Product Familiarity, Information Processing, and Country-Of-Origin Cues

ABSTRACT - Past research on consumers' propensity to use 8 product's "made in" label shows conflicting findings regarding the role of product familiarity. When evaluating products, are country-of-origin cues used more by individuals familiar with the products or by novices? By interpreting the country-of-origin cue as a heuristic or proxy for intrinsic product attributes and adopting a simplifying information processing framework, the present study develops a theoretical rationale for why individuals familiar with a product class would make greater use of the product's "made in" label. The developed model is tested with empirical data.

Citation:

Arthur E. Heimbach, Johny K. Johansson, and Douglas L. MacLachlan (1989) ,"Product Familiarity, Information Processing, and Country-Of-Origin Cues", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 460-467.

Advances in Consumer Research Volume 16, 1989      Pages 460-467

PRODUCT FAMILIARITY, INFORMATION PROCESSING, AND COUNTRY-OF-ORIGIN CUES

Arthur E. Heimbach, University of Washington

Johny K. Johansson, University of Washington

Douglas L. MacLachlan, University of Washington

[The authors acknowledge financial support from the schools' Research Fund and from the last two authors' Affiliate Program professorships.]

ABSTRACT -

Past research on consumers' propensity to use 8 product's "made in" label shows conflicting findings regarding the role of product familiarity. When evaluating products, are country-of-origin cues used more by individuals familiar with the products or by novices? By interpreting the country-of-origin cue as a heuristic or proxy for intrinsic product attributes and adopting a simplifying information processing framework, the present study develops a theoretical rationale for why individuals familiar with a product class would make greater use of the product's "made in" label. The developed model is tested with empirical data.

Published research demonstrating the important role played by "made-in" labels in consumers' product evaluations has by now reached an age of acceptance and maturity (Bilkey & Nes, 1982). Although the exact workings of the country-of-origin cues remain controversial, there is little doubt about the basic fact that "made-in" labels powerfully influence people's perceptions and attitudes towards products and brands (Erickson, Johansson, and Chao 1984). On the other hand, there has been almost no published research on the question of which consumers are more likely to make use of country-of-origin cues or exactly how the cue is used in the evaluation process.

One of the earliest explanations for the effect of "made in" labels on product evaluation is that where little direct information about product attributes is available, more or less relevant indirect evidence (like the country-of-origin) is employed. In fact Schooler (1965), in the first country-of-origin effect study, relied on this assumption in his choice of products that had very few discernible attributes, thus enabling the subjects' biases to be projected onto the products to be evaluated. This has led to the notion that country-of-origin effects are strongest among consumers who have little or no product familiarity (e.g. Nagashima 1970). Such an argument is consistent with the usage of brand names described by Bettman and Park (1980). Further, many of the early studies showed strong country of origin imagery effects when intangible product concepts or general product categories were presented for evaluation (e.g., Reierson 1966; Nagashima 1970, 1977; Lillis & Narayana 1974).

It was then only a short step to the argument that the effect was in fact spurious, and that "made-in" labels merely substituted for the attributes omitted in the particular study. Partial evidence in support of this induction was provided by Schooler (1971). It follows then, that country of origin might merely be a proxy variable for missing product characteristics, only having value when more conclusive evidence is unavailable.

This approach leads to the notion that country of origin effects are strongest among consumers who have little or no product familiarity. In the world of imperfect information on actual product attributes, individuals are forced to rely on stored knowledge. Since people with less familiarity have a lower supply of such stored information, they end up having to rely more heavily on the stored stereotypic biases they have toward the specific country identified in the "made-in" label.

Such an argument, however, is inconsistent with the greater usage of brand names among people with high familiarity described by Bettman and Park (1980). Furthermore, recent studies have yielded results involving product familiarity and use of the country-of-origin cue that are at odds with the low familiarity hypothesis. Johansson, Douglas dc Nonaka (1985), for example, uncovered a positive correlation between product familiarity and influence from country-of-origin. Similarly, Johansson & Nebenzahl (1986) reported a positive and significant (.23) correlation between "Knowledge about the product class" and "Importance of country-of-origin."

These later studies suggest that individuals who consider themselves- familiar with the brands in a product class are more willing to let country-of-origin cues enter their evaluations than are people who are less familiar with the product class. Therefore, while the earlier perspective suggests that country-of-origin influences are "unwarranted biases" employed by the less familiar, the more recent findings indirectly suggest that country stereotypes are used more by people who should know better. This somewhat contradictory circumstance deserves further study, both to more clearly assess the direction of the effect that product familiarity has on country-of-origin cue usage as well as for developing a greater understanding of the causal mechanisms involved in their relationship.

The purpose of this research is to try to reconcile the conflicting findings in the country-of-origin literature and to solve the apparent paradox of positive correlation between product familiarity and use of the country-of-origin cue. In this study, we will provide evidence that the use of "made-in" labels is partially a function of the interaction of two components: product class familiarity and the perceived variance in quality between the products marketed from within a single country and across all the countries offering products in the product class.

CONCEPTUAL FRAMEWORK

Two conceptual approaches are integrated for understanding the linkages of product familiarity and use of the country-of-origin cue. First, a simplified information processing paradigm is introduced whereby consumers learn to assess the information value (in a sense, the validity) of the country-of-origin cue in order to minimize the processing effort involved in product evaluations. Second, through repeated product exposures, increased product familiarity is suggested to instill a sense of reliability in the assessed information value of the cue, thereby increasing the confidence in using the cue. This then provides the driving force behind the familiarity-cue usage effect.

Simplified Information Processing

One way to view country-of-origin is as a cue useful in the simplified information processing consumers employ in low involvement settings (Wright 1975). In order to handle complex processing tasks, consumers often attend only to a few features of- the total bundle available for evaluation. One of the strategies employed is to look for "summary statistics" which encompass all or most of a set of correlated attributes. The most common of these summaries is the brand name. The country-of-origin cue, or "made in" label, could be employed as another such summary statistic.

Under this interpretation, the country-of-origin cue is used as a mental "short-cut", a way of avoiding unnecessary information processing. "If it is made in Hong Kong it is a fake" or "It's a Swiss watch--what else do you need to know?" are sentiments which exemplify the process of judging the product by a summary cue. Use of the country-of-origin cue as such a summary statistic, however, rests on there being some minimum level of perceived correlation between the cue and one or more relevant product attributes. In other words, consumers must somehow assess the information value of the cue--its predictive value.

Predictive Value

As discussed above, a cue, in order to be useful in simplifying information processing, must first be perceived as a valid predictor of some product attributes. For example, if people perceive various countries to be equally capable producers (perhaps because of standardized manufacturing technology), there would be little justification for attending to the "made-in" label. The cue, in this case, is not "informative." Thus, the first criterion for expecting the country-of-origin cue to be useful would be for consumers to perceive variations across countries. It is this perceived between-country variability which gives information content to the country-of-origin "summary statistic."

Cox (1965) introduced a similar notion in a cue theoretic perspective, labeling the perceived information value as the cue's predictive value. Cox posited that the use of a cue in product evaluations is directly related to the perceived probability that the cue is associated with a specific product attribute. In the country-of-origin case, the belief that a set of product attributes (e.g., workmanship, material durability and product reliability) vary consistently across countries gives meaning to the cue, making it useful as a summary statistic for evaluating the product (e.g., on overall quality).

However, it is easy to see that when the variations in quality differences are large for the brands offered from a single country, the use of the "made-in" label will be inhibited rather than encouraged. For example, if American brands of the product vary widely in quality, the "Made in America" label will carry little predictive value as to quality. Conversely, where a country generally is perceived to possess brands within a narrower band of quality variations, the label is indeed useful. Shoes "Made in India" might be an informative label on this account, as might be "Swiss" watches.

Thus the incentive to pay attention to the "made-in" label varies (1) directly according to perceived variations in mean product quality across countries, and (2) inversely with variations of brands within a country. Pursuing the notion of "summary statistic" one can see that the ratio of the across country or "between" quality variations and the "within" variations can be viewed as an "F-statistic". Kelly (1973): employed this concept by suggesting that individuals calculate a "naive F" statistic in order to make causal attributions. The higher the F, the higher the predictive value of the cue. Similarly the "between"/"within" statistic can be thought of as a type of signal-to-noise ratio. When perceived "within" country quality variations are large relative to the "between" country differences, the noise is high, the F-value and the predictive value fall--the "made-in" labels convey very little information.

The relationship between this interpretation of a cue's predictive value and the individual's familiarity with the product class is not straightforward. First, it is not clear that product class familiarity is even a necessary prerequisite for an individual to be able to calculate an F-statistic. The beliefs of between and within country variances could be made through naive assumptions or inferences from other product classes. Further, even if it were assumed that at least a minimal amount of exposure was needed to establish estimates of the within and between variances, greater product familiarity, per se, would not necessarily increase the F-statistic nor lead to greater use of a cue. Instead, we argue that greater product familiarity will improve the true accuracy of the consumer's naive F-statistic, thereby allowing the individual to be more confident in his or her use of the cue.

Product Familiarity and Confidence in Cue Usage

Product familiarity, typically defined as the "number of product-related experiences that have been accumulated by the consumer" (Alba and Hutchinson 1987; p. 411), has been shown to have a complex role in product attitude formation and information processing. In particular, previous research suggests that product familiarity influences the choice and use of summary cues (such as price and brand) in the assessment of product evaluations. Park and Lessig

FIGURE A

CUE USAGE AS A FUNCTION OF THE INTERACTION BETWEEN THE CUE'S CONFIDENCE WALUE AND PREDICTIVE VALUE

(1981) reported that a high product-class-familiar group was significantly more confident in their use of price and brand cues for making satisfaction maximizing product choices than were low- and medium-familiarity groups. Wilkinson et al. (1984) reported further empirical evidence of a familiarity confidence relationship for the price cue. These results suggest a causal mechanism of product familiarity leading to confidence in one's use of summary cues followed by greater usage of the cues.

The linkage between familiarity and confidence in cue usage is also logically consistent with theories of covariation assessment and inferencing. Covariation assessment research (Bettman, Roedder John and Scott 1986) suggests that increased product familiarity reinforces the "real world" or actual correlations between product attributes and summary cues. Theories of inferencing (Alba and Hutchinson 1987) suggest that as familiarity increases, the inferences made from the cues become more veridical. When knowledge is low, inferences will still be made; however, they will also be less trusted since there is, by definition, less supporting evidence for the inference. Both of these perspectives support the proposition that increased familiarity increases confidence in the use of a cue.

Cox (1962) and Olson and Jacoby (1972) suggested a framework for the relationships between confidence in the use of a cue, the predictive value of a cue and the actual usage of a cue. In their framework, the degree to which a cue is used in a product evaluation is= a function of the interaction between one's assessment of a cue's predictive value and his or her confidence in the use of that cue (termed the cue's confidence value). Thus the cue's confidence value has a multiplicative moderating role (what Sharma, Durand and Gur-Arie (1981) define as a pure moderator) where increased confidence enhances the effects of predictive value in leading to use of a cue. In other words, cues perceived to have high predictive value are used to a greater extent when confidence in the use of the cues is also high. It follows then that consumers with low confidence in their cue usage will, across the board, make less use of cues, regardless of their assessment of any particular cue's predictive value. In the extreme, complete lack of confidence would lead to complete disregard for a cue. This relationship can be seen in Figure A.

The distinction between the confidence value of the cue and the predictive value of the cue can be seen clearly through another statistical analogy. Whereas the proposed naive F-statistic is a point estimate of the predictive value of the cue, the confidence value of a cue describes the error term associated with the estimated F value. The greater the familiarity, the lower the error term and therefore the greater the confidence in using the cue.

It is important to distinguish between product familiarity and expertise. Expertise is defined by Alba and Hutchinson (1987, p.411) as "the ability to perform product-related tasks successfully." Product familiarity is a necessary but not sufficient condition for consumer expertise. "True" experts may not use cognitive simplification strategies. There is plenty of evidence, however, that one major benefit of product familiarity is the reduction of cognitive effort expended in making product judgments (Alba and Hutchinson, 1987).

FIGURE B

To summarize, consumers, for the most part, can be seen as misers of cognitive resources and therefore employ summary statistics to simplify their evaluation tasks. The ratio of "between" to "within" country variances in product quality is presented as one such statistic in that it measures the "made-in" label's predictive value--an estimate of the degree to which the country of origin information summarizes a set of product attributes. Increases in product familiarity are expected to increase the consumer's confidence in the use of the cue, thereby moderating the effect of the cue's predictive value on cue usage. Finally, we suggest that at any level of predictive value for the country-of-origin cue, greater product familiarity will lead to greater use of the country-of-origin cue.

The Model

The driving effect of the perceived quality variations on the propensity to use the country-of-origin cue and the interactive role of familiarity can be portrayed in a simple diagram, Figure B.

As shown in the diagram, the hypothesized linkage between perceived quality variability and use of the country-of-origin cue is moderated by the degree of familiarity. As the predictive value increases, so does the propensity to use the cue--the basic relationship is positive. The moderating effect from familiarity is also positive, the cue utilization increasing as confidence goes up.

Hypotheses. There are three hypotheses that guide our research:

H1: The degree of country-of-origin cue utilization in attitude judgments is positively correlated with the perceived information value of the cue--i.e., the cue's predictive value.

H2: The degree of country-of-origin cue utilization in attitude judgments is positively correlated with a person's familiarity with the product class under consideration--i.e., the cue's confidence value.

H3: The degree country-of-origin cue utilization in attitude judgments is a function of the interaction between the cue's predictive value and its confidence value.

Notice that the hypotheses concern cue utilization in making overall attitude judgments about the brands, not in making actual choices among the brands. Cue utilization would be expected to differ depending on whether one is investigating judgment or choice. See, for example, Johnson and Russo (1984) and Bettman and Park (1980), who found evidence for an inverted-U relationship between product familiarity and information acquisition in choice situations.

METHOD

Survey Questionnaire

To test the proposed linkages a survey design was employed. The target population consisted of teaching and research assistants (male and female) at a west coast school. The survey obtained respondents' evaluations of automobiles produced in three different countries (Germany, Japan, and the U.S.). It was expected that respondents would exhibit wide variations in familiarity with cars and also in their perceptions of the quality variations "between" and "within" these countries.

The questionnaire consisted of three parts. In the first section the respondents were asked to rate three described autos on expected price, overall quality, and interest in purchase. The profile descriptions of the cars were kept deliberately vague, incorporating fifteen attributes, so as to allow the country-of-origin cue to have differential impact across individuals. The order of the descriptions and the assignment of country labels (Germany, U.S., and Japan) were randomized across respondents.

A brief section on background variables, including ethnic background and country experience followed. The questionnaire elicited the makes of cars owned in the last ten years and asked for makes of cars which the respondent associated with each of the three countries (unaided recall). This section ended with an exercise in which the respondents named and rated the best and the worst car from each country in terms of "overall quality". This exercise generated the data necessary to measure both within and between country variability.

The last section of the questionnaire covered questions about familiarity with car magazines, autos from different countries, and ended with a self-reported assessment of the degree to which the country-of-origin information was used when rating the initial three automobiles.

Data

The data were collected during the winter term 1987. The mailout comprised 2226 questionnaires and, after eliminating those potential respondents who had moved and those who did not respond by the end of the term, the usable number of questionnaires returned was 541, resulting in a 24 percent response rate. After further eliminating those with incomplete responses (the country variability exercise created fatigue problems) the actual number of usable responses was 258. These were the responses used for the present research.

The incomplete responses to the country variability questions might have created a nonresponse bias. Fatigued respondents would be likely to have a relatively low level of involvement and familiarity with cars from different countries, making the exercise seem tiresome and pointless. Thus, the median level of familiarity in the final data base is probably biased upwards. This is a correlational study, however, and unless all nonfamiliar respondents have been eliminated (so that the familiarity distribution has been truncated), one would still expect the relationship between familiarity and the other constructs to remain unbiased. (Anticipating the results, the familiarity variable did in fact exhibit a full range from 0-10 in the data as in the questionnaire .)

MEASUREMENT

Familiarity

The familiarity measures were elicited for cars from each of the three countries separately. The responses were in the form of bipolar adjective items ("low" to "high") on a scale from zero to ten. The specific question asked was "In general, how much familiarity do you believe you have with: USA autos, German autos, Japanese autos?" These scales were then averaged to obtain an overall familiarity measure.

Such a self-report measure has been commonly used in the product familiarity research done to date. For example, Johnson and Russo (1984) employed a five-point familiarity scale (also for automobiles). On the other hand, Park and Lessig (1981) augmented such a five-point familiarity rating with self-reports regarding prior behavioral conditions such as the subjects' search experience, usage experience, and ownership status for the product under study. Similarly, in the present study responses were elicited on readership of auto magazines and the number of correctly recalled auto makes from various countries.

Quality Variations

The within and between country variations were computed from the responses given for quality ratings for the best and worst brands from each country. "Within" country quality variation was computed by taking the mean of the two quality ratings (scaled between 0-low and 10-high) for the best and worst brand within each country, then subtracting the average from the reported best and worst quality values. Squaring and summing these two differences yields a standard sum-of-squares within measure.

The "between" country variation measure was computed by using the average brand quality value (of best and worst) computed above and averaging these over the three countries to obtain an overall quality value. Then, in the same manner as described above, a between sum-of-squares was computed as the between i measure.

Use of Cue

"Use of the country of origin cue"--the dependent variable--was measured by the self-reported assessment at the end of the questionnaire and scored between 0-low and 10-high. The specific question was "Lastly, to what extent did you use the country of manufacture information in your assessment of the automobiles described in the first part of this survey?"

Validity and Reliability

Familiarity. The measurement reliability of the product familiarity variable was tested through assessing the internal consistency of the three country specific product familiarity variables that were used to compute the overall variable. The Cronbach alpha for this analysis was .80, a quite acceptable value indicating that the average familiarity variable is reliable.

Multiple measures of product familiarity collected in the survey included (1) readership of auto magazines and articles on cars from Consumer Reports and the local newspapers and (2) the total number of correctly recalled brands from the three target countries (Japan, USA and Germany). The readership variables were combined into an average readership score and correlations between the familiarity measure and the two alternative measures were computed (see Table 1). The results evidence convergent validity, with each of the correlations being positive and statistically significant.

Use of the Cue. The reliability of this measure was tested for stability across sub-samples of the respondents. Two sub-samples were drawn randomly, one a fifty percent sample, the other a twenty percent sample. Comparing the values for the total sample against those of the sub-samples in terms of means and standard deviations, the differences were negligible (means: 5.14, 5.20, 5.00; SDs: 2.76, 2.73, 2.94, respectively).

Although some demand effect bias might exist in this measure, since it was taken at the end of the survey, one would expect that, if present, the bias would work against the hypothesized effects. That is, one would expect people familiar with automobiles to report less reliance on the country-of-origin cue than others, because they might have rehearsed or guessed what the attributes were

TABLE 1

PAIRWISE CORRELATION OF MULTIPLE FAMILIARITY MEASURES

ESTIMATION RESULTS

Descriptive Statistics

The simple correlation between familiarity and the use of cue variables equalled .21, very similar to the .23 value reported by Johansson & Nebenzahl (1986). The respondents more familiar with automobiles tended to use the country-of-origin cue more than other respondents did. Thus, the main result from previous work is replicated.

The mean levels, standard deviations, the minimum and maximum values of the relevant variables are given in Table 2. All measures had reasonably high levels of variation among sample elements. Some respondents obviously considered themselves very familiar with cars, others not at all. The wide variability in the quality variations was especially heartening, since only one product class was studied. One might have feared that the perceived quality variations would be similar across all respondents, but this was clearly not the case.

Regressions

Table 3 presents the results of the regression analyses used to test the research hypotheses. In part a of the table, we see that both H1 and H2 are supported. That is, use of the country-of-origin cue is positively associated with both Between/Within variation (the cue's predictive value) and product familiarity (the cue's confidence value). The regression coefficients are highly statistically significant.

When both main effects and the interaction term are entered in the equation (Table 3b), all coefficients become relatively insignificant (i.e., small standardized beta values and high significance probabilities), but the interaction term captures most of the variation in the use of the cue. Adjusted R2 increases somewhat. It is apparent that multicollinearity is obscuring the true relationship between the variables on the left and right sides of the equation. (Indeed, the simple correlation between familiarity and the interaction variable is 0.74.)

The obvious next step is to regress use of the cue on the interaction variable alone. The result is given in Table 3c. This model captures the largest explanation of variation in use of the cue (i.e., it has the highest R2), the standardized coefficient being large and significant. Therefore, hypothesis H3 is supported. This is evidence that use of the country-of-origin cue depends interactively upon product familiarity and the particular cue's predictive value.

DISCUSSION

The empirical results from the study have supported the notion that country-of-origin labels can be interpreted as cues which are used more by people with higher product familiarity. Because of the nature of the sample and the limitation to one product class, one can of course not infer that country-of-origin effects are of the same magnitude in other contexts. There is no particular reason to suppose, however, that the strength of the uncovered relationship between familiarity and use of the cue is not translatable to other situations. Allowed to interact with the predictive value of the cue, the explanatory power of familiarity amounted to about 10% of the observed propensity to use the "made-in" label, not a large percentage but quite significant.

The view of the "made-in" label as a simplifying proxy for underlying product attributes is well-supported by these results. So too is the distinction between a cue's predictive and its confidence value. The two constructs are empirically separable and the results validate previous theoretical work in the area (Cox, 1962; Olson & Jacoby 1972).

The low R-squares suggest that other factors might be equally or more important than familiarity in explaining the use of the country-of-origin cue. But there are no obvious variables omitted that would seem likely to render the estimated relationship spurious. Moreover, cross-sectional data across individual respondents are notorious for yielding weak explanatory relationships--there are simply too many individuals with idiosyncratic behavior.

TABLE 2

DESCRIPTIVE STATISTICS OF RELEVANT VARIABLES

TABLE 3

REGRESSION RESULTS

Alba and Hutchinson (1987) have described convincingly the complexity of consumer expertise and have shown how product familiarity, as commonly defined, cannot capture this complexity. They argue, as is done here, that "the effects of knowledge on consumer behavior cannot be regarded only as main effects and must be studied along with a wide range of moderating variables." (p. 438). This study suggests that product familiarity yields confidence in a cue whose predictive value is also ascertained through product knowledge and that these factors interact to influence use of the cue.

CONCLUSION

The major contribution of this study is to explain why the correlation between product familiarity and use of the country-of-origin cue is so relatively high and significantly positive. According to our model, the correlation reflects the increased confidence that people with product familiarity have in the information value of the cue. If the model is correct, whatever systematic differences people with high familiarity perceive between countries will in fact be reflected in their actual use of the "made-in" labels in evaluative judgments, because of the higher confidence value of the cue.

REFERENCES

Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411-454.

Bettman, James M. and C. Whan 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, 7 (December), 234-48.

Bettman, James M., Deborah Roedder John and Carol A. Scott (1986), "Covariation Assessment by Consumers," Journal of Consumer Research, 13 (December), 316-26.

Bilkey, Warren J. and Erik Nes (1982), "Country of Origin Effects on Product Evaluations," Journal of International Business Studies, 12 (Spring/Summer), 88-99.

Cox, Donald F. (1962), 'The Measurement of Information Value: A Study in Consumer Decision-Making," in Emerging Concepts in Marketing, ed. W.S. Decker, Chicago: American Marketing Association, 413-421.

Erickson, Gary M., Johny K. Johansson, and Paul Chao (1984), "Image Variables in Multiattribute Product Evaluations: Country-of-Origin Effects," Journal of Consumer Research, 11 (September), 694-99.

Johansson, Johny K., Susan P. Douglas and Lkujiro Nonaka (1985), "Assessing the Impact of Country-of-Origin on Product Evaluations: A New Methodological Perspective," Journal of Marketing Research, 22 (November), 388-396.

Johansson, Johny K. and Israel D. Nebenzahl (1986), "Multinational Expansion: Effect on Brand Evaluations," Journal of International Business Studies, 17 (Fall) 101-26.

Johnson, Eric J. and J. Edward Russo (1984), "Product Familiarity and Learning New Information,'' Journal of Consumer Research, 11 (June), 542-50.

Katz, Elihu and Paul F. Lazarsfeld (1955), Personal Influence (New York: The Free Press).

Katz, Elihu (1957), 'The Two-step Flow of Communication: An Up-to-Date Report on a Hypothesis," Public Opinion Quarterly, 21 (Spring), 61-78.

Kelly, Harold H. (1973), "The Process of Causal Attribution," American Psychologist, 28, 107-128.

Lillis, Charles M. and Chem L. Narayana (1974), "Analysis of 'Made In' Product Images - An Exploratory Study," Journal of International Business Studies, 5 (Spring), 119-127.

Meyer, Robert J. (1981), "A Model of Multiattribute Judgments Under Attribute Uncertainty and Informational Constraint," Journal of Marketing Research, 18 (November), 428-41.

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

Nagashima, Akira (1977), "A Comparative 'Made In' Product Image Survey Among Japanese Businessmen," Journal of Marketing, 41 (July), 95-100.

Olson, Jerry C. and Jacob Jacoby (1972), "Cue Utilization in the Quality Perception Process,' Proceedings, Third Annual Conference of the Association for Consumer Research, ed. M. Venkatesan, Ann Arbor, Michigan: Association for Consumer Research, 167-79.

Park, C. Whan and V. Parker Lessig (1981), "Familiarity and Its Impact on Consumer Decision Biases and Heuristics," Journal of Consumer Research, 8 (September), 223-30.

Reierson, Curtis (1966), "Are Foreign Products Seen as National Stereotypes?" Journal of Retailing (Fall), 33-40.

Schooler, Robert D. (1965), "Product Bias in the Central American Common Market," Journal of Marketing Research, 2 (November), 394-397.

Schooler, Robert D. (1971), "Bias Phenomena Attendant to the Marketing of Foreign Goods in the U.S.," Journal of International Business Studies (Spring), 71 -80.

Sharma, Subhash, Richard M. Durand, and Oded Gur-Arie, "Identification and Analysis of Moderator Variables," Journal of Marketing Research, 18 (August), 291-300.

Wilkinson, J. B., J. Barry Mason, Christie H. Paksoy and Richard M. Durand (1984), "Confidence and Price Knowledge: The case of Advertised Food Specials," in Advances in Consumer Research, 11, ed. Thomas C. Kinnear, Provo, Utah: Association for Consumer Research, 772-5.

Wright, Peter (1975), "Consumer Choice Strategies: Simplifying Vs. Optimizing," Journal of Marketing Research, 12 (February), 60-7.

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