The Effects of Country of Origin, Brand, and Price Information: a Cognitive-Affective Model of Buying Intentions

Wai-Kwan Li, University of Illinois, Urbana-Champaign
Kent B. Monroe, University of Illinois, Urbana-Champaign
Darius K-S Chan, University of Illinois, Urbana-Champaign
ABSTRACT - The present study examines the effects of three extrinsic cues, country of origin, brand, and price information, on buying intentions. The results indicate that these cues affect perceived quality in a similar fashion. However, while price and expected price have direct effects on liking and perceived value, country of origin and brand information have no significant direct effect on these two variables. More importantly, a proposed cognitive-affective model of buying intentions is supported across the two products studied, and has more explanatory power than a cognitive model of buying intentions.
[ to cite ]:
Wai-Kwan Li, Kent B. Monroe, and Darius K-S Chan (1994) ,"The Effects of Country of Origin, Brand, and Price Information: a Cognitive-Affective Model of Buying Intentions", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 449-457.

Advances in Consumer Research Volume 21, 1994      Pages 449-457

THE EFFECTS OF COUNTRY OF ORIGIN, BRAND, AND PRICE INFORMATION: A COGNITIVE-AFFECTIVE MODEL OF BUYING INTENTIONS

Wai-Kwan Li, University of Illinois, Urbana-Champaign

Kent B. Monroe, University of Illinois, Urbana-Champaign

Darius K-S Chan, University of Illinois, Urbana-Champaign

ABSTRACT -

The present study examines the effects of three extrinsic cues, country of origin, brand, and price information, on buying intentions. The results indicate that these cues affect perceived quality in a similar fashion. However, while price and expected price have direct effects on liking and perceived value, country of origin and brand information have no significant direct effect on these two variables. More importantly, a proposed cognitive-affective model of buying intentions is supported across the two products studied, and has more explanatory power than a cognitive model of buying intentions.

Currently, globalization effects of business firms have brought about important changes in the manufacturing locations of products. Along with these changes, an important issue is whether buyers' evaluations of a product are influenced by knowing the country in which it is manufactured. Recent research has provided some evidence that country of origin, brand name, and price information affect product evaluations and buying intentions (Bilkey and Nes 1982; Ozsomer and Cavusgil 1991; Monroe 1973; Rao and Monroe 1989; Dodds, Monroe, and Grewal 1991). Although these three extrinsic cues have attracted significant research attention, little research has been done to compare whether they affect product evaluations and buying intentions in a similar fashion, or in different ways. Therefore, the first research issue of the present study is to explore how these three extrinsic cues affect product evaluations and buying intentions.

In general, researchers have assumed that consumers are rational, and therefore the measurements of product evaluations are essentially cognitively oriented (an example is available in the Appendix of Dodds, Monroe, and Grewal 1991). However, Zajonc and Markus (1982) have argued that, in addition to cognitive factors, affective factors also play an important role in preferences. They illustrated this argument by a food preference example:

"Dog meat is a delicacy in some parts of East Asia, but few Americans would find it appetizing... Most Americans like corn, but in various countries corn has been thought suitable only for pigs."

Recently, Cohen and Areni (1991) proposed a dynamic model of affect in consumer behavior. Essentially, the model describes the interactions between affect and the cognitive system across three phases, which can affect consumers' behaviors.

Recognizing affect is an important component in consumers' preferences, we will extend the cognitively oriented models of buying intentions to a cognitive-affective model of buying intentions, and examine the predictive power of the extended model.

The Cognitive Models of Buying Intentions

Effects of extrinsic cues. Scitovszky (1945) noticed that consumers might use price as a surrogate of product quality, if they had difficulties in evaluating the product. Based on this notion, Monroe (1979), and Monroe and Krishnan (1985) proposed a conceptual model describing the relationships among price, perceived quality, perceived sacrifice, perceived value, and willingness to buy. Recently, Zeithaml (1988), and Dodds, Monroe, and Grewal (1991) extended the model to include brand name and store name (Figure 1A).

In the Dodds et al. (1991) study, price had a positive effect on perceived quality, but a negative effect on perceived value, while the overall effect on willingness to buy was positive. Moreover, a favorable brand name and store name had positive effects on perceived quality. Consistent with these findings, the meta-analysis conducted by Rao and Monroe (1989) suggested that both price and brand name have positive effects on perceived quality. Turning to the country of origin effects, two comprehensive reviews (Bilkey and Nes 1982; Ozsomer and Cavusgil 1991) have consistently concluded that a favorable country of origin image has a positive effect on perceived quality.

Based on these empirical findings, it is logical to hypothesize that (Figure 1B):

H1a: As price increases, buyers' perceptions of quality increase.

H1b: When buyers' perceptions of brand name are more favorable, their perceptions of quality are higher.

H1c: When buyers' perceptions of country of origin are more favorable, their perceptions of quality are higher.

H1d: As price increases, buyers' perceptions of value declines.

H1e: When buyers' perceptions of quality increase, their perceptions of value increase.

H1f: When buyers' perceptions of value increase, their willingness to buy increase.

Effects of reference price. The existence of the concept of reference price is well-documented in the pricing literature. Essentially, it refers to an internal price to which consumers compare the observed prices (Monroe, Grewal, and Compeau 1991). One suggested operationalization is that it is an "expected price" (Winer 1988). By comparing the observed price with the expected price, consumers judge whether the product being considered is of high or low perceived value-for-money. More concretely, if a consumer has an expected price of $199, and the observed price is $250, s/he may judge the product as expensive. In another case, if his/her expected price is $299, with the same observed price, s/he probably will judge the product as not expensive. Therefore, ceteris paribus, the higher the expected price, the higher the perceived value-for-money. Incorporating this construct in the present model, we hypothesize that:

H1g: For a given observed price, when the expected price is higher, buyers' perceptions of value are greater.

The Affective Component in Buying Intentions

Although affect is identified as an important antecedent of preferences, its relationships with other cognitive antecedents of preferences are unclear. According to the traditional approach, the affect component is influenced by a cognitive component (Zajonc and Markus 1982). For example, before you like a Honda, you must know what it is. In contrast, Zajonc (1980) also suggested that affective reactions can precede cognitive reactions. In this situation, the cognitive reactions perhaps serve as a justification for liking. For example, you like blue, therefore you may evaluate a blue shirt favorably.

FIGURE 1A

EXTENDED MODEL PROPOSED BY DODDS ET AL. (1991)

FIGURE 1B

COGNITIVE MODEL OF BUYING INTENTIONS

Recently, Cohen and Areni (1991) argued that affect can be elicited automatically, as well as after some cognitive processes. In addition, an affective response can influence the subsequent cognitive responses. More concretely, imagine a baby powder advertisement, the audience may elicit some positive affective reactions automatically when the happy babies appear in the advertisement. According to Cohen and Areni's model, these reactions are the first phase affective responses, which may occur before any cognitive reactions. As the advertisement moves on, the audience receives the message about the hygienic feature of the product. This advertisement message may lead to some positive affective reactions. These reactions are the second phase affective responses, which occur after some cognitive reactions. After watching the advertisement, the audience may continue their elaborative interpretation. This elaborative interpretation, a cognitive response, may be affected by the phase two affective responses; however, it may also elicit phase three affective reactions.

The Cognitive-Affective Model of Buying Intentions

Using in-depth interviews, Li and Monroe (1992) reported that country of origin could play two roles in product evaluations, namely, signalling and attribute roles. Specifically, the signalling role of country of origin information suggests that consumers use it as a signal to infer whether a country possesses the necessary skills in manufacturing a product (a cognitive response). Notice that this is a cognitive product evaluation process.

On the other hand, the attribute role of country of origin refers to the liking effect on product evaluations. For example, consumers may like a product more (an affective response) if it is from their home countries, which is an affective product evaluation process.

If country of origin can elicit both cognitive and affective responses, it is likely that price and brand name can also induce these two types of responses. For instance, people may infer a high price product, or a national brand product, is of high quality (cognitive responses). Alternatively, they may simply like expensive products more, or like and be loyal to a certain brand (affective responses).

First phase affective responses. It should be noted that the affective responses elicited directly from extrinsic cues (some external stimulus) correspond to the first phase affective responses in Cohen and Areni's (1991) model. Therefore, we hypothesize that (see Figure 2A):

H2a: As price increases, buyers' liking (first phase) toward a product increase.

H2b: When buyers' perceptions of brand name are more favorable, their liking (first phase) toward a product is greater.

H2c: When buyers' perceptions of country of origin are more favorable, their liking (first phase) toward a product is greater.

Second phase affective responses. By definition, the second phase affective responses occur after some cognitive responses, that were induced by some external stimulus. As stated in hypotheses 1a to 1c, the three extrinsic cues will trigger some cognitive responses, such as perceptions of quality. Since a consumer should like a product of high quality more than a product of low quality, liking should be positively influenced by perceived quality. Hence we hypothesize (see Figure 2B):

H2d: When buyers' perceptions of quality are more favorable, their liking (second phase) toward a product is greater.

In addition, consumers may compare the observed price with their internal expected prices. If their expected price is greater than the observed price, then the perceived good deal will elicit some positive affective responses. Therefore, we hypothesize (see Figure 2B):

H2e: For a given observed price, when the expected price is higher, buyers' liking (second phase) toward a product is greater.

H2f: For a given expected price, when the observed price is higher, buyers' liking (second phase) toward a product is lower.

Third phase affective responses. As the cognitive process continues, consumers may evaluate the perceived value of a product. This cognitive response may then elicit the third phase affective responses. Therefore, we expect (see Figure 2C):

H2g: As buyers' perceptions of value increases, their liking (third phase) toward a product is greater.

Effects of affective responses on cognitive responses. According to Cohen and Areni's (1991) model, an affective response can also activate some cognitive responses, because consumers may want to justify their preferences by some subsequent cognitive responses (see also Zajonc 1980). Based on this argument, if a consumer likes a product, s/he may justify this preference by considering the product is of higher quality, or is of higher value-for-money. To further extend this reasoning, the affective responses may directly influence one's buying intentions. It should be pointed out that an affective response can only affect the subsequent (but not the preceding) cognitive responses. That is, the first phase liking may affect the subsequent perceptions of quality; however, the second and third phases liking cannot influence the preceding perceptions of quality. Hence, we hypothesize:

H2h: As buyers' liking (first phase) toward a product increases, their perceptions of quality increase (see Figure 2A).

H2i: As buyers' liking (first or second phase) toward a product increases, their perceptions of value increase (see Figures 2A and 2B).

H2j: As buyers' liking (first, second, or third phase) toward a product increases, their willingness to buy increase (see Figures 2A, 2B and 2C).

It should be noted that buyers' liking toward a product can occur in the first, second, or third phase. However, since little research has been done in this area, which phase of liking may occur remains an empirical question. Moreover, since we will measure liking only once in the experiment described below, the liking measurement can only represent the affective response which is the strongest one among the three phases. Therefore, we expect that the affective responses described in either (1) phase 1 (hypotheses 2a, 2b, 2c, 2h, 2i, and 2j), or (2) phase 2 (hypotheses 2d, 2e, 2f, 2i, and 2j), or (3) phase 3 (hypotheses 2g, and 2j), will be supported.

METHOD

To investigate these hypotheses, price, brand and country of origin information were provided to subjects with short product descriptions about two electronic products (CD stereo system, and coffeemaker). We conducted a pilot study and a main study.

Pilot Study

In the pilot study, 82 subjects were asked to suggest a price (that they would expect to see in the marketplace) for a coffeemaker and a CD stereo system. [A total of 18 subjects were screened out due to the following reasons. First, subjects who suggested a price that was beyond "3 standard deviations were regarded as outliers and therefore screened out. Second, subjects who could not recognize the brand as a Japanese brand were discarded. Third, foreign students were also screened out, since they might not be familiar with the market price.] These products were described as either with well-known (Sony and Toshiba) or unknown brand names (Mishita and Yichiban), and manufactured in a favorable (Japan) or an unfavorable (Mexico) country. Pictures of the two products were provided in order to make the stimuli more concrete.

The major purposes of the pilot study were to obtain the expected price of the products, as well as to verify the effectiveness of the selected countries and brand names. This information was later used as the price, country of origin, and brand manipulations for the main study.

FIGURE 2A

COGNITIVE-AFFECTIVE MODEL OF BUYING INTENTIONS (PHASE 1)

FIGURE 2B

COGNITIVE-AFFECTIVE MODEL OF BUYING INTENTIONS (PHASE 2)

Based on the results of the pilot study, the average expected price for the CD stereo system was $399, and for the coffeemaker was $69. These two prices were used as the medium price levels of the two products in the main study. The lowest expected prices suggested by subjects for the CD stereo system and coffeemaker were $179 and $29, respectively, and were used as the low price levels of the two products. These two prices were $220 and $40 lower than the average prices. To obtain a symmetrical manipulation, $619 (i.e., $399+$220) and $109 (i.e., $69+$40) were set as the high price levels for the main study. Also, the known brand was perceived as of higher quality than the hypothetical unknown brand (X=8.00 vs. 5.18, F(1,38)=78.73, p<.01, for CD stereo system; X=6.70 vs. 4.93, F(1,29)=14.04, p<.01, for coffeemaker). In addition, products from Japan were perceived to be of higher quality than products from Mexico (X=7.62 vs. 4.08, F(1,59)=161.93, p<.01, for CD stereo system; X=6.71 vs. 4.60, F(1,57)=65.48, p<.01, for coffeemaker).

Main Study

In the main study, price, brand, and country of origin information were manipulated; cognitive product evaluations (perceived quality, perceived value) and affective product evaluation (liking), buying intentions and expected prices for the two products were measured.

Design. The main study was a 2 (Japan vs. Mexico) x 2 (known vs. unknown brand) x 3 (low, medium, high price level) between-subjects factorial design, with replication across two products. Again, the brands used were either real well-known Japanese brand names (Sony, Toshiba) or hypothetical unknown brand names that appeared to be Japanese (Mishita, Yichiban). To ensure subjects would be able to perceive them as Japanese brand names, subjects were explicitly told that the brands are "famous manufacturers in Japan" for well-known brands; and just "manufacturers in Japan" for unknown brands. As in the pilot study, pictures of the products were provided with the product descriptions to make the stimuli more concrete.

FIGURE 2C

COGNITIVE-AFFECTIVE MODEL OF BUYING INTENTIONS (PHASE 3)

Sample. A total of 134 students in an introductory business course at a major mid-western university participated in this study. However, a total of eight subjects were screened out because their expected price of the products were located out of the +3 standard deviation range. A total of 126 usable cases were used in the following data analyses.

Dependent measurements. First, subjects were asked to suggest an expected price (a specific price point or a price range, if they wished) for each product described. Then, after reading the information about the Japanese brands, they were asked to suggest an expected price for each product again. The objective of this second measure was to detect the effect of knowing the brand's country of origin on expected price, if there was any. Third, subjects were asked to evaluate the products' quality, value, liking, and buying intentions. Each construct was measured by 5 items, on 9-point semantic differential scales. Fourth, demographics such as gender, education level, and subjects' nationality were recorded.

Manipulation checks. Subjects were asked to evaluate the perceived quality of the two products from the two countries, as well as of different brand names using 9-point scales (ranging from very poor to very good). Since hypothetical brand names were used here, subjects were allowed to use 0 to represent they really had no idea about the perceived quality. Subjects were also asked to evaluate the level of technology necessary to manufacture the two products. Nine-point scales, ranged from "not at all" to "requires a lot", were used.

RESULTS

Manipulation Checks

All the manipulations were successful. First, products from Japan were perceived as of significantly higher quality than products from Mexico (X=7.95 vs. 3.84, F(1,108)=409.13, p<.01, for CD stereo system; X=7.37 vs. 4.25, F(1,104)=213.38, p<.01, for coffeemaker). Second, products of the known brand were perceived as of higher quality than products of unknown brand (X=7.87 vs. 4.76, F(1,62)=113.12, p<.01, for CD stereo system; X=6.50 vs. 4.65, F(1,59)=35.05, p<.01, for coffeemaker). Finally, subjects perceived the level of technology required in manufacturing a CD stereo system to be higher than that for coffeemaker (X=7.61 vs. 5.42), F(1,124)=298.63, p<.01).

Reliability Checks

The Cronbach's alpha for the five items measuring perceived quality, perceived value, liking, and buying intentions, for both products, ranged from .90 to .94, suggesting these items were highly internal-consistent. Therefore, the average ratings of the 5 items of each variable were used to represent the variables.

Country of Origin, Brand, and Price Effects on the Dependent Variables

Since there are four dependent variables (perceived quality, perceived value, liking, and buying intentions), a 4-way mixed MANCOVA was performed to reveal all the main and interaction effects (with country of origin, brand, and price being the between-subjects factors; product being the within-subjects factor; and expected price being the covariate). Results reported in Table 1 show that expected price has a significant covariate effect on the dependent variables. Moreover, the results indicate that only the country of origin, brand, price, and product main effects on the four dependent variables were significant, and no interaction effect was significant. Therefore, only main effects will be considered in the subsequent model testing.

Criteria used for Testing Models

LISREL VI's maximum likelihood estimation procedure was used to analyze the four models (Joreskog and Sorbom 1986). The cognitive model (Figure 1B) and the three cognitive-affective models (Figure 2A, 2B, and 2C) were tested by estimating the goodness-of-fit between the hypothesized models and the observed data. [It should be noted that all four models are viable. Our objective is to identify the model that best fit with the data, among the four models. It is not our intention to challenge the cognitive model with three different competing explanations.] Each model was first tested by using coffeemaker as the product, and replicated by using CD stereo system as another product.

TABLE 1

MANCOVA RESULTS OF THE MAIN STUDY

Several indices were considered when assessing the likelihood that the data fit the hypothesized models. They include goodness-of-fit statistics, (i.e. the overall chi-square measure, the goodness-of-fit index, and the adjusted goodness-of-fit index), the root-mean-square residual (RSR), and the percentage of variances of buying intentions being explained by the model.

Testing the Cognitive and Cognitive-Affective Models of Buying Intentions

The LISREL results of all four models, for both products are reported in Table 2. First, the goodness-of-fit statistics (X2, GFI, AGFI) suggest that both the cognitive model and the cognitive-affective model for the second phase fit the data adequately, but the cognitive-affective models for the first and the third phases do not. Second, the RSR residuals for all four models reflect that the hypothetical covariance matrices do not deviate substantially from the observed covariance matrices. Third, among the two adequately fitted models, the cognitive-affective model for the second phase has higher predictive power on buying intention than the cognitive model. To sum up, the results suggest that the cognitive-affective model for the second phase fits best with the data. The path coefficients of the two adequately fitted models are reported in Figures 3A and 3B.

DISCUSSION

Extrinsic Cues

The results from MANCOVA and LISREL analyses revealed some interesting findings about the three extrinsic cues. First, all three extrinsic cues affected the perceived product quality, which implies that subjects did rely on these extrinsic cues to make quality evaluations. Second, among the three extrinsic cues, while price of the product had significant direct effects on liking and perceived value, brand and country of origin had no significant direct effects on them. [It is possible that the coupling of brand and country of origin information (e.g. Sony and Mexico) earned little credibility, which made brand and country of origin have no direct effect on liking and perceived value. Future research should measure this information.] This finding suggests that price was the major consideration in subjects' product evaluations, although country of origin and brand name might have some indirect effects via perceived quality. However, since this finding may be limited to the student sample used, future research should validate the present finding using a different sample of subjects.

Expected Price

The results concerning expected price provided another interesting line of research. First, the MANCOVA results indicated that expected price had a significant influence on product evaluations. The LISREL analyses clarified that expected price significantly influenced liking and perceived value, but not perceived quality. Moreover, the LISREL results suggest that expected price and price together influenced both liking and perceived value of the product. This finding is consistent with the transaction value literature that consumers compare their internal reference price with the external observed price to evaluate the transaction value (Thaler 1985; Monroe, Grewal, and Compeau 1991). The liking here probably represents the subjects' positive affective responses towards the good deal. Future research should explore if country of origin, or brand names, affect consumers' expected price, which in turn may affect perceived value or even buying intentions.

TABLE 2

LISREL RESULTS FOR THE FOUR MODELS OF BUYING INTENTIONS

Cognitive versus Cognitive-Affective Models of Buying Intentions

It is interesting to compare the cognitive model of buying intentions (Figure 3A) with the cognitive-affective model of buying intentions for the second phase (Figure 3B). First, the data confirmed all seven paths of the cognitive model, in terms of statistical significance as well as predicted directions, which explained 47.9% to 64.0% of the variance in buying intentions, for coffeemaker and CD stereo system, respectively. This result replicated the findings reported in the literature (for example, Dodds et al. 1991; Zeithaml 1988; Rao and Monroe 1989). Second, the results also indicated that the cognitive-affective model for the second phase fits the data very well. All 12 paths, except one, were statistical significant with the expected direction. The only non-significant one was the perceived quality- perceived value path, which suggests that perceived quality had no direct effect on perceived value. However, the significant perceived quality-liking, and liking-perceived value paths suggest that perceived quality had an indirect effect on perceived value, and was mediated by liking. This finding reveals that it is important to include an affective component in the buying intention model. This model explained 63.6% and 77.5% of the variance in buying intentions, for coffeemaker and CD stereo system, respectively. This pattern of results suggested that the cognitive-affective model not only can postulate the specific relations among the various cognitive and affective components, but also seems to be a better predictor of buying intentions, explaining an additional of 13.5% to 15.7% of the variance in buying intentions.

Further, the present findings also indicate that liking is similar in strength with perceived value in influencing buying intentions (path coefficients being .44 vs. .50 for coffeemaker, and .41 vs. .61 for CD stereo system). This result implies that the "affective" component is almost as important as the "cognitive" component in influencing buying intentions, which until recently has been ignored in buying intention studies. Inspired by this finding, future research on product evaluations and buying intentions should pay more attention to the affective component in addition to the cognitive component.

Another interesting issue remaining unanswered is why the cognitive-affective model for phase 1 and 3 is not supported, but that for phase 2 is supported. We speculate that it is due to the products (coffeemaker and CD stereo system) used in this study. The products considered in this study are neither extremely "affective" nor "cognitive", therefore they may provide little support for the cognitive-affective models of phase 1 (affective responses occurred first) and phase 3 (cognitive responses occurred first). It would be interesting to replicate this study by using affective products (e.g. perfume, wedding rings) and cognitive products (e.g. computers, dictionaries), and see if the cognitive-affective models for phase 1 and phase 3 will be supported.

To conclude, the proposed cognitive-affective model of buying intention (phase 2) is generally supported across two products. It is found that the three extrinsic cues affect perceived quality in a similar fashion. However, while price and expected price have direct effects on liking and perceived value, country of origin and brand name do not have a direct significant effect on them. Furthermore, the effect of perceived quality on perceived value is mediated through liking. Finally, liking toward a product and perceived value seem to be of similar importance in predicting buying intention.

FIGURE 3A

RESULTS FOR THE COGNITIVE MODEL OF BUYING INTENTIONS

FIGURE 3B

RESULTS FOR THE COGNITIVE-AFFECTIVE MODEL OF BUYING INTENTIONS

REFERENCES

Bilkey Warren J. and Erik Nes (1982), "Country-Of-Origin Effects on Product Evaluation," Journal of International Business Studies, 13(spring/summer), 89-99.

Cohen, Joel B. and Charles S. Areni (1991), "Affect and Consumer Behavior," in Handbook of Consumer Behavior, eds Thomas S. Robertson and Harold H. Kassarjian, Prentice Hall, NJ, 188-240.

Dodds, William B., Kent B. Monroe and Dhruv Grewal (1991), "Effects of Price, Brand, and Store Information on Buyers' Product Evaluations," Journal of Marketing Research, 28(August), 307-19.

Joreskog, Karl and Dag Sorbom (1986). LISREL VI: Analysis of Linear Structural Relationships by Maximum Likelihood and Least Square Methods. Mooresville, IN: Scientific Software, Inc.

Li, Wai-kwan and Kent B. Monroe (1992), "The Role of Country Of Origin Information on Buyers' Product Evaluations: An In-depth Interview Approach," in AMA Educators' Proceedings: Enhancing Knowledge Development in Marketing, Volume 3, eds. Robert P. Leone and V. Kumar, Chicago, 274-280.

Monroe, Kent B. (1973), "Buyers' Subjective Perceptions of Price," Journal of Marketing Research, 10(February), 70-80.

Monroe, Kent B., Dhruv Grewal, and Larry D. Compeau (1991), "The Concept of Reference Prices: Theoretical Justifications and Research Issues," Working paper, University of Illinois.

Monroe, Kent B. and R. Krishnan (1985), "The Effect of Price on Subjective Product Evaluations," in Perceived Quality: How Consumers View Stores and Merchandise, eds. Jacob Jacoby and Jerry C. Olson, Lexington, MA, 209-232.

Ozsomer, Aysegul and S. Tamer Cavusgil (1991), "Country-Of-Origin Effects on Product Evaluations: A Sequel to Bilkey and Nes Review," in AMA Educators' Proceedings: Enhancing Knowledge Development in Marketing, Volume 2, eds. Mary Gilly et al., Chicago, 269-277.

Rao, Akshay R. and Kent B. Monroe (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Perceptions of Product Quality: An Integrative Review," Journal of Marketing Research, 26(August), 351-7.

Scitovszky, Tibor (1945), "Some Consequences of the Habit of Judging Quality by Price," Review of Economic Studies, 12(Winter), 100-105.

Thaler, Richard (1985), "Mental Accounting and Consumer Choice," Marketing Science, 4(Summer), 199-214.

Winer, Russell S. (1988), "Behavioral Perspective on Pricing: Buyers' Subjective Perception of Price Revisited," in T. M. Devinney, ed., Issues in Pricing: Theory and Research, Lexington, MA: Lexington Books, pp.35-58.

Zajonc, Robert B. (1980), "Feeling and Thinking: Preferences Need No Inferences," American Psychologist, 35 (February), 151-175.

Zajonc, Robert B. and Hazel Markus (1982), "Affective and Cognitive Factors in Preferences," Journal of Consumer Research, 9(September), 123-131.

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence," Journal of Marketing, 52 (July), 2-22.

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