Thus, this concern is threatening the established position of foreign firms in China. Tucker (2006) issued that the foreign companies in fact have helped the Chinese firms to develop higher quality products and thus created future competition. An example is IBM, who outsourced his production to Lenovo, a Chinese company founded in 1984. Lenovo had to produce computer parts for IBM, but started an own company in China and is now capturing 25% of the Chinese computer market. The rise of the Chinese brands give the Western firms a new challenge to understand the consumer behavior of Chinese consumers. The emerge of domestic brands will continue as outlined in the 12th Five Year Plan. In this Five Year plan, the focus is on strengthen the social safety net, boost wages, increase consumer product imports, promote urbanization and optimize the consumer market for consumption, which will help the Chinese consumer to a higher spending consumption. The grow would lead to domestic consumption growth instead of relying on the investments, which now still is the highest contributor of the GDP growth as seen in figure 2.

Figure 2: China Domestic Demand (in % of GDP)

Source: CEIC Data Company Ltd and IMF staff calculations

China's import of consumer goods is still marginal, it account for 2 percent of the global consumer goods imports. China's global consumer goods import has decreased since 1995 as its share in global consumption, seen in figure 3. Chinese consumers have given more emphasis on buying domestic produced goods.

Figure 3: China Share in Global Consumption versus Share in Global Consumer Goods Imports (In percent of total)

Sources: United Nations, Comtrade database and IMF staff calculations

According to a survey of Forbes (2011), the most of the survey respondents believe that Chinese local brands will have an advantage towards global well-known brands, as seen in figure 3. The penetration into the Chinese market has not been an easy job for foreign companies, because Chinese consumers are very loyal to their domestic brands. Yet, as the middle income class rises, more emphasis will be put on purchasing imported products because Chinese consumers perceive these as higher quality and safe. Domestic consumption and its ratio of GDP decreased to 36 percent in 2009, against 45 percent in 2000 (CNBS, 2009).

Figure 4: Chinese brands will have an advantage over global brands within China

Source: Forbes Insight 2011

2.3.1 Demographics

China has the largest population in the world. According to the World Fact Book (2009), China's has a population of 1.34 billion people and accounts for about one fifth of the world's population (Table 2). The population is divided in urban and rural areas, as 43% of the total population live in the urban regions. There are about 65 million young urban consumers who are aged from 20 to 39. The young consumers in the urban areas are influenced by the Western society and are therefore a target group for Western companies. The most dense cities are Shanghai and Beijing, which have more than 5 million inhabitants.

The one-child policy has led to an unequal ratio of male and female in China. A research of McDonald (2009) identified that in 2005 there were 32 million more males than females who were younger than 20 years. He issued that this unequal ratio will remain in the next two decades. Also, because of the one-child policy, the population in China is aging. The average age in China in 2009 is 34.1 years.

Age Range

% of Population

# of Males

# of Females

0-14 years

19.8

140,877,745

124,290,090

15-64 years

72.1

496,724,889

469,182,087

65 years and over

8.1

51,774,115

56,764,041

Table 2: Population Breakdown by Gender and Age Group, 2009

Source: The World Fact Book, China (June 2009), estimates for 2009

Although China faces a male-oriented society, Chinese women are a significant target group for firms. Research of Zamcheck (2008) have shown that Chinese women hold 60% of the spending power and 78% of the daily consumptions in households. Rein (2009) identified that Chinese mothers have the tendency to spend about 20% more for food products, clothing and items for babies, knowing that these products are from high quality and safe. Also, the middle-class consumers prefer to pay more for food products that are safe (A.T. Kearney, 2007). This issue will gain more intention, because of the 2008 food safety incidents in China.

Chinese consumers tend to spent much more of their income on food (36%), transportations and communications (14%) and cultural and recreation services (13%) than on clothing (10%), residence (10%), household facilities, articles and services (6%), health care and medical services (7%) and miscellaneous goods and services (4%) as seen in figure 5.

Figure 5: Average per Capita Annual Expenditure of Urban Households, 2007

Source: China Statistical Yearbook 2008

Due to China's rapid economic growth, consumers have changed its consumption patterns. According to Pingali (2004) and Gale (2003) Chinese consumers are consuming more poultry, eggs, dairy products, fruits, vegetables and fish than before, also seen in figure 6.

Figure 6: Per Capita Annual Purchases of Major Commodities by Urban Households (kg), 1990 and 2007

Source: China Statistical Yearbook 2008

4. Conceptual framework Consumer Ethnocentrism

In this chapter, the conceptual framework will be developed. The factors, effects and moderators on CE regarding food products will be discussed. The basic model van Sharma, Shimp & Shin (1995) will be used. Many studies have used the conceptual model of Sharma, Shimp & Shin (1995) and in some of those studies the model has been modified or new concepts have been added. First, in 4.1 the conceptual model of Sharma, Shimp & Shin (1995) will be discussed. Second, the demographic variables will be explained in 4.2. Further, the effects of CE will be described in 4.3. The moderators will be discussed in 4.4, that are influencing the relationship between Consumer Ethnocentrism and the effects. At last, the final adjusted conceptual model will be presented in 4.5.

4.1 Basic framework of Sharma, Shimp & Shin (1995)

According to Sharma, Shimp & Shin (1995) ethnocentric tendencies are being caused by a set of social-psychological and demographic influences. The framework of the researchers show that Consumer Ethnocentrism is the focal concept that is related to demographic variables (age, gender, education) and social-psychological variables (openness to foreign cultures, patriotism ,collectivism and conservatism). Figure 7: basic framework of Sharma, Shimp & Shin (1995)

The effect of Consumer Ethnocentrism is to overestimate domestic products and underestimating foreign goods. Consumers tend to favor domestic products and feel obligated to purchase the products produced in own country. Thus, Consumer Ethnocentrism is decisive for the attitude towards product import. In their research, they also identify two moderators. The first moderator is the expectation that CE will have a stronger effect on the product attitude towards product import when these products are perceived as not necessary. Second, Sharma, Shimp & Shin (1995) assume that the relationship between CE and product attitude varies because of the perceived economic threat: people expect that CE has a strong influence on the attitude towards product import if the consumer perceive the imported products as a threat to his own well-being or the wealth of the their national economy.

Sharma, Shimp & Shin (1995) have tested this conceptual framework in Korea. Findings have shown that the correlation between openness to foreign cultures and Consumer Ethnocentrism is negative. People who are more open to foreign cultures are less ethnocentric. The correlation between patriotism and conservatism and CE is positive. The demographic variable 'gender' is also significant correlated with Consumer Ethnocentrism: women have more ethnocentric tendencies than men. Furthermore, higher educated people are less ethnocentric according to their research. There has been no correlation found between age and Consumer Ethnocentrism in their research (Sharma, Shimp & Shin, 1995).

The study of Sharma, Shimp & Shin (1995) have shown that the impact of CE on the product attitude of imported goods varies with the moderators 'perceived product necessity' and 'economic threat'. First, when a good is perceived as less necessary, the impact of CE on the attitude towards imported goods increases. Second, if consumers perceive imported products as a threat to the personal and economic well-being, they are less likely to purchase imported products.

4.2 Demographic variables

Studies have shown that demographic and social-economic factors are essential in cultural and ethnic aspects of consumer behavior (Sheth, 1977). Some individuals perceive importing foreign goods as a threat to his quality of life and economic wealth and some individuals see this as a benefit for their national economy. Therefore, Shimp & Sharma (1987) state that the degree of Consumer Ethnocentrism varies through different populations with different demographic characteristics.

4.2.1 Gender

Sharma, Shimp & Shin (1995) have found a positive correlation between women and Consumer Ethnocentrism and a negative relation between men and CE in the United States. Women are more conservative and tend to be behave more patriotic than men (Han, 1988). Balbanis, Mueller & Melewar (2002) identified that women in Turkey have more ethnocentric tendencies than men but they did not find the same results in Czech. Kaynak & Kara (2002) have found the same evidence in Turkey but not in Azerbeidzjan. Vida & Fairhurst (1999) have found a positive correlation between women and CE in Czech, Estonia, Hungary and Poland. Klein & Ettenson (1999) have identified the same positive correlation in the United States. Also Good & Huddleston (1995) have identified in their research that women in Poland have a positive correlation with CE, but not in Russia. These studies have shown that results differ in different countries. Therefore my second hypothesis is:

H1: Ethnocentric tendencies of Chinese female consumers are higher than Chinese men.

4.2.2 Age

According to the model of Sharma, Shimp & Shin (1995), the correlation between age and CE is not significant but older people tend to show more conservative and patriotic behaviour. Older people also have more experienced conflicts with other countries. Balbanis, Mueller & Melewar (2002) have found a positive correlation with CE and age in Turkey. They have done the same research in Czech and here, they did not find any correlation. Vida & Fairhurst (1999) did found a positive correlation in Czech, Estonia, Hungary and Poland. Klein & Ettenson (1999) on the other hand, did not find a significant correlation in Czech and the United States. Good & Huddleston (1995) found a positive correlation in Czech but older people in Russia did not had more ethnocentric tendencies. Thus, no significant correlation is found in Russia between CE and age. Witkowski (1998) did his research in Mexico and Hungary and found a positive correlation between CE and age. Steenkamp (1993) found a positive correlation in Belgium Spain, Greece and Great Britain. Also, Orth & Fibrasová (2003) confirmed a positive correlation with higher age and CE. Research have shown that there are different outcomes between countries concerning CE and age. Although there has been several studies that there is a positive correlation between age and CE, this correlation that older consumers, have more ethnocentric tendencies, cannot be generalized. Therefore the second hypothesis is:

H2: Ethnocentric tendencies of Chinese consumers of higher age groups is expected to be positive correlated.

4.2.3 Education

Sharma, Shimp & Shin (1995) have found that there is a negative correlation between education and CE. The higher educated a person, the less conservative he or she is (Ray, 1990). Higher educated people are likely to have a positive attitude towards imported goods because they don't have ethnic prejudices (Watson & Johnson, 1972). Klein & Ettenson (1999) and Good Huddleston (1995) have confirmed this correlation. Witkowski (1995) did find a negative correlation in Mexico but not in Hungary. Balbanis, Mueller & Melewar (2002) did not find a significant correlation between education and Consumer Ethnocentrism. In general, we can't conclude that higher education consumers have lower scores on the CETSCALE. This leads to the third hypothesis:

H3: Education and ethnocentrism are negatively related for Chinese consumers

4.4 Effects of Consumer Ethnocentrism

In this paragraph several potential effects of Consumer Ethnocentrism will be discussed. In the framework of Sharma, Shimp & Shin (1995), Consumer Ethnocentrism leads to overestimating domestic products and underestimating foreign imported products. People feel obligated to buy domestic products and have a preference for these products. Thus, Consumer Ethnocentrism is a determinant factor for the product attitude of importing foreign goods. Another factor that is an effect of CE is the purchase intention.

4.4.1 Product attitude

Sharma, Shimp & Shin (1995) find a negative correlation between Consumer Ethnocentrism and the attitude towards importing products. Balabanis & Diamantopoulous use a 'multidimensional unfolding approach', which visually examines the preference of two objects (domestic and foreign goods). The authors have made a separation between 'preference for domestic products' which is called the home-country bias and 'preference for foreign products' which is called the country-of-origin effects. Balabanis & Diamantopoulos (2004) conclude that Consumer Ethnocentrism is able to explain the preference for domestic products. The attitude towards Western brand personality has between investigated by Supphellen and Grønhaug (2003) in Russia. Results show that brand personality dimensions have no influence on brand attitude for ethnocentric people. People who do not have ethnocentric tendencies, dimensions of brand personality does have an effect on the attitude toward Western brands. Thus, low ethnocentric consumers are being influenced by brand personalities of foreign brands.

4.4.2 Purchase intentions

In the research of Good & Huddleston (1995) there has been looked if there is a relationship between Consumer Ethnocentrism and the purchase intentions of a domestic product. The study is done in Russia. They found a positive correlation in Russia. Russian consumers prefer buying Russian clothing than German and American. Wu, Zhu & Dai (2010) have investigated if CE effects the purchase intention of consumers in China. They found a significant relationship between CE and the purchase intention in China with the Ajzen and Fishbein's attitudes model. Therefore, there can be concluded that CE does influence the purchase intention of consumers.

4.5 Moderators

In this paragraph the most significant part of this study will be discussed. Which moderators are influencing the relationship between CE and the effects of CE. Sharma, Shimp & Shin (1995) have found two moderators in their framework: 'product necessity' and 'economic threat'. They conclude that both moderators are both significant and that the relationship between CE and attitude towards import products is being influenced. However, some studies did not found this significant relationship.

4.5.1 Product necessity

Sharma et al. (1995) identified that perceived product necessity moderated the effect between ethnocentric behavior and attitude towards imported products. Sharma et al. (1995) assumed that when consumers tend to find a product quite necessary, they are more willing to accept the imported products.

Balabanis & Diamantopoulos (2004) have researched if the relation between CE and attitude towards foreign goods varies by product category. The product categories that they have investigated are: cars, food, television, clothes, toys, do-it-yourself products and furniture. According to their research consumers tend to favor domestic food and do -it-yourself products than the other product categories. This lead to the fourth hypothesis:

H4:Perceived necessity of foreign goods moderates the effect of ethnocentrism on attitude toward imports for Chinese consumers

4.5.2 Economic threat

Sharma, Shimp & Shin (1995) divide the moderator 'economic threat' into 'personal threat' and 'domestic threat'. When consumers perceive imported products as a bigger threat for his personal and economic wealth, CE has more influence on the resistance against imported products. Sharma, Shimp & Shin (1995) conclude that 'economic threat' has a significant role on the effects of CE.

4.5.3 Anonymity as moderator

Herche (1994) assumed that future research could show that ethnocentric tendencies will increase due to face-to-face interaction with sellers on public places, such as convenient stores. The possibility to buy products anonymously, such as placements orders by phone, can develop a situation where ethnocentric consumers are being approached by the importers. If consumers can buy foreign products anonymously (such as the internet or a mail-order firm, there will be a negative relationship between Consumer Ethnocentrism and the product attitude towards foreign goods. Therefore, the last hypothesis is:

H5: De relationship between Consumer Ethnocentrism and product attitude towards foreign goods is less negative if there is a possibility to buy products anonymously.

4.6 Final conceptual model

Moderators

Perceived necessity

Purchase anonymity

Attitude towards imports

Purchase intention

Demographics

Age

Gender

Education

Consumer Ethnocentrism tendencies

Figure 8: Adjusted conceptual model

5. Methodology

In this following chapter, the methodology will be applied in order to look if current Chinese consumers have strong ethnocentric tendencies toward food products. The conceptual model that is shown in figure 9, will be tested empirical to look if the variables, moderators and outcomes are applicable for Chinese consumers. In this chapter first, there will be an operationalization of the model. Second, there will be an explanation of the pre-test. Third, the results of the empirical research will be discussed. All SPSS-outcomes can be found in the appendix (results of the representation of the respondents, results of the scales, results of the relationship between CE and the outcomes of CE and results of the effects of the moderators).

Consumer ethnocentrism

The CETSCALE is a reliable scale to test the consumer ethnocentrism in China, as discussed in the literature review (Lindquist et al. 2001; Klein et al, 2006). In this research, I will select ten items of the CETSCALE of Shimp and Sharma (1987) to measure CE. The scale will be a Likert-scale with 7-points, where respondents have to answer in ''Totally disagree = -3" to ''Totally agree= +3". This scale will consist with statements regarding consumer ethnocentrism.

Totally disagree

Disagree

Disagree somewhat

Undecided

Agree somewhat

Agree

Totally agree

CETSCALE

-3

-2

-1

0

+1

+2

+3

A real Chinese should always buy Chinese-made food products

Foreigners should not be allowed to put their food products on our markets

Chinese people should always buy Chinese-made products instead of imports

Only those products that are unavailable in China should be imported

Buy Chinese-made products. Keep the Chinese economy working

It is not right to purchase foreign products, because it puts Chinese people out of jobs

We should purchase products manufactured in China, instead of letting other countries get rich off us

There should be very little trading or purchasing of food products from other countries unless out of necessity

It may cost me in the long-run but I prefer to support Chinese food products

Chinese consumers who purchase products made in other countries are responsible for putting their fellow Chinese people out of work

Table 3: CETSCALE of Shimp and Sharma (1987)

Demographic variables

In this questionnaire the respondents will be asked to answer three questions to provide the demographic factors: "Gender", "Education'' and ''Age".

Demographic variables

Gender

Education

Age

Table 4: Items demographic variables

Effects of Consumer Ethnocentrism

To measure the effects of Consumer Ethnocentrism, it is necessary to choose products that are imported to China. Different studies have shown that the chosen products play a significant role. Archaya & Elliot (2003) have shown that Australian consumers prefer Japanese cars and American jeans over domestic products. Australian consumers only prefer pineapple from Australia when choosing domestic goods over foreign goods. According to the research, Australian consumers have more trust in their food products, produced in own country. Balabanis & Diamantopolous (2004) also assume that the preference for domestic food products is higher than other product categories like cars, televisions, beauty products, clothes, toys and furniture. Therefore, food products is chosen in this research: 'vegetables, 'yoghurt', 'beer' and 'infant milk'. No research is done in this product category.

Attitude towards imported products

To measure the product attitude towards imported products the respondents will be asked if they think that 'beer', 'vegetables', 'infant milk' and 'yoghurt' should be imported to China. In this research, the product attitude will be measured on a scale of 1 (Strongly disagree) to 5 (Strongly agree).

Product attitude

Strongly disagree

Disagree

Neither

Agree

Strongly Agree

Beer should be imported to China

Vegetables should be imported to China

Infant milk should be imported to China

Yoghurt should be imported to China

Table 5: Items Product attitude

Purchase intention

Imports purchase intention is being measured using a five-point Likert-type scale ranging from 1 (definitely would not buy) to 5 (definitely would buy) for the four food products categories (Beer, vegetables, infant milk and yoghurt).

Purchase intention

Definitely would not buy

Would not buy

Neither

Would buy

Definitely would buy

What is the likelihood that you would buy beer from a foreign country?

What is the likelihood that you would buy vegetables from a foreign country?

What is the likelihood that you would buy infant milk from a foreign country?

What is the likelihood that you would buy yoghurt from a foreign country?

Table 6: Items Purchase intention

Moderators

The moderator ''Anonymity" will be measured by asking the respondents how often they buy products on the internet and mail-order-company.

Anonymity

Never

1-2 times a year

3-5 times a year

5-10 times a year

More than 10 times a year

How often do you buy products through a mail-order-firm?

How often do you buy products on the internet ?

Table 7: Items Anonymity

The moderator 'Product necessity' is measured according to the model of Sharma, Shimp & Shin (1995). They asked the respondents to provide answers for three products if they perceive these products as necessary on a scale of 1 (strongly disagree ) to 5 (strongly agree.

Product necessity

Absolutely not necessary

Not necessary

Neither

Necessary

Absolutely necessary

Do you think beer is necessary in our daily life?

Do you think vegetables are necessary in our daily life?

Do you think infant milk is necessary in our daily life?

Do you think yoghurt is necessary in our daily life?

Table 8: Items Perceived necessity

5.1 Pilot test

To acquire the essential information of the respondents, a structured questionnaire has been developed (Appendix). It consists four major sections. In the first part the demographics of the survey participants will be described. The consumer ethnocentrism will be measured in the second part by using the CETSCALE, consisting of 17 items, using a 7-point Likert-scale. The third part consist the measurement of the attitude toward import products and the purchase intention. A 5-point Likert-scale will be used. The fourth part consists the measurement of anonymity by asking the participants how often they buy products through the internet and mail-order firm.

In this study, the quantitative data will be collected with pre-structured online questionnaires. The questionnaire has been translated into Chinese by a Chinese bilingual teacher. The questionnaire is then back translated by a second bilingual person. The data for this research will be collected in Shanghai which locates in the South east area of China. A total of 210 questionnaires will be delivered randomly to adult participants.

The purpose of this research will not be told to the participants. This can manipulate the results. The questionnaire has been send to twenty Fudan-University students. Focus points during this pilot test are the correctness of the questions, the amount of needed time and the evaluation of the tests. Some statements had to be translated differently because of the vague translations. The most students understood the questionnaire and no remarks were given about the length and content. This pilot study has improved the design of the questionnaire.

5.2 Demographic characteristics

The characteristics of the respondents with a total number of 208 questionnaires are shown in table 9. As it can be seen, the sample shows an average age of the respondents of 35 years. There is a large number of respondents in the age range of 20-39 years. The percentage of this age category is 76.5%. During the research, there were more men than women, who cooperated on this study. When it comes to education, the most of the respondents had a high education level. The percentage of the Undergraduate and Master or above are 80.4%.

Variable

Frequency

Percentage

Gender

Male

53.1%

Female

46.9%

Age

20-29

42.5%

30-39

34%

40-49

12%

51-58

2.4%

60-69

3.4%

70-79

5.7%

Education

Elementary school

0.5%

High school

3.3%

College

15.8%

Undergraduate

49.3%

Master and above

31.1%

Table 9: characteristics of the respondents

5.3 Results

In this paragraph, the results of the empirical results will be explained. First, the relationship between Consumer-ethnocentrism and the demographic variables will be described. Second, the relationship between CE and the effect of CE will be explained. Third, the effect of the moderators will be analyzed. At last, there will be a summary of the tested hypotheses.

5.3.1 Reliability of the scales

It is essential to determine the reliability of the questions in the questionnaire, whether they measure the constructs correctly. The internal consistency of the questions has to be significant to measure the aspect. As mentioned in paragraph 3.2, the Cronbach's Alpha is an indicator to measure the internal consistency. A low alpha (alpha ≤ 0.60) indicates that the scale is not internal consistent (Hair et al. 1998). When this is the case, respondents do not have a consistent image of the aspect that the questions are supposed to measure. Thus, the indicator is not reliable. The scale can be interpreted as internal consistent when the alpha is higher than 0.60. The questions can be seen as reliable indicators of the underlying aspect.

5.3.2 Measuring the hypotheses

To measure the relationship between the demographic variables and Consumer Ethnocentrism the correlation analyses, t-test and the one-way ANOVA can be used (Dalen & Leede, 2009). The relationships between the effect of Consumer ethnocentrism (product attitude) and CE can also be measured by correlation analyses. The moderating effects of the relationship (CE and the effects) can be measured by a regression analysis. In this regressions analysis, the dependent variable is ' product attitude toward imported products' and the independent variables are 'Consumer Ethnocentrism' and 'the interaction-effects between CE and the moderators'.

5.3.3 Testing the scales

In this paragraph, the descriptive statistics of the scales will be described. To look if the scales are internal consistent, the Cronbach Alpha of the scales will be determined.

CETSCALE

In table 10, the results of the average scores per item is shown. The factor that scored highest in the scale is item 5, namely '' Buy Chinese-made products. Keep the Chinese Economy working''. This high score identifies that Chinese people should buy Chinese-made products. They are likely to agree that they should help the Chinese economy. The average score is 3.55 with a standard deviation of 0,373, which means that Chinese consumers have a moderate degree of ethnocentrism. The scores for items 2,3 and 9 are more than the total average score. Items 10, 4 and 7 ('Chinese consumers who purchase products made in other countries are responsible for putting their fellow Chinese people out of work ', 'Only those products that are unavailable in China should be imported ', 'We should purchase products manufactured in China, instead of letting other countries get rich off us') have the lowest scores. These low scores of the items show that Chinese people do not think that people who buy import products are causing China a higher degree of unemployment.

The Cronbach's Alpha of the ten items is 0.911. Which indicates that the scale is internal consistent.

CETscale items

Mean

Corrected Item-Total Correlation

Cronbach's Alpha if Item deleted

1

A real Chinese should always buy Chinese-made food products

3.45

0.582

0.908

2

Foreigners should not be allowed to put their food products on our markets