Socialization and Acculturation Impacts on Country-Of-Origin Image: Initial Results

ABSTRACT - Country-of-Origin (CO) is a potentially powerful image variable which may be used to influence competitive positioning and success in the global market-place. A weakness in CO research and implementation is to treat a country's consumers as homogenous. Dividing the country market into its natural segments before studying CO perceptions would be an added advantage. The current study segments the market by immigrant status (foreign born, first generation, and local ancestry) to study the impact of acculturation and socialization on perception of foreign products. Findings reveal subtle differences in how these groups perceive foreign products.

Citation:

R. Mohan Pisharodi and Ravi Parameswaran (1994) ,"Socialization and Acculturation Impacts on Country-Of-Origin Image: Initial Results", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 109-115.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 109-115

SOCIALIZATION AND ACCULTURATION IMPACTS ON COUNTRY-OF-ORIGIN IMAGE: INITIAL RESULTS

R. Mohan Pisharodi, Oakland University

Ravi Parameswaran, Oakland University

ABSTRACT -

Country-of-Origin (CO) is a potentially powerful image variable which may be used to influence competitive positioning and success in the global market-place. A weakness in CO research and implementation is to treat a country's consumers as homogenous. Dividing the country market into its natural segments before studying CO perceptions would be an added advantage. The current study segments the market by immigrant status (foreign born, first generation, and local ancestry) to study the impact of acculturation and socialization on perception of foreign products. Findings reveal subtle differences in how these groups perceive foreign products.

INTRODUCTION

Country-of-Origin (CO) image refers to "buyers' opinions regarding the relative qualities of goods and services produced in various countries" (Bilkey 1993, p. xix). CO serves as a useful extrinsic cue (and as a surrogate for difficult-to-evaluate intrinsic characteristics such as quality and performance) because consumers tend to be less familiar with foreign than with domestic products (Han and Terpstra 1988).

An indication of the importance of country-of-origin images in international marketing strategy can be gauged by the sheer volume of research on the topic. Consumer perceptions of foreign products have been examined from a wide variety of perspectives. The consumer behavior stream represents a fifth of all the empirical research conducted in the field of international marketing and is also the stream that has made the most progress in theory development (Aulakh and Kotabe 1993).

A major weakness of country-of-origin image studies is that they tend to view consumers in a particular country as being a homogeneous group and, therefore, report on the country's consumers' perceptions of foreign products without distinction to any segmentation (other than gender and age) that may exist. In fact, in their assessment of the theoretical and methodological developments in international marketing, Aulakh and Kotabe (1993) highlight the paucity of research in market segmentation (the subject matter of only 1.3%-11 journal articles-of all research in international marketing) as one of the serious limitations.

PURPOSE OF THE STUDY

A significant segmenting variable in the study of country-of-origin images is ethnicity, ancestry or national origin. An interesting question relating to country-of-origin images in a country like the U.S. with a history of immigration is whether ancestry/national origin moderates a person's perceptions of foreign products? If it does, what is the impact of the length of domicile (in a country) on the person's perceptions about foreign products? In other words, what are the impacts of consumer socialization and acculturation on the perception of foreign products? Consumer socialization and acculturation refer to "the acquisition of consumption related cognitions, attitudes and behavior" (Engel, Blackwell and Miniard, 1993, p. 64). Socialization is the process by which the individual learns of the norms and values of the society in which he/she is raised whereas acculturation is the process of learning the norms and values of a society different from the one in which the person was raised (Assael 1992). Specifically, do recent immigrants respond to foreign products differently than first generation citizens (children of immigrants who are born in the country)? Again do first generation citizens respond to foreign products differently than those who can trace their ancestry in the country further back? In other words, at what point in time in the immigration continuum (from the time a person lands in a new country to the point where he/she can trace ancestors in that country) does a person's response to foreign products become indistinguishable from the response of typical consumers of that country? The current study attempts to answer some of these questions.

The purpose of this study is to address some of the issues raised in the above discussion. Specifically, we would like to investigate the nature and dimensionality of the country-of-origin construct broadly defined (that is, country image as a function of both perceptions towards the people and products of the country in question) and the effects of CO image on purchase intentions in the context of the acculturation/socialization process.

THE COUNTRY-OF-ORIGIN CONSTRUCT

Nagashima (1970) has been credited with first defining country image as "the picture, the representation, the stereotype that businessmen and consumers attach to products of a specific country. This image is created by such variables as representative products, national characteristics, economic and political background, history and tradition." The impact of the CO cue on consumption behavior has been related to producing country characteristics. Papadopoulos, Heslop and Bamossy (1989) summarized this notion by stating that the perceptions of sourcing countries are impacted by cognition about, and affect and conative orientation toward, that country's peoples. They further noted that a consumer's image of a people with whom he/she is not familiar may well be formed upon the basis of knowledge about that people's capacity for producing quality products in generalCand that perception impacts his/her evaluation of specific products from that country. Yaprak and Parameswaran (1986) hypothesized that purchase intentions and behavior are impacted by CO effects (General Country Attribute [GCA] and General Product Attribute [GPA]) as well as by specific product attributes (SPA). This three-factor conceptualization of the CO construct forms the basis of our current investigation.

MEASUREMENT METHODOLOGY AND SAMPLING

In order to fulfil the objectives of this study, GCAs for a source country (Germany) were measured using twelve statements that are commonly found in the country-of-origin literature (Table 1). Consumers' attitudes toward the general nature of products from Germany were measured using eighteen statements again commonly cited in the CO literature. Their perceptions in regard to specific product related cues (SPA's) for a German (Volkswagen Jetta) car were measured using 10 relevant automobile attributes. These perceptions were measured by identifying relevant attributes through a key informant procedure involving automobile dealers. The above variables-GCAs, GPAs, and SPAs-were measured on a 10-point scale where '1' represented that the statement was 'not at all appropriate' while '10' represented that it was 'most appropriate.' A respondent's likelihood of purchasing (IP) the Volkswagen Jetta was operationalized as a 10 point likelihood scale where '1' represented 'not at all likely' and 10 represented 'extremely likely.

TABLE 1

LIST OF INDICATORS

Data were gathered from the adult population of a large midwestern metropolitan area which is highly heterogenous in terms of ethnic (national origin) composition. Blank questionnaires were hand delivered to the selected (systematic sampling procedure) respondents comprising the relevant sampling population (namely immigrant, first generation and US consumers) and completed questionnaires were collected within the next two weeks. A total of 678 completed and useable questionnaires were returned from the 1025 that were originally placed, a 66% response rate. 583 questionnaires containing valid responses regarding acculturation/socialization (305 immigrants, 123 first generation, and 155 US consumers) ["Immigrant" was defined as respondents who were foreign-born but residing in the United States. "First generation" refers to respondents who were born and are residing in the United States but whose parents are foreign-born, and "US consumers" alludes to those who were born and are residing in the United States and whose parents are also US-born.] were used in the analysis of the acculturation/socialization effects.

STATISTICAL ANALYSIS OF DATA

The statistical analysis of data consisted of three steps: the purification of the measurement model, the analysis of CO effects within the proposed model, and the assessment of the impact of socialization and acculturation on CO effects. Our investigation subjected a scale developed to measure country image (Parameswaran and Yaprak 1987) to the confirmatory factor model whose "ability to test specific structures suggested by substantive theory gives it a major advantage over the exploratory factor model (Long 1983)".

Purification of the Measurement Model

The indicators used to measure the underlying theoretical structure of the CO construct were analyzed for unidimensionality and reliability using ITAN (Gerbing and Hunter 1988), a statistical package for item analysis with correlational data. ITAN has the capability of using multiple groups confirmatory factor analysis (MGRP) and generates inter-item correlations, item-factor loadings, factor-factor loadings, and standardized coefficient alpha. The output generated by ITAN can be used to evaluate internal and external consistencyCtwo statistical criteria for establishing the unidimensionality of the indicators of a construct.

The measures were partitioned a priori into three different sets measuring GCA, GPA, SPA, and one indicator measuring IP (Intention to Purchase). Through ITAN analysis of data collected from all respondents (n=678), indicators were dropped and reclustered progressively until smaller sets of measures reflecting fairly high degrees of unidimensionality and reliability were identified.

Two dimensions underlying GCA perceptions were identified. The first dimension (GCA1) consisted of a set of indicators measuring the cognitive and affective aspects of the respondent's CO perceptions while the second dimension (GCA2) consisted of indicators measuring the respondents' conation (level of interaction [in this case perceived similarity]) with the source country of Volkswagen Jetta (Germany). The same procedure resulted in the selection of sets of indicators representing the dimensions underlying GPA perceptions of all the respondents for both makes of car. The first dimension (labeled GPA1) consisted of a set of indicators measuring negatively oriented product attributes which Hong and Wyer (1989) term undesirable attributes. This occurred although the directions of these indicators had been reversed earlier through appropriate data transformation. The second and third consisted of indicators measuring positively oriented product attributes. The results of ITAN analysis of the indicators of SPA did not yield multiple dimensions in any of the data sets studied. [C3, C4, C6, C7, and C8 (standardized a=.872) were selected for measuring GCA1; C9, C10, and C11 (a=.849) for GCA2; P4, P7, P9, P13, and P14 (a=.75) for GPA1; P6, P8, P12, and P17 (a=.735 for GPA2; P11, P16, and P18 (a=.796) for GPA3; and S3, S4, S5, and S9 (a=.819) for measuring SPA.]

The high item-factor loadings of each indicator on its underlying factor combined with significantly lower loadings on the other factors, and the noticeable drop in similarity coefficients between different sets of indicators provided empirical support for convergent and discriminant validity of the measurement model.

LISREL analysis was employed as a complementary technique to ITAN in the process of respecification. The improvement in fit (represented by lower chi-square values) resulting from the progressive respecification of the measurement model was assessed using incremental fit analysis (Bentler and Bonett 1980). The results of incremental fit analysis (Table 2) indicate that the process of respecification had indeed resulted in a significantly better-fitting measurement model.

Analysis of Country-of-Origin Effects

The proposed theoretical structure was modified to recognize the multidimensional nature of GCA and GPA and was represented in the form of a causal model which was analyzed using the LISREL 8 (J÷reskog and S÷rbom 1993). Four models (M0, M1, M2, and M3) were analyzed. The first model (M0) was the most restricted model and consisted of seven unrelated constructs. In the models that followed it, guided by the proposed theoretical structure, selected structural relationships were freed. [In M1, IP is the dependent variable and the other six variables are independent. In M2, SPA is an intervening variable with IP as the dependent variable and the rest as independent variables. In M3 (see Figure 1), the GPA variables intervene between the GCAs and SPA.] The final model (M3) is presented in Figure 1a with selected structural notations and estimates.

The chi-square statistic generated by the models do not indicate a good fit of any of the final models to their respective data. However, the drop in chi-square values with each model modification is noticeable and statistically significant (p_0.05) indicating that relaxing the structural links significantly improves the fit between the specified models and their respective data. The values of the Non-normed Fit Index and the Normed Fit Index, two indices for evaluating the increment in fit obtained in step-up comparison of two models (Bentler and Bonett 1980), indicate that the final model explains a large portion of the variance unexplained by the first model (although the nature of these indices make it impossible to pinpoint the exact percent of variance explained by it). [Each set of fit indices represent only a part of the modeling effort (either measure purification or analysis of CO effects) reported in this paper.]

One of the interesting findings of this research is that the structural links connecting GPA1 (undesirable product attributes) to all the other unobservables are generally weak (Figure 1a). The findings demonstrate that GPA1 does not fit well into the proposed theoretical structure. The extremely low factor-factor loadings of GPA1 (generated earlier through analysis using ITAN) with the other factors in the theoretical structure also support this observation. Among the facets of GCA, the relationship between GCA2 (conative aspects) and the remaining variables in the model are relatively weak when compared to those associated with GCA1 (cognitive and affective aspects)

In general, direct or proximal effects (between adjacent variables) were found to be much stronger (average of absolute values=0.34) than indirect or remote ones which bypass intervening variables, as can be observed in Figure 1a. The average (absolute) for one-variable bypass (bypassing GPA or SPA) is 0.11, while the average (absolute) for two-variable bypass (bypassing both GPA and SPA) is 0.09. The effect of GPA on SPA is much stronger than its effect on IP. This indicates that GCA impacts IP significantly through its impact on the GPA variables and on specific product information (SPA).

TABLE 2

RESULTS OF INCREMENTAL FIT ANALYSIS

Assessment of the Impact of Socialization and Acculturation

In order to assess the impact of socialization and acculturation, as previously mentioned, the data set was subdivided into three categories: immigrants, first generation, and US consumers. The differences between the first two categories would highlight the acculturation effect and the differences between the last two categories would highlight the socialization effect.

The measurement model and the structural equation model used for analyzing CO effects (M3: the final model) were applied to the three subsamples with no alteration. The sets of items used in the analysis of CO effects were found to possess acceptable or high levels of reliability in all three subsamples. The standardized values of coefficient alpha in the subsamples ranged from .622 to .878 with a mean of .788. The standardized structural coefficients estimated by LISREL during the analysis of the subsamples are presented in Figures 1b through 1d.

Most of the general conclusions derived from the analysis of the complete data set appear to hold true in each one of the subsamples. While the relationship between SPA and IP was strong in all subsamples, the relationship of SPA with GPA2 and GPA3 did not show a consistent pattern. Among first-generation respondents, GPA2 possesses a stronger link with SPA while in the other two subsamples, GPA3 has a stronger link.

GCA2 possessed stronger structural links with the other variables in the theoretical structure in the subsamples (Figures 1b-1d) than in the overall sample (Figure 1a). Furthermore, the direction of relationship differed in the different subsamples. Among the US consumers, the relationships (between GCA2 and most of the other variables) were positive and strong. Among the immigrants, the same relationships were negative and strong. The relationships for the first generation were relatively weaker and either positive or negative. The fact that subsamples displayed relationships not obvious in the overall sample makes a good case for introducing market segmentation concepts in CO image studies.

These results indicate that US-consumers tend to evaluate a product from a source country similar to their own favorably. This contrasts with the results obtained from the immigrants who give poorer ratings when the source country of the product is similar to their origin country. All the estimates obtained for the first generation respondents were between those obtained for the US consumers and immigrants. This may be the because the first generation respondents are less clear about their cultural and ethnic identity than either of the other categories.

In order to assess the impact of acculturation and socialization on the dimensions of CO image, a measure of acculturation-socialization was developed (number of years in the US/age). That is, the proportion of a person's life spent in the US was used as an indicator of acculturation, thus generating a continuum with the most recent immigrants on one extreme and first generation respondents on the other end. Since it was felt that the US consumers might have been influenced by immigrant source-country culture even less that first generation, the continuum was further extended and US consumers were positioned on the continuum beyond first-generation respondents.

A causal model with the acculturation indicator as the exogenous variable and the dimensions of CO image as endogenous variables was developed and was analyzed using LISREL 8. The resulting structural coefficients indicated that respondents with higher levels of acculturation (i.e., those who had spent most or all of their lives in the US) tended to view the people of Germany as similar to themselves culturally, economically, and in political views. Higher levels of acculturation were also found to be positively related to the other components of CO image.

FIGURE 1

ANALYSIS OF COUNTRY-OF-ORIGIN EFFECT

The items of GCA2 (S8, S9, and S10) were also analyzed using MANOVA (Norusis 1988) to determine whether the subsamples differed significantly in how they viewed German people as being similar or dissimilar to themselves. As was expected a priori, both the multivariate and the univariate F-tests were significant indicating that the immigrants viewed Germans as being most dissimilar to themselves and the US consumers viewed Germans as being most similar to themselves. The perceptions of the first generation were consistently between the perceptions of the two other categories.

CONCLUSIONS

The results of confirmatory factor analysis of the measurement model indicate that the three-factor model (specified a priori) may not be appropriate for the measurement of CO image. As discussed earlier, during our analysis of CO perceptions, GCA was found to consist of two dimensions and GPA was found to consist of three dimensions. Our analysis of CO effects on intention to purchase showed that CO perceptions have an impact on intention to purchase primarily through their impact on specific product information. The analysis of subsamples of respondents provided further support for this conclusion.

Our study also indicates that CO image variables and their effect on purchase intention are influenced by level of acculturation and socialization. Their purchase intentions and perceptions of CO image variables appear to indicate that immigrants do not value the economic, political, and cultural similarity (with oneself) as highly as US consumers do when it comes to the purchase of cars. This may be because the vast majority of the immigrants in our sample were born in developing nations. Perhaps immigrants value cars from highly industrialized countries (with people who differ from "us" a great deal) very highly. Similarly, US-born respondents value cars from highly industrialized countries (with people who are similar to "us") favorably. Thus our findings demonstrate that acculturation-socialization is a potential criterion for market segmentation in international marketing. Our study is an initial attempt at introducing the need for market segmentation consideration in country-of-origin studies.

We do not conclude that perception of similarity (or lack of it) is a factor that directly causes a respondent to have favorable CO image perceptions. The findings of our research may be the result of a third variable(s) which is causally linked to both variables of interest. The exploration of the nature of this relationship may be a promising avenue for future research.

We recognize that our results may have been influenced by the product studied (car) and the source country of the product (Germany). Future research should explore the generalizability of our findings to other product categories, other source countries, and immigrants from highly developed countries.

REFERENCES

Assael Henry (1992), Consumer Behavior and Marketing Action, Fourth Edition, Boston, MA: PWS-Kent Publishing Company.

Aulakh Preet S. and Masaki Kotabe (1993), "An Assessment of Theoretical and Methodological Development in International Marketing: 1980-1990," Journal of International Marketing, 1 (2), 5-28.

Bentler, Peter M. and Douglas G. Bonett (1980), "Significance Tests and Goodness-of-Fit in the Analysis of Covariance Structures," Psychological Bulletin, 88 (November), 588-606.

Bilkey, Warren J. (1993), "Foreword," in Product-Country Images: Impact and Role in International Marketing, Nicolas Papadopoulos and Louise A. Heslop, eds., New York: International Business Press, xix-xx.

Engel, James F., Roger D. Blackwell, and Paul W. Miniard (1993), Consumer Behavior, Seventh Edition, Fort Worth, TX: The Dryden Press.

Gerbing, David W. and John E. Hunter (1988), ITAN: A Statistical Package for Item Analysis with Correlational Data Including Multiple Groups Confirmatory Factor Analysis, Portland, OR: Portland State University.

Han, C. Min and Vern Terpstra (1988), "Country-of-Origin Effects for Uni-National and Bi-National Products," Journal of International Business Studies, 19 (Summer), 235-254.

Hong, Sung-Tai and Robert S. Wyer, Jr. (1989), "Effects of Country-of-Origin on Product Attribute Information on Product Evaluation: An Information Processing Perspective," Journal of Consumer Research, 16 (September), 175-187.

J÷reskog, Karl and Dag S÷rbom (1993), LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language, Hillsdale, NJ: Scientific Software.

Long, J. Scott (1983), Confirmatory Factor Analysis, Beverly Hills: SAGE Publications.

Nagashima, Atkira (1970), "A Comparison of Japanese and U.S. Attitudes Towards Foreign Products," Journal of Marketing, 34 (January), 68-74.

Norusis Marija (1988), SPSSx Advanced Statistics Guide, 2nd ed., Chicago: SPSS Inc.

Papadopoulos, Nicolas, Louise A. Heslop, and Gary J. Bamossy (1989), "International Competitiveness of American and Japanese Products," in Dimensions of International Business (No. 2), Nicolas Papadopoulos, ed., Ottawa, Canada: International Business Study Group, Carleton University.

Parameswaran, Ravi and Attila Yaprak (1987), "A Cross-National Comparison of Consumer Research Measures," Journal of International Business Studies, 18 (Spring), 35-49.

Yaprak, Attila and Ravi Parameswaran (1986), "Strategy Formulation in Multinational Marketing: A Deductive, Paradigm-Integrating Approach," in Advances in International Marketing, 1, S. Tamer Cavusgil, ed., Greenwich, CT: JAI Press, Inc., 21-45.

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