Choice Strategies and Involvement: a Cross-Cultural Analysis

Dana L. Alden, University of Texas
Wayne D. Hoyer, University of Texas
Guntalee Wechasara, Chulalongkorn University
ABSTRACT - Interest in the relationships between affective factors and cognitive processes in consumer decision making continues to grow. However, the emerging body of theory in this area is thus far culture bound--that is, it has yet to be exam led across different cultures. This study explores hypothesized relationships between involvement and the use of decision making strategies in samples from three cultures: West Germany, Thailand and the United States. Results indicate that higher levels of involvement lead to greater use of both affective and cognitive decision-making heuristics in all three cultures.
[ to cite ]:
Dana L. Alden, Wayne D. Hoyer, and Guntalee Wechasara (1989) ,"Choice Strategies and Involvement: a Cross-Cultural Analysis", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 119-126.

Advances in Consumer Research Volume 16, 1989      Pages 119-126

CHOICE STRATEGIES AND INVOLVEMENT: A CROSS-CULTURAL ANALYSIS

Dana L. Alden, University of Texas

Wayne D. Hoyer, University of Texas

Guntalee Wechasara, Chulalongkorn University

ABSTRACT -

Interest in the relationships between affective factors and cognitive processes in consumer decision making continues to grow. However, the emerging body of theory in this area is thus far culture bound--that is, it has yet to be exam led across different cultures. This study explores hypothesized relationships between involvement and the use of decision making strategies in samples from three cultures: West Germany, Thailand and the United States. Results indicate that higher levels of involvement lead to greater use of both affective and cognitive decision-making heuristics in all three cultures.

INTRODUCTION

During the 1970's and early 1980's, consumer decision-making research in the United States focused largely on cognitive processes. Typically, consumers were thought to cognitively evaluate and integrate information on product attributes in order to reach a decision (Peterson et al. 1986). More recent research has indicated that basic affective processes (i.e., basic feelings of like or dislike) may also play a key role in influencing consumer information processing (Zajonc and Markus 1982; Batra 1986). In fact, for many purchases, it appears that cognitive decision processes don't occur, even on the first purchase (Olshavsky & Granbois 1979).

While a detailed review of this vast literature is beyond the scope of the present paper, one general conclusion may be drawn: research in consumer decision making has tended to focus on three general types of processes: (1) extensive cognitive evaluation (e.g., multi-attribute models such as Fishbein and Ajzen 1975), (2) basic affective processes (e.g., Zajonc and Markus 1982), and (3) simple cognitive processes (e.g. the use of choice heuristics - Hoyer 1984). Although it has not yet been studied in detail in the consumer literature, one could also include strong emotion as an important potential determinant of choice processes. Thus one could posit two general continuums, one cognitive and one affective. The cognitive continuum would range from simple choice rules to extensive cognitive evaluation, the affective continuum from basic affect to strong emotion. Also, these two processes need not necessarily be independent (i.e., affective processes may influence cognitive processes and vice-versa)*

Most importantly, a variety of different factors could potentially determine where a particular consumer is placed on the continuum for a particular decision context (Peter and Olson 1987). Previously studied factors would include perceived risk (Deshpande and Hoyer 1983), involvement (e.g., Bloch and Richins 1983) and product knowledge (e.g., Bettman and Park 1980). One problem with this large body of research, however, is that it has the potential to be culture bound. Thus, a major question which can be posed is: Do consumers in other cultures process information in a manner similar to American consumers? The purpose of the present study is to examine the general types of decision strategies used in different cultures and to relate these tendencies to a key factor identified in many studies: product involvement.

The Cross-Cultural Context

Cross-cultural consumer studies of affective, cognitive and/or heuristic components of decision-making appear to be non-existent in the consumer behavior literature. Although researchers have explored external information search issues (Anderson and Engledow 1977; Tan and Dolich 1983; Thorelli 1985), cross-cultural study of internal processing issues has been left almost exclusively to psychologists working in non-consumption contexts. Furthermore, the use of decision-making strategies and the moderating factor of involvement across cultures appears to be relatively unstudied in both disciplines.

On the cognition side, work has focused on differences and similarities in attitude and belief content (see Davidson 1980 for a thorough review of the literature). Other psychologists have looked at such diverse topics as: object representation (classification, memory and conservation); spatial representation, intelligence and cognitive development (Pick 1980). While early studies found cross-cultural differences, recent studies have more effectively controlled potential confounding factors such as familiarity with the stimulus. Once such contextual factors have been controlled, researchers have reported few if any differences in basic information processing abilities (Pick 1980). Based on findings such as these, many cross-cultural psychologists have concluded that basic processing abilities are universal while cognitive content may vary depending on the types of cognitions and processing styles the culture deems important (Cibrowski 1979; Price-Williams 1980).

The bulk of cross-cultural research on affect has explored the question of whether or not facial and vocal expressions and their attendant emotions can be recognized across all cultures (Izard 1980). This focus has grown out of the debate, raging since the time of Darwin (1872), over the question of whether emotions are biological or learned (Boucher 1979). Generally, certain universal emotions (roughly six to eight) have been identified using facial expressions (Izard 1980; Boucher and Carson 1980; and McAndrew 1986) and vocal expressions (McCluskey and Albas 1980; Bezzoijen, Otto and Heenan 1983). These basic expressions and emotions have been found to hold across both Western and non-Western, developed and developing cultures.

What appears to vary by culture are more complex expressions and emotions as well as affect display rules (Friesen 1972). Yet, with the exception of facial expression research and more recent studies of emotional antecedents (Scherer et. al. 1986), our limited understanding of affect in a cross-cultural context leads Boucher (1979) to conclude:

The paucity of available studies of emotion in the cultural domain suggests that a high rate of practical return could be yielded from an investment in studies such as replication of American-based research in other cultures.

Boucher's (1979) comments appear to apply equally well to the cross-cultural study of consumer behavior in general and the study of affective and cognitive consumer decision-making heuristics in particular. For example, in an analysis of published, empirical research in the international marketing Field from 1976 to 1982 (112 studies in all), Albaum and Peterson (1984) found that only 5% of the studies reviewed focused on cross-cultural consumer behavior.

Despite the relative scarcity of cross-cultural research, there appears to be general consensus that far more comparative analyses are needed. Indeed every major review or comment on the published research in international marketing has emphasized the culture-bound nature of the present body of knowledge and called for additional studies of external validity (Cauvisgil and Nevin 1981; Boddewyn 1981; Albaum and Peterson 1984; Cunningham and Green 1984).

Methodological Issues

In addition to calling for research, numerous authors have noted that cross-cultural studies often contain serious methodological problems. Hui and Triandis (1985) point to a potential problem involving research designs which attempt to analyze differences in beliefs, attitudes or behaviors using only two or three cultures. According to Hui and Triandis (1985), the validity of this approach is questionable because "culture" serves as the sampling unit. As such, it is unlikely that a two or three unit sample of cultures will provide enough variability in the construct of interest to validate attribution of observed differences to the cultural treatment rather than several other possible factors (see also, Davidson 1980; Davidson and Thomson 1979).

Davidson (1980), however, argues that the use of two or three cultures for testing the "boundary conditions" of existing theory is methodologically sound. Such designs are said to be valid because of their focus on relationships among a number of variables within one culture at a time rather than mean differences across cultures. This approach has two additional advantages.

When the study is not concerned with comparing mean differences across cultures, only conceptual/functional equivalence of measurement constructs is required (Hui and Triandis 1985). Conceptual/functional equivalence exists when a given operationalization is capable of capturing a behavioral construct that is meaningful to both cultures and directed to similar goals (Berry and Dasen 1974). It is generally agreed that establishing conceptual/functional equivalence across cultures is less difficult than verifying conceptual, functional and measurement equivalencies (for a thorough review of equivalence issues, see Hui and Triandis 1985). Second, it is often argued that the nomological validity of cross-cultural differences can only be assumed following replication of general patterns of similarity. Davidson (1980), for example, states that without knowledge of such cultural similarities, it is not possible to distinguish true differences from the large number of methodological complications which may explain observed patterns.

Study Purpose

As previously noted, this study explores the impact of involvement on the use of alterative decision-making heuristics in samples from three cultures: the United States, Thailand and West Germany. The study's objectives are two-fold. First, by focusing on similarities in relationships within selected cultures, the study seeks to illustrate a general cross-cultural research strategy suggested by many psychologists but as yet used by few consumer researchers. Second, by looking at how decision-making heuristics vary depending on involvement with the purchase decision in the three cultures, the study serves as an exploratory step toward better understanding the external validity of U.S. based models of information processing in general and consumer choice heuristics in particular.

METHOD

Subjects - The sample was composed of university undergraduate students in each of the three cultures. The American sample consisted of 264 students in an introductory marketing class. The West German sample was composed of 115 undergraduate business students from several different universities. Finally, the Thai sample of 93 students was obtained from a business college within one of the nation's major universities. While convenience sampling was employed in each culture, the fact that all respondents came from similar academic levels and programs was expected to produce generally comparable samples.

Independent Variables

Culture - The three cultures used in this study were selected for both convenience and theoretical reasons. For example, West Germany was chosen as test culture because it shares a generally common heritage with the United States. Thailand was examined because of its more dramatic departure in cultural heritage (i.e. a non-Western culture).

Involvement - Involvement was operationalized by means of product categories. Although it is recognized that involvement is more accurately determined by product X individual interaction (i.e., the same product is not equally involving for all individuals), it was felt that this method was the most useful for making cross-cultural comparisons. In particular, pretests suggested that the concept of involvement toward a product may be foreign to the German culture and language. Thus, it was felt in this instance to be less problematic to remain at the product level.

The six products chosen for study were based on several U.S. pretests First, based on a review of the literature, a pretest was conducted to identify items which exhibited high criterion reliability in tapping the involvement construct. Four items emerged from the larger set as fulfilling this criterion and were employed for the next set of pretests.

At the next stage, the author(s) generated a large number of product categories which are purchased by students. Each of these product categories was then rated on the involvement battery of questions. Each statement was rated on a 1 (strongly disagree) to 5 (strongly agree) scale (with a 0 indicating "don't use"). An examination of scores for the total pool of products revealed 3 products which should be classified as strongly high on involvement: wristwatch (x = 16.67), jeans (x = 17.72), and athletic shoes (x = 17.72). Likewise there were three which were rated low in involvement: peanut butter (x = 8.31), cleaning powder (x= 7.76), and chewing gum (x= 10.14). In addition the main study included a manipulation check for involvement to ensure that the manipulation was emically valid for the other two cultures. Due to the cross-cultural problems with the concept of involvement mentioned earlier, average time spent making a decision was employed as a pseudo-measure of the construct.

Dependent Variables

The dependent variables of interest were the extent to which affective and cognitive decision rules are employed to make a variety of consumer decisions. A pretest was conducted to develop items to tap each of these dimensions. A group of 24 American student subjects was asked in an open ended format to describe the processes by which they decide to purchase a variety of consumer products. A detailed analysis of these protocols was then employed to develop two sets of dependent measures.

Choice Heuristics - The first set of dependent measures consisted of a series of statements which reflected choice heuristics which one might employ when making a decision (similar to Hoyer 1984). These statements covered a wide variety of decision strategies including affective rules (such as "I buy the brand that I have a gut' liking for" or "I buy the brand that makes me feel good"), cognitive rules (such as "I buy the brand that I have collected a lot of information about" or "I buy the brand I am most familiar with" or "I buy the brand I have used most often in the past"). In all, 22 different choice heuristics were evaluated. Each statement was rated on a 5 point scale ranging from 1 (=always) to 5 (=never).

It is important to note that verbal reports of decision making behavior have been shown in previous research to be a less effective means for obtaining a description of detailed information processing behaviors. However, the use of this technique in the present study is justified on two counts. First, given data collection resources available to the researchers, it would have been quite difficult to employ the more common and current process techniques used in consumer information processing research. In particular, this was problematic in collecting data in West Germany and Thailand where only very limited resources were available. Second, the purpose of the present study is not to provide a detailed account of the exact nature of information processing. Rather, the intent is to examine the broad level existence of certain types of processing across the different cultures. The present measures were judged to be suitable for this purpose.

Process Descriptions - As a second indicator of decision processes, respondents were presented with a series of paragraphs which described the way in which some consumers make a product choice. These descriptions were taken directly from the verbal protocols mentioned earlier. In all, 10 different paragraphs were generally able to cover the universe of mentioned decision processes. Subjects were asked to indicate which paragraph most closely represented their decision process for each product.

Development of German and Thai Questionnaires

A double-back translation technique (Green and White 1976) was employed to ensure the development of comparable versions of the questionnaire in German and Thai. In each case, the questionnaire was translated by a native of that country who was also quite fluent in English. The translated versions were then back translated into English by a second set of individuals who were also fluent in both languages. Any problems or discrepancies were discussed and corrected.

Procedure

The study materials were administered by means of questionnaire booklets. Each booklet contained a set of dependent measures for three of the six test products. The order and inclusion of the products was equally counterbalanced across subjects. Students were provided with the materials in class and were asked to answer the questions at their own leisure and return them during the next class period.

RESULTS

Convenience sampling of upper division college students in all three cultures produced samples which were comparable on fewer factors than expected. For example, the Thai and West German and the U.S. and the West German samples were significantly different in age, based on Tukey's HSD test with p < .05. However, the effect size of these differences was rather small. In addition, all three samples were significantly different in gender composition. However, this variable was not expected to be strongly related to dependent variables, as verified in subsequent analysis.

More importantly, emically designed economic indicators suggested that the overwhelming majority of respondents in all three cultures came from families which were middle to upper middle income within their respective societies. Finally, the existence of generally similar educational processes in each culture indicated that the junior and senior level undergraduates who participated in the study all had roughly equivalent years of formal education.

TABLE 1

MANIPULAIION CHECK ON INVOLVEMENT (TIME SPENT EVALUATING)

TABLE 2

FACTOR LOADINGS FOR AFFECT/COGNITION HEURISTICS

Validity Tests - Prior to analysis of substantive issues, several validity tests were undertaken. First, low and high involvement treatments were checked for emic validity by measuring time generally spent evaluating decisions for each product set on a seven point scale. Only for two products were there significant differences on "time spent evaluating" across cultures at p < .05 (cleaning powder and wristwatch). More importantly, in all three cultures, the mean differences between the two product sets on the manipulation check measure were strongly significant (see Table 1).

Second, to guard against functional/conceptual non-equivalence, an analytic approach referred to as ''internal structure congruence" was undertaken on the instrument's twenty-two item choice heuristic battery. As described by Hui and Triandis (1985), internal structure congruence typically involves factor analyzing items in each culture and determining the extent to which theoretically similar measures load on the same factor in a similar manner. The greater the similarity in "structure" across cultures the greater the likelihood that conceptual/functional equivalence has been achieved.

Principal components analyses with varimax rotation revealed that the two dimensions of interest in this study, cognition and affect, produced generally similar loadings in all three cultures. (See Table 2). Loading strongly on one factor were two primarily cognitive items: "is the best on one or two key things I am looking for" and "best on more than two factors". This factor would appear to reflect a fairly strong cognitive evaluation (as opposed to a simple choice heuristic). Loading strongly on a second factor (with the exception of "strong feeling" in W. Germany) were three affective items: "makes me feel good"; and "I have a 'gut' liking for" and "I have a strong feeling for." Taken together, these items suggest the presence of a stronger level of affect than is typically studied in the previously cited research. The presence of similar structures indicates that these items are tapping two conceptually/functionally equivalent dimensions (affective vs. cognitive choice strategies) in all three cultures.

Having found evidence of functional/ conceptual equivalence, convergent and discriminant validity tests were administered to further verify the orthogonality of the choice strategy dimensions. These tests u ere made by comparing correlations between the affective and cognitive measures just discussed and script-like descriptions of alternative choice heuristics included in the research instrument. As reported in Tables 3 and 4, correlations were all in the expected direction with theoretically similar (though operationally different) questions generally resulting in significant, positive correlations. This pattern would suggest the presence of convergent validity. Furthermore, theoretically dissimilar operationalizations generally resulted in correlations insignificantly different from zero, indicating discriminant validity.

TABLE 3

CONVERGENT VALIDITY TEST

TABLE 4

DISCRIMINANT VALIDITY TEST

Finally, the relationships within each culture between involvement and use of alternative choice heuristics were analyzed using two derived scale criterion variables. These were operationalized as the average scores on the cognitive and affective items which were validated previously. A full factorial MANOVA model was first tested with three predictors (culture with three levels, involvement with two, and sex). Neither the "sex" main effect nor any of the interaction terms were related to frequency of affective or cognitive strategy usage.

A reduced main effects only Manova model (3X2) with both criterion variables and two predictors (involvement and culture) was next tested. This model appears to capture the underlying phenomena in a parsimonious yet valid manner and is presented in Table 5. Both main effects were related to increased usage of both strategy sets, affective and cognitive. Furthermore, inspection of mean differences across levels of involvement shows that the construct is positively related to usage of both affective and cognitive strategies. Finally, analysis of mean differences across cultures reveals that the West German sample indicated the lowest usage of both heuristic types for the six products. Further, the Thais indicated the highest usage of the affective heuristic and the US sample the highest for the cognitive. For the reasons stated earlier, without additional equivalence testing and replication, it is not possible to conclude that this last set of scores actually reflects underlying cultural differences in strategy usage.

To further validate the final model, step down (or sequential) analysis was performed. The relative size of the univariate F tests along with the canonical variate correlations (.907 for the affective criterion and .621 for the cognitive) suggested that the affective criterion variable was more strongly related to the predictors in the model than was the cognitive criterion variable. Hence, the affective variable was first modeled alone in a main effects only, ANOVA (3X2) design. Both main effects were significant at p < .01 (involvement F = 30.8 and culture F = 32.3).

TABLE 5

FINAL MODEL (3X2) MAIN EFFECTS ONLY

Next, the affective measure was used as a covariate in a main affects only ANCOVA (3X2), with ''cognitive strategy usage" as the criterion variable. In this test, "cognitive strategy usage" remained significantly related to both main effects at p < .01 (involvement F = 6.11 and culture F = 5.73). This finding supports the conclusion that involvement has a significant impact on the frequency of both affective and cognitive strategy usage in all three cultures.

CONCLUSION

This study has demonstrated a useful and appropriate methodology for studying U.S. based consumer behavior theory in a cross-cultural context. By focusing on within culture similarities in relationships between major consumer behavior variables and by using equivalence validation techniques such as those described herein, insight into the global applicability of consumer behavior theory can be gained. Such effort will also pave the way for eventual mapping of differences across cultures. As a step in this direction, a detailed analysis employing functionally, conceptually and metrically equivalent measures is needed to determine whether or not culture truly impacts on frequency of choice strategy use.

Second, this paper provides preliminary support for the notion that involvement impacts on the type of information processing which ensues within cultures other than in the US. The affective and cognitive factors found in all three cultures reflected fairly strong levels of both cognition and affect processing and the usage of both processing strategies increased at the higher levels of involvement. Such evidence suggests that certain relationships of current interest to consumer behavior researchers may indeed possess cross-cultural validity.

Finally, the fact that both cognition and affect moved in the same direction as involvement increased may provide support of the dual-continuum proposition described earlier. Furthermore, it suggests that these processes should not be studied independently. The present study did not, however, span the entire length of both continua and future research is needed to explore this possibility.

REFERENCES

Albaum, Gerald and Robert A. Peterson (1982), "Empirical Research in International Marketing," Journal of International Business Studies (Spring/Summer), 161-173.

Anderson, Ronald D. and Jack L. Engledow (1977), "A Factor-Analytic Comparison of U.S. and German Information Seekers," Journal of Consumer Behavior, 3 (March), 185-196.

Batra, Rajeev (1986), "Affective Advertising: Role, Processes and Measurement," in Robert A. Peterson, Wayne D. Hoyer and William R. Wilson (Eds.), The Role of Affect in Consumer Behavior: Emerging Theories and Applications, Lexington, MA: Lexington Books.

Berry, J.W. and Dasen, P. (Eds.) (1974), Culture and Cognition London: Metheun.

Bettman, J.R. and C.W. Park (1980), "Effects of Prior Knowledge and Experience and Phase of the Choice Process on Consumer Decision Processes: A Protocol Analysis," Journal of Consumer Research, (December), 234-245.

Boddewyn, Jean J. (1981), "Comparative Marketing: The First Twenty-Five Years," Journal of International Business Studies, (Spring/Summer), 61 -79.

Bezzoijen, Renee Van, Stanley A. Otto and Thomas A. Heenan (1983), "Recognition of Vocal Expressions of Emotion," Journal of Cross-Cultural Psychology, 14 (4), 387-406.

Bloch, Peter H. and Marsha L. Richins (1983), "A Theoretical Model of the Study of Product Importance Perceptions," Journal of Marketing (Summer), 6981.

Boucher, Jerry D. (1979), "Culture and Emotion," Perspectives on Cross-Cultural Psychology, NY Academic Press.

Boucher, Jerry D. and Gary E. Carson (1980), "Recognition of Facial Expression in Three Cultures," Journal of Cross-Cultural Psychology, 11 (3), 263-280.

Cauvisgil, S. Tamer and John R. Nevin (1981), "The State of the Art in International Marketing: An Assessment," in Ben Enis and Ken Roering (eds.), Review of Marketing, Chicago: American Marketing Association.

Cibrowski, Thomas J. (1979), "Cross-Cultural Aspects of Cognitive Functioning: Culture and Knowledge," in Anthony J. Marsella, Robert G. Tharp and Thomas J. Cibrowski (Eds.), Perspectives on Cross-Cultural Psychology, NY Academic Press.

Darwin, Charles (1872, 1955), The Expressions of Emotions in Man and Animals, NY: Philosophical Library.

Davidson, Andrew R. (1980), "Culture and Attitude Structure and Change," in Anthony J. Marsella et al. (eds.), Perspectives on Cross-Cultural Psychology, NY: Academic Press, 137- 155.

Davidson, Andrew R. and Elizabeth Thomson (1980), "Cross-Cultural Studies of Attitudes and Beliefs," in Harry C. Triandis and Richard W. Brislin (Eds.), Handbook of Cross-Cultural Psychology: Social Psychology, Volume 5, Boston, MA: Allyn and Bacon, Inc.

Deshpande, Rohit and Wayne D. Hoyer (1983), "Low Involvement Decision Processes: The Importance of Choice Tactics," in Marketing Theory: Philosophy of Science Perspectives, in Robert F. Bush and Shelby D. Hunt, (Eds.), Chicago: AMA, 155-158.

Fishbein, M. and I. Ajzen (1975), Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research, Reading, MA: Wesley PubLishing Co.

Friesen, W.V. (1972), Cultural Differences in Facial Expressions in a Social Situation: An Experimental Test of the Concept of Display Rules Unpublished Doctoral Dissertation, University of California.

Green, Robert T. and Phillip D. White (1976), "Methodological Considerations in Cross-National Consumer Research," Journal of International Business Studies, Fall/Winter, 80-88.

Hoyer, Wayne D. (1984), "An Examination of Consumer Decision-making for a Common Repeat Purchase," Journal of Consumer Research, 11 (December), 82^829.

Hui, C. Harry and Harry C. Triandis (1985), "Measurement in Cross-Cultural Psychology," Journal of Cross-Cultural Psychology, Vol. 16 (2), 13 1 - 15 2.

Izard, Carroll E. (1980), "Cross-Cultural Perspectives on Emotion and Emotion Communication," in Handbook of Cross-Cultural Psychology, Vol. 3, in Harry C. TriandIs and Walter J. Lormer (Eds.), pp. 185-221.

McAndrew, Francis T. (1986), "A Cross-Cultural Study of Recognition of Thresholds for Facial Expressions of Emotion," Journal of Cross Cultural Psychology, 17 (2), 211-224.

McCluskey, Ken W., and Albas, Daniel C. (1980), "Perception of Speech by Canadian and Mexican Children, Adolescents and Adults," International Journal of Psychology, 16, 119- 132.

McGuire, W.J. (1985), "Attitudes and Attitude Change," in Gordon Lindzey and Elliot Aronson (Eds.), Handbook of Social Psychology 3rd Ed., Reading, MA: Addison-Wesley.

Olshavsky, Richard W. and Donald H. Granbois (1979), "Consumer Decision-Making -- Fact or Fiction?" Journal of Consumer Research, 6, 93-100.

Peter, J. Paul and Jerry C. Olson (1987), Consumer Behavior: Marketing Strategy Perspectives, Homewood, IL: Irwin.

Peterson, Robert A., Wayne D. Hoyer and William R. Wilson (1986), "Reflections on the Role of Affect in Consumer Behavior," in Robert A. Peterson, Wayne D. Hoyer and William R. Wilson (Eds.), The Role of Affect in Consumer Behavior, Lexington, MA: Lexington Books.

Pick, Anne D. (1980), "Cognition: Psychological Perspectives," in Harry C. Triandis and Walter J. Lonner (Eds.), Handbook of Cross-Cultural Psychology (Vol 3), Boston: Allyn and Bacon, Inc., 117-153.

Price-Williams, Douglass (1980), 'Toward the Idea of Cultural Psychology," Journal of Cross-Cultural Psychology, 11 (1 March), 75-88.

Scherer, Klaus R., Harold G. Wallbott and Angela A. Summerfield (eds.) (1986), Experiencing Emotion: A Cross-Cultural Study, London: Cambridge University Press.

Tan, Chin Tiong and Ira J. Dolich (1983), "A Comparative Study of Consumer Information Seeking: Singapore Versus U.S., " Journal of the Academy of Marketing Science, 11 (3), 313-322.

Thorelli, Hans B . (1985), "Product Information: Search and Satisfaction: The PRC, Overseas Chinese and Thailand," International Marketing Review, (Winter), 12-20.

Wober, M. (1969), "Towards an Understanding of the Concept of Intelligence," in John W. Berry and P R. Dasen (Eds.), Culture and Cognition: Readings in Cross-Culture Psychology, London: Methuen.

Zajonc, R.B. and H. Markus (1982), "Affective and Cognitive Factors in Preference," Journal of Consumer Research, 9, 123-131.

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