Effects of Partitioned Country of Origin Information on Buyer Assessment of Binational Products

Dongdae Lee, Tongmyung University of Information Technology
SangWook Bae, Dongseo University
ABSTRACT - This study deals with the effects of partitioned country of origin, brand image, and interaction between them on consumer evaluation of binational brands. Country of origin was divided into three subconstructs: country of manufacture product specific image (Cmpi), country of manufacture overall image (Cmoi), and country of brand overall image (Cob). Relationships between these three subconstructs of country of origin and brand image were hypothesized based on previous studies and categorization theory. Data were analyzed via structural equation models using Lisrel 8. To deal with an interaction term in the model, Ping’s (1995) Aparsimonious@ approach was employed.
[ to cite ]:
Dongdae Lee and SangWook Bae (1999) ,"Effects of Partitioned Country of Origin Information on Buyer Assessment of Binational Products", in NA - Advances in Consumer Research Volume 26, eds. Eric J. Arnould and Linda M. Scott, Provo, UT : Association for Consumer Research, Pages: 344-351.

Advances in Consumer Research Volume 26, 1999      Pages 344-351

EFFECTS OF PARTITIONED COUNTRY OF ORIGIN INFORMATION ON BUYER ASSESSMENT OF BINATIONAL PRODUCTS

Dongdae Lee, Tongmyung University of Information Technology

SangWook Bae, Dongseo University

ABSTRACT -

This study deals with the effects of partitioned country of origin, brand image, and interaction between them on consumer evaluation of binational brands. Country of origin was divided into three subconstructs: country of manufacture product specific image (Cmpi), country of manufacture overall image (Cmoi), and country of brand overall image (Cob). Relationships between these three subconstructs of country of origin and brand image were hypothesized based on previous studies and categorization theory. Data were analyzed via structural equation models using Lisrel 8. To deal with an interaction term in the model, Ping’s (1995) "parsimonious" approach was employed.

A binational brand is now a ubiquitous phenomenon. Consumers commonly purchase General Electric refrigerators made in Mexico, Sony telephones manufactured in Malaysia, and Honda Accord sedans assembled in America. These are brands manufactured outside the country that originally manufactured the brand (Chao 1993; Ettenson and Gaeth 1991; Han and Terpstra 1988).

Proliferation of such binational brands requiresa new perspective in dealing with country of origin (Coo) concept. Traditionally, Coo concept was treated as equivalent to the "made in" concept (Head 1988; Thorelli, Lim, and Ye 1989). Studies dealing with Coo focused on images of manufacturing countries. Binational brands, however, brought us to consider a new concept called "country of brand (or Cob)". Cob means country image derived from the country in which the brand name originated. For a binational brand, a country from which the brand originate is no longer the country in which the brand is manufactured. In assessing a binational brand, consumers may utilize image of the brand’s country. Suppose there are two unfamiliar brands, and one brand is originated from Japan and the other originated from Philippine. If all others are the same, most consumers would develop more positive initial perceptions toward the Japanese brand than the Philippine brand. This suggests that Cob might have an effect on the image of brand. While Cob is a relatively new concept in the area of Coo studies, an analogous concept, country of design, has been explored by Chao (1993).

Besides Cob, another subconcept needs to be further partitioned out from Coo. Coo has been commonly defined and measured at a specific product class level (Chao 1993; Cordell 1992; Eroglu and Machleit 1989; Han 1989; Han and Terpstra 1988; Hooley, Shipley and Krieger 1988; Lawrence, Marr, and Prendergast 1992; Thorelli, Lim, and Ye 1989; Wall, Liefeld, and Heslop 1991). Roth and Romeo’s (1992) definition is a good example. In their study, they define Coo as "the overall perception consumers form of products from a particular country, based on their prior perceptions of the country’s production and marketing strengths and weaknesses (Roth and Romeo 1992, p. 480)." Studies based on this approach always employ certain products or product categories and measure Coo at the level of specific product categories (We call this Cmpi or Country of Manufacture Productspecific Image).

However, this traditional approach of Coo covers only a limited domain. A more general facet of country image remain uncovered(Johansson 1989; Parameswaran and Yaprak 1987). Consider an Iraqi carpet from the viewpoint of American consumers. While American consumers may evaluate manufacturing ability of carpets by Iraq very high, they may not evaluate the country or people of Iraq as high. In assessing an Iraqi carpet, consumers may be affected not only by the carpet manufacturing ability by Iraq, but also by the overall country image of Iraq. German automobiles assessed by Israeli buyers can be another example (Johansson 1989). To deal with this rather basic yet more general domain of Coo, a group of researchers suggests measuring Coo at the overall country image level (Hooley, Shipley, and Krieger 1988; Lawrence, Marr, Prendergast 1992; Parameswaran and Yaprak 1987). Definition of Coo at the overall country image level (We call this Cmoi or Country of Manufacture Overall Image) is well illustrated by Bannister and Saunders (1978): " generalized images created by variables such as representative products, economic and political maturity, historical events and relationships, traditions, industrialization and the degree of technological virtuosity will have effects upon consumer attitudes ". This approach does not tie Coo effect with a specific product or products produced in a specific country. Rather, it implies general images such as cultural, economical, or political images associated with a country have effects on consumer evaluation of products made in the country.

Along with the issue of types of country image, relationship between brand image (Bi) and Coo (more specifically Cmpi) requires a new spotlight. Proliferation of binational brands made roles of brand image and Cmpi, in consumer evaluation of binational brands (Ebb), one of the foremost concerns of many multinational companies (Ettenson and Gaeth 1991). According to a Wall Street Journal report, Chrysler dealers refused to carry Plymouth and Dodge K-Cars manufactured in Mexico because they feared American consumers will notpurchase the brand once they realized that the cars are manufactured in Mexico (Nag 1984). Due to such a possible consumer backlash, multi-national companies take various precautionary approaches to protect their established brand image when they determine a sourcing country (Johansson and Nebenzahl 1986; Johannson and Thorelli 1985). However, relationships between Brand image and Cmpi in consumer evaluation of a binational product remain unclear (Ettenson and Gaeth 1991). While Han and Terpstra (1988) and Wall, Liefeld, and Heslop (1991) reported that Cmpi effect is stronger than brand image, Cordell (1992) reported that effect of brand image could be stronger or weaker depending on product types. More investigations are required for the interaction relationships between brand image and Cmpi.

This study has two major objectives. The first is to explore effects of three different types of Coo: Cob, Cmoi and Cmpi. As explained, country image cannot be nor should not be confined as a single dimensional concept (Chao 1993). A new conceptual framework which can include and explain roles of these three types of Coo is needed. Roles of Cmoi and Cob on consumer evaluations of binational brands in the context of brand image and Cmpi are largely unknown. The second objective is to test relationships between brand image and Cmpi in consumer evaluation of binational brands. Despite importance of relationships between brand image and Cmpi, studies on this issue are rare and roles of these two are still unclear.

LITERATURE REVIEW

Brand name is an image or extrinsic variable which works as summary statistics in consumer decision making (Erickson, Johansson, and Chao 1984; Han 1989; Huber and McCann 1982). As an extrinsic or image variable, brand name is commonly utilized as a proxy of quality evaluation or some other overall evaluations of a product whenever consumers do not have any specific ideas about the product (Jacoby, Olson, and Haddock 1971; Szybillo and Jacoby 1974). According to categorization theory, attitudes toward a stimulus are directly related to attitudes associated with the activated category (Alba and Hutchinson 1987; Fiske 1982; Sujan 1985). When consumers face a brand name associated with positive image, consumers will infer positive attitudes toward the brand. Even though a product is made in a different country to the one originally associated with the brand name, consumers may regard the product as belonging to the brand. Therefore, when the image generated from a brand name (i.e., brand image or BI) is good, consumer evaluation of a binational brand (Ebb) will be favorable. When brand image is bad, consumer evaluation of a binational brand (Ebb) will be unfavorable.

H1:  Brand Image (Bi) has a direct positive effect on (Ebb).

Like brand name, Coo is considered as an image or extrinsic variable. Thus, Coo serves as a proxy of quality evaluation for consumers (Han 1989; Johansson 1989). Consumers are willing to pay premium prices for a product made in a specific country such as French perfumes, English china, German machineries, Italian fashions, and so on (Kapferer 1994). Effect of Coo on consumer evaluation of a binational brand is generally tied with image levels of the related country. In other words, when a country name in the Coo label is associated with good image, consumer evaluation of a binational brand becomes favorable. When a country name in the Coo label is tied with poor image, consumer evaluation of a binational brand becomes unfavorable (Cordell 1991; Roth and Romeo 1992; Wall, Liefeld, and Heslop 1991). In short, effects of Coo on consumer evaluation of a binational product are positive.

As discussed early, Coo is divided into three subconstructs: Cmoi which represents overall image of manufacturing country, Cmpi which represents specific product related image of manufacturing country, and Cob which represents overall image of brand country. Because Cmoi and Cmpi are concepts derived from the traditional Coo concept which deals with image of manufacturing country, effects of these two subconstructs of Coo on consumer evaluation of binational brands (Ebb) will be positive. Role of Cob on Ebb, however, requires an additional explanation because Cob is not related with manufacturing but with the origin of brand. According to Chao (1993), image of country of design has a positive effect on consumer evaluation of binational products (Ebb). Although country of design is not the same as Cob, they may overlap significantly in domains because many binational brands are designed in the country of brand. Therefore, it may be safe to assume that Cob’s role is similar to the role of country of design. Consequently, all three subconstructs of Coo are hypothesized to have a direct positive effect on consumers evaluation of binational brands (Ebb).

H2:  Cmoi has a direct positive effect on Ebb.

H3:  Cmpi has a direct positive effect on Ebb.

H4:  Cob has a direct positive effect on Ebb.

Relationship between Cmoi and Cmpi is similar to the one between a basic category and a subcategory. While Cmoi represents consumers’ categorical knowledge of a country at a basic and primary level, Cmpi represents consumers’ categorical knowledge at a specific product level. According to categorization theory, knowledge at the basic level category affects knowledge at the specific subcategory (Mervis and Rosch 1981; Tversky and Hemenway 1984), and when facing a categorical cue, consumers retrieve categorical information at the basic level before categorical information at the subordinate level (Lee 1994). Therefore, if a country name stimulates consumers retrieval of information from memory, consumers will retrieve general information about the country (Cmoi) first and specific product related information (Cmpi) later. Also, if a consumer retrieves a positive Cmoi upon facing a country name, it will affect positively to the development of specific product related image of the country. In case of the retrieval of a negative Cmoi from a country name, however, product related image (Cmpi) will be negatively affected. In short, relationship between Cmoi and Cmpi is a direct positive one. Since both Cmoi and Cmpi are regarded as having direct positive effects on Ebb and Cmoi has a direct positive effect on Cmpi, Cmpi can be regarded as a mediating variable.

H5:  Cmoi has a direct positive effect on Cmpi.

H6:  Cmpi is a mediator of Cmoi’s effect on Ebb.

Although Cob and Bi relationship is between a country and a brand, it can be considered as a basic category and subcategory relationship as in the Cmoi and Cmpi case. Cob, which represents general image of a country, can be considered as a basic category while Bi, which represents specific image of a brand originated from the country, can be considered as a subcategory. In a categorical knowledge structure, therefore, Cob is likely to be formed and thus retrieved before Bi. For example, upon facing a brand name, consumers will retrieve more basic and general Cob first, which affects forming of specific Bi later. More specifically, if consumers hear that an unknown brand is from a country for which high Cob is likely to occur, image of the unknown brand will go up immediately. On the contrary, if consumers hear that the unknown brand is from a low Cob country, image of the brand will drop instantly. Consequently, Cob’s efect on Bi could be hypothesized as positive. Because both Bi and Cob have an effect on Ebb and Cob has an effect on Bi, Bi can be considered as a mediator. Based on this;

H7:  Cob has a direct positive effect on Bi.

H8:  Bi is a mediator of Cob’s effect on Ebb.

Besides direct effects, brand image and Coo may have an interaction effect on consumer evaluation of a binational product. Two studies reported significant interaction effects between Bi and Coo on consumer evaluation of a binational brand. Specifically, Han and Terpstra (1988) found significant positive interaction effects in five out of six models in their study. Wall, Liefeld, and Heslop (1991) also found a significant result out of 12 tested models in their study. It is quite likely for consumers to purchase a brand with good images than a brand with poor images because a brand with good images can give more satisfaction, self esteem, and prestige to them. Naturally, consumers will show more interests and higher involvement in brands with good images than in brands with poor images. Therefore, Coo information related to a brand with good images will be a much more important decision making factor than Coo information related to a brand with poor images. As a consequence, effect of Coo on consumer evaluation of binational products will be greater when the involved brand is associated with good images, and will be weaker when the brand image is poor. As discussed in the previous section, because Cmpi is regarded as a mediator of Cmoi’s effect on consumer evaluation of binational products, interaction between brand image and country image means interaction between brand image and Cmpi. Based on this;

H9:  Bi and Cmpi have a positive interaction effect on Ebb. Specifically, Cmpi’s effect is greater when Bi is high than when Bi is low.

In normal situation, a brand name will be one of the first inferential cues consumers employ to evaluate a product. Consumers’ utilization of Cmpi information as an inferential cue will occur following the recognition of the brand name because consumers would find Cmpi information at the back or bottom of the product after noticing the brand name of the product. In case of Sony t.v. made in China, for example, a consumer will form his or her initial affective responses toward the t.v. set the moment s/he sees the brand name "Sony". However, his or her affective responses toward the Sony t.v. set could be changed after s/he notices that the t.v. set is made in China, not Japan. This example shows, if other cues are absent or ignored, a consumer’s evaluation of a binational brand will be primarily determined by brand image rather than by Cmpi. From this viewpoint, Han and Terpstra’s (1988) finding in which Cmpi has much bigger impact on consumer evaluation of binational products needs a careful review. While the importance of brand image in consumer decision making is largely accepted without being challenged, effect of Cmpi is often questioned (Peterson and Jolibert 1995). Moreover, Han and Terpstra’s (1988) finding was based on a repeated measure design which is often criticized for a possible implication with threats of demand artifacts. Based on the above reasoning;

H10:  Bi’s effect on Ebb will be stronger than Cmpi’s effect.

Figure 1 summarizes all the proposed hypotheses.

FIGURE 1

PROPOSED MODEL OF Ebb

METHOD

Six brands of t.v. were used for the study. They were two American brands (Rca and Emerson), two Japanese brands (Sony and Sanyo), and two brands from developing countries (Goldstar from Korea and Funai from China). Brand pairs were selected to reflect high and low brand image in each country group. Brand countries were selected to represent both developed and developing countries. Manufacturing countries were determined with Japan and Thailand for American brands, America and China for Japanese brands, and Canada and Mexico for Korean/Chinese brands. Each pair of manufacturing countries was chosen to reflect high and low Cmpi condition for each brand. Countries for Cmoi and Cob were determined by countries for Cmpi and brand countries respectively. That is, if Cmpi is about the image of Japan, then Cmoi is also about the Japan. If a brand is originated from America, then Cob is about the image of America. A total of twelve different stimulus conditions was made by combining six brands with two manufacturing countries. (i.e., Rca made in Japan, Rca made in Thai, Emerson made in Japan, Emerson made in Thai, Sony made in Usa, Sony made in China, Sanyo made in Usa, Sanyo made in China, Goldstar made in Canada, Goldstar made in Mexico, Funai made in Canada, and Funai made in Mexico).

Current study deals with five constructs and their observed measures. Cob and Cmoi are both for overall country image. Based on Parameswaran and Yaprak’s (1987) measures, six item scales were developed. They were about political maturity, education level, skill level, economic development, motivation to raise living standards, and achievement of high standards. Since both Cob and Cmoi deal with overall country image, scales for each construct contain the same contents except the specific country name. For example, a statement for Cob such as 'Canada is a politically matured country’ was changed into 'Korea is a politically matured country’ for Cmoi. Items measuring Bi and Cmpi were generated from Nagashima’s (1970) and Johansson and Nebenzahl’s (1986) studies. Five items were developed. They were about reliability, workmanship, longevity, style, and performance. As in the scales measuring overall country image, the same format was applied for Bi and Cmpi. Specifically, an item such as 'Sony t.v. is reliable’ for Bi was transformed into 't.v. made in Canada is reliable’ for Cmpi. Five items measuring attitudes toward binational brands (Ebb) were borrowed from Lee’s (1994) study. They were about the likeness, favor, other people’s approval, satisfaction, and happiness. All scales were Likert type with 17 strongly disagree strongly agree categories. Cronbach alpha coefficient of each multi-item scale was uniformly high (Cmoi=.9035; Cob=.8785; Bi=.9254; Cmpi=.9070; Ebb=.9234).

A total of 231 students in a state school in the southwest region participated in the study. Subjects were randomly assigned to each of 12 stimulus conditions. Before filling out questionnaires, subjects were requested to read carefully first two pages which contain information about quality ratings of various t.v. brands. Subjects were allowed to spend 23 minutes for the reading. Questionnaire items were arranged in the order of Cob, Cmoi, Bi, Cmpi, Ebb, and demographic information. Subjects spent 10-15 minutes to fill out the questionnaire. Among 231 responses, 21 questionnaires were deleted due to omissions in responses. A total of 210 questionnaires were subsequently used for the data analysis.

TABLE 1

MEAN RATINGS OF KEY VARIABLES

FIGURE 2

MEASUREMENT MODEL OF Ebb WITHOUT AN INTERACTION TERM

Data Analysis Method

Data were analyzed via structural equation models using Lisrel 8. One problem in the analysis of the current model is dealing with an interaction term. As shown in Figure 1, an interaction term between Bi and Cmpi is involved in the current model. Analysis of an interaction term in a structural equation model generally requires complicated procedures. For example, procedures suggested by Kenny and Judd (184) and Hayduk (1987) require to calculate many dummy observed and latent variables. Unless the number of observed variables for the related main effects (latent variables) is very small, their suggested procedures could be impractical. To overcome such a problem, Ping (1995) suggested a parsimonious approach. In the current study, Ping’s parsimonious approach using Lisrel 8 was employed as a major analysis technique for the proposed model. Hierarchical regression was also utilized as a supporting analysis tool.

RESULTS

Preliminary Data Analysis

Prior to testing the model, descriptive statistics for key variables (Bi, Cob, Cmpi, and Cmoi) were calculated. Mean Bi were assessed first. Sony’s rating was the highest with 6.08, followed by Rca(5.19), Sanyo (4.18), Emerson (3.97), Goldstar (2.98), and Funai (2.41). Cob levels were high for Japan (5.60) and America (5.39), but relatively low for China (4.14) and Korea (4.14). Mean Cmpi values were also compared. Japan was the highest with 5.31, followed by America (4.99), Thailand (4.43), Canada (4.14), China (4.14), and Mexico (3.47). Order of Cmoi levels for the six countries was slightly different from that of Cmpi levels. Cmoi level of America was the highest with 5.73, followed by Japan (5.66), Canada (5.31), China (4.06), Thailand (3.98), and Mexico (3.23). Mean ratings of each key variable are summarized in Table 1.

Hypothesis Tests

To solve an interaction term (Bi*Cmpi) in the structural equation model, this study follows Ping’s (1995) parsimonious estimating technique. Ping (1995) suggested a three step approach for latent variable interactions in structural equation models. The first step is to estimate the measurement model parameters. In other words, model parameters in Figure 2 which does not contain the interaction between Bi and Cmpi should be estimated first. Table 2 summarizes all the estimated parameters (loadings and variances).

The next step is calculating the values for the loading and error variance of the interaction term using the measurement model parameter estimates. Estimating the structural model with fixing the loadings of observed variables and their error variances is the last step. Model in Figure 3 which contains the interaction between Bi and Cmpi is used in this step. Estimation results of structural equation model are summarized in Table 3. Fit indices show that fitting between data and model is less than ideal (c2=734.3, Gfi=.749, Agfi=.704, Nfi=.992). This less than ideal fit level (Gfi>.8 is generally regarded as an acceptable fit level) is probably due to the small sample size. Compared with the number of observed variables (n=28) in the model, sample size 210 was relatively small.

TABLE 2

MEASUREMENT MODEL RESULTS

Eight hypotheses were tested with the Lisrel analysis results. Tests of H2 and H4 were not included in the analysis of structural equation model due to the presence of mediating variables. All the hypotheses except one (H9) were supported. Specifically, H1 was supported (b41; t=5.98). Consumers preferred binational brands with higher brand image to those with lower brand image. H3 was also supported (b42; t=2.57). Consumer evaluations of binational brands got higher as Cmpi increased. H5 proposed that Cmoi has a positive effect on Cmpi. H5 was supported (g22; t=6.41). H6 proposed that Cmpi would be a mediator between Cmoi and Ebb. To be a mediator, Cmoi should have a weaker effect on Ebb after Cmoi’s effect on Cmpi is controlled. Cmoi’s effect on Ebb becomes weaker when Cmpi is controlled (g42; t=3.15 for H6 vs. t=5.54 for H2). H6 was supported. H7 is about the relationship between Cob and Bi. Cob had a direct positive effect on Bi (g11; t=6.70). H7 was supported. H8 proposed that Cob would mediate BI’s effect on Ebb. Because Cob_ffect on Ebb was insignificant when Bi was controlled (g41; t=1.38 for H8 vs. t=4.59 for H4), Cob was revealed as a pure mediator. H8 was strongly supported. H9 is about a positive interaction by Bi and Cmpi. H9 was not supported (b43; t=.05). H10 is about the strength between Bi and Cmpi on Ebb. Comparison of estimated coefficients revealed that Bi’s effect on Ebb (b=.64) is much stronger than Cmpi’s effect on Ebb(b=.23). H10 was supported.

H2 and H4 were tested via hierarchical regression. Both hypotheses were supported. When mediators (Bi and Cmpi) were not controlled, Cob and Cmoi had a direct positive effect on Ebb individually.

DISCUSSION AND CONCLUSIONS

This study shows that Bi and Cmpi have direct positive effects on Ebb. Subjects preferred binational brands with high brand image and/or binational brands made in countries with good image. These findings are consistent with the results of previous studies (Cordell 1992; Han and Terpstra 1988). Cmoi’s direct effect on Ebb was also identified. Although Cmoi’s effect decreased by the presence of Cmpi, its direct effect on Ebb remained. In other words, Cmpi was not a pure but a simple mediator for Cmoi’s effect on Ebb. According to this result, companies need to focus on a country with high Cmpi as well as a country with high Cmoi when deciding a sourcing country or promoting Coo of its brand to potential customers. Both factors have independent effects on Ebb.

Unlike the Cmoi’s effect, Cob’s direct effect on Ebb disappeared when the effect of Bi on Ebb was controlled. In other words, Bi was identified not as a simple but as a pure mediator between Cob and Ebb. According to this result, a firm need not build its communication strategy focusing on delivering brand country’s image. Such a strategy would be ineffective in enhancing consumer evaluation of a binational brand. A firm needs to focus on increasing its brand image instead. This study also found that brand image had a much stronger effect on Ebb than Cmpi (.63 vs. .23). While this finding contradicts to a previous report by Han and Terpstra (1988), it concurs with the finding reported by Lee and Ganesh (1997). Because brand image’s effect is stronger than Cmpi’s effect, consumer evaluation of a binational brand would be primarily determined by the brand name. This means a firm having a brand with a good image could have a smaller risk of sourcing from the less developed countries than a firm having a brand with a poor image. It warns, however, while brand image has a stronger effect on Ebb than country image does, Coo information can alienate loyal customers when the image of sourcing country is poor.

Nonexistence of an interaction effect by Bi and Cmpi is somewhat an unexpected result. Possibilities of an interaction effect by Bi and Cmpi on Ebb were raised by several researchers (Cordell 1992; Ettenson and Gaeth 1991; Han and Terpstra 1988; Lee and Ganesh 1997). Nonetheless, the current study found no evidence of any interaction effect by Bi and Cmpi. There is a possibility that previous findings about the existence of an Bi*Cmpi effect on Ebb could have been biased because most of them used regression which relies on a biased and inefficient coefficient estimation technique. More studies are necessary on this issue.

Limitations

This study has a few limitations. First, this study used students as subjects. Generalization of the finding with student subjects is limited. Second, the current study has a poor fit of the structural equation model. Compared with the number of variables, a larger size of sample was required. Third, items in the questionnaire were arranged in the fixed order of Cob, Cmoi, Bi, Cmpi, and Ebb. Thus, results are not free from possible order effects. Presentation order of variables should have been treated as an experimental factr and included in the study design. Fourth and finally, the present study has a weakness in design. A new design considering nested relationships among variables will produce more convincing results.

FIGURE 3

STRUCTURAL EQUATION MODEL OF Ebb WITH A Bi*Cmpi INTERACTION TERM

TABLE 3

STRUCTURAL MODEL ESTIMATION RESULTS

TABLE 4

RESULTS OF REGRESSION ANALYSIS

REFERENCES

Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411454.

Bannister, J. P. and J. A. Saunders (1978), "United Kingdom Consumers’ Attitudes towards Imports: The Measurement of National Stereotype Image," European Journal of Marketing, Vol. 12 (8), 56270.

Bilkey, Warren J. and Erik Nes (1982), "Country of Origin Effects on Product Evaluations," Journal of International Business Studies, Vol. 13 (Spr/Sum), 8999

Chao, Paul (1993), "Partitioning Country of Origin Effects: Consumer Evaluations of a Hybrid Product," Journal of International Business Studies, Vol. 24 (2), 291306.

Cordell, Victor V. (1992), "Effects of Consumer Preferences for Foreign Sourced Products," Journal of International Business Studies, Vol. 23 (2), 25169.

Erickson, Gary M., Johny K. Johansson, and Paul Chao (1984), "Image Variables in MultiAttribute Product Evaluations: Country of Origin Effects," Vol. 11 (September), 69499.

Eroglu, Sevgin A. and Karen A. Machleit (1989), "Effects of Individual and ProductSpecific Variables on Utilizing Country of Origin as a Product Quality Cue," International Marketing Review, Vol. 6 (6), 2741.

Ettenson, Richard and Gary Gaeth (1991), "Commentary: Consumer Perceptions of Hybrid (Binational) Products," Journal of Consumer Marketing, Vol. 8 (4), 1318.

Fiske, Sujan T. (1982), "Schematriggered Affect: Applications to Social Perception," in Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, M. Clark and S. Fiske eds., Hillsdale, New Jersey: Erlbaum.

Han, C. Min (1989), "Country Image: Halo or Summary Construct?,"Journal of Marketing Research, Vol. 26 (May), 222-229.

Han, C. Min and Vern Terpstra (1988), "CountryofOrigin Effects for Uninational and Binational Products," Journal of International Business Studies, Vol. 19 (Summer), 23555.

Hayduk, Leslie A. (1987), Structural Equation Modeling with Lisrel: Essential and Advances, Baltimore, Maryland: Johns Hopkins Press.

Head, David (1988), "Advertising Slogans and the "Made-in" Concept," International Journal of Advertising, Vol. 7, 237-252.

Hooley, Graham J., David Shipley, and Nathalie Krieger (1988), "A Method for Modelling Consumer Perceptions of Country of Origin," International Marketing Review, Vol. 5 (Autumn), 6776.

Huber, Joel and John McCann (1982), "The Impact of Inferential Beliefs on Product Evaluations," Journal of Marketing Research, Vol. 19 (August), 32433.

Jacoby, Jacob, Jerry C. Olson, and Rafael A. Haddock (1971), "Price, Brand Name, and Product Composition Characteristics as Determinants of Perceived Quality," Journal of Applied Psychology, Vol. 55 (6), 570579.

Johansson, Johny K. (1989), "Determinants and Effects of the Use of "Made in" Labels," International Marketing Review, Vol. 6(1), 4758.

Johansson, Johny K., and Israel D. Nebenzahl (1986), "Multinational Production: Effect on Brand Value," Journal of International Business Studies, Vol. 17 (Fall), 10126.

Kapferer, Jean-Noel (1994), Straegic Brand Management: New Approaches to Creating and Evaluating Brand Equity, New York: Free Press.

Kenny, David A. and Judd, Charles M. (1984), "Estimating the Nonlinear and Interactive Effects of Latent Variables," Psychological Bulletin, 96: 20110

Lawrence, C., N. E. Marr and G. P. Prendergast (1992), "CountryofOrigin Stereotyping: A Case Study in the New Zealand Motor Vehicle Industry," European Journal of Marketing, Vol. 30, 3751.

Lee, Dongdae (1994), Effects of Perceived Quality, Product Category Similarity, and Brand Breadth on Consumers’ Perceptions of Brand Extensions: Tests of Categorization Theory and Cognitive Response Theory," Unpublished Doctoral Dissertation, University of North Texas, Denton, Texas.

Lee, Dongdae, and G. Ganesh (1997), "Effects of Partitioned Country Image in the Context of Brand Image and Familiarity: A Categorization Theory Perspective," 1997 AmA Educators Proceedings, Chicago, 5051

Mervis, Carolyn B. and Eleanor Rosch (1981), "Categorization of Natural Objects," Annual Review of Psychology, Vol. 32, 89-115

Nag, Amal, "Chrysler Tests Consumer Reaction to Mexican-Made Cars Sold in the U.S.," Wall Street Journal, July 23, 1984, Section 2.

Nagashima, Akira (1970), "A Comparison of Japanese and U.S. Attitudes Toward Foreign Products," Journal of Marketing, Vol. 34 (January), 6874.

Parameswaran, Ravi and Attila Yaprak (1987), "A CrossNational Comparison of Consumer Research Measures," Journal of International Business Studies, Vol. 18 (Spring), 3549.

Peterson, Robert A. and Alain J. P. Jolibert (1995), "A Meta Analysis of Country of Origin Effects," Journal of International Business Studies, 4. 883-900

Ping, Robert A. (1995), "A Parsimonious Estimating Technique for Interaction and Quadratic Latent Variables," Journal of Marketing Research, 32 (August), 33647

Roth, Martin S. and Jean B. Romeo (1992), "Matching Product Category and Country Image Perceptions: A Framework for Managing Country of Origin Effects," Journal of International Business Studies, Vol. 23 (3), 47797.

Sujan, Mita (1985), "Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments," Journal of Consumer Research, vol. 12 (June), 3146.

Szybillo, George J. and Jacob Jacoby (1974), "Intrinsic versus Extrinsic Cues as Determinants of Perceived Product Quality," Journal of Applied Psychology, Vol. 59 (1), 7478.

Thorelli, Hans B, JeenSu Lim, and Jongsuk Ye (1989), "Relative Importance of Country of Origin, Warranty and Retail Store Image on Product Evaluations," International Marketing Review, Vol. 6(1), 3546.

Wall, Majorie, John Liefeld, and Louise A. Heslop (1991), "Impact of Country of Origin Cues on Consumer Judgments in MultiCue Situations: a Covariance Analysis," Journal of the Academy of Marketing Science, Vol. 19 (2), 10513.

----------------------------------------