Store Price Perception: Hong Kong Versus North America: Researching Cultural Differences in Consumer Behavior

Citation:

Roger M. Heeler and Kineta Hung (1998) ,"Store Price Perception: Hong Kong Versus North America: Researching Cultural Differences in Consumer Behavior", in AP - Asia Pacific Advances in Consumer Research Volume 3, eds. Kineta Hung and Kent B. Monroe, Provo, UT : Association for Consumer Research, Pages: 162-165.

Asia Pacific Advances in Consumer Research Volume 3, 1998      Pages 162-165

STORE PRICE PERCEPTION: HONG KONG VERSUS NORTH AMERICA: RESEARCHING CULTURAL DIFFERENCES IN CONSUMER BEHAVIOR

Roger M. Heeler, York University, Canada

Kineta Hung, Chinese University of Hong Kong

The consumer behavior of Americans has been extensively researched and documented in a quarter of a century of Journal of Consumer Research and elsewhere. The findings may often have universal validity, but where there are plausible grounds to suspect that effects may be culturally mediated, it behooves researchers to replicate the research elsewhere (Monroe, 1992). Consumer research should "leave home" and interact with the world outside the United States (Wells, 1993). Despite these opinions, Madden, Easley, and Dunn (1995) found that while natural science editors generally endorse replication as a necessary part of research, social science journal editors are less than enthusiastic. The first part of this paper will review some of the reasons for cross-cultural replication.

The first reason for cross-cultural replication of consumer behavior research is to validate the original research, that is to seek similarity. A recent example was Leong, Ang, and Tham (1996). They replicated American brand-name-recall research in Singapore with (English speaking) Chinese race students in Singapore. The effects of pictures, consumer information-processing level, ad meaning, and ad exposure were reproduced in the new cultural setting, so the authors could conclude "our findings lend some degree of generality to those obtained previously in the North American context." They found that better recall was obtained with ads containing pictures and words than with words-only ads, with ads processed semantically than with ads processed sensorially, with ads having a high level of meaning, and with ads that were repeated.

The econd reason for replication is to create a natural experiment. Hofstede (1980) has described culture as an aggregate of characteristics. The researcher identifies a characteristic(s) that has a different level in two (or more) cultures, hypothesizes the effect of this difference on another variable, and measures to see if reality corresponds or not to the hypothesis. This is replication seeking differences.

A recent example of this second approach is Schmitt, Pan, and Tavassoli (1994). The characteristic of interest was language. The hypothesis was that Chinese (Beijing students) and Americans (Chicago students) would differ in recall of brand names depending on whether the presentation was phonic or visual. In particular they hypothesized and found that the Americans relied more on phonologically representation and the Chinese relied more on visual representation.

A third reason for replication is managerial imperative. Marketers must determine their optimal international strategy. For example should marketers standardize the strategy across countries, or should they adapt the strategy by country or region (Levitt, 1983). For example Schmitt, Pan, and Tavassoli (1994) deduced from their results that "marketers should further enhance the natural tendency of Chinese consumers to rely on visual representations and English-speaking consumers’ natural tendency to rely on phonological representations."

CONSUMER’S PERCEPTIONS OF COMPARATIVE STORE PRICES

This paper uses replication for all three reasons. First it is hypothesized that differences in (a) information processing due to language and (b) price value motivation would cause there to be a difference between Canadian and Chinese in their perceptions of store prices, when that processing is performed in a particular fashion (side-by-side). This is reason two (differences). Then it is hypothesized that when processing is performed in a different fashion (one-at-a-time) that Canadian and Chinese store price perceptions will be the same. This is reason one (similarities). Both of these hypotheses have managerial implications (reason three). The research replicated parts of a study by Alba, Broniarczyk, Shimp, and Urbany (1994) to test these hypotheses.

Alba et al hypothesized and found that as a cue to price perception, frequency dominated magnitude. That is, a store advertising small discounts on a large number of items was perceived as having lower prices than another store with the same value in savings represented by large discounts on a small number of items.They believe that this effect is caused by the consumer’s need to rely on visible cues in newspaper advertisements and flyers, and that frequency is an easy cue to encode and remember.

The nine Alba et al experiments differed in detail, as several potential moderator variables were examined. A basic experiment in the sequence is experiment 1B. This experiment included manipulation of another variable, prior information, that is not relevant to the current research. The condition of 1B that omits this extra manipulation, the no-priors condition, is described below.

The subjects were 35 American business undergraduates. All the experimental materials were contained in booklets distributed to each subject. The first experimental page contained an introduction that stated that this was a study concerning supermarket prices. Information regarding 60 grocery store items would be shown for two competing stores, Clarks and Taylors. After reading the introduction, subjects were allowed four minutes to examine the next two pages. These pages contained prices for both Clarks and Taylors for 60 grocery store items. Twenty items were lower priced at Taylors (magnitude condition) by an average of 14 cents. Forty were lower priced at Clarks (frequency condition) by an average of 7 cents. Thus overall, the 60 items cost the same at both stores.

Subjects then turned to three question pages, wih instructions not to look back. The first page asked respondents to estimate the total price for the products shown in each store. The measure used was the difference between the price estimates for frequency and the price estimates for magnitude, that is frequency advantage. The second question asked if a family spent $500 per month at the grocery store, and could buy all the groceries they needed for $500 at Taylors, how much would they spend for the same basket at Clarks. This question also yielded a difference measure. The third page asked, "how did you decide to assign prices the way you did to the two stores".

The results showed Clarks (frequency) perceived $6.32 lower on the first question, and $23.72 lower on the second question. Responses to the third question were dominated by frequency. These results, together with those for the other eight experiments argue persuasively for a managerial focus on frequency. Is this a globally applicable result, or is it limited to the USA?

CHINESE VERSUS NORTH AMERICAN PRICE PERCEPTIONS

In this section two cultural characteristics are suggested which may lead to different store price perceptions between Chinese and North Americans (USA and Canada). The first characteristic is based on language. English native speakers phonologically recode visually presented words, because there is a correspondence between the written word and the spoken word. By contrast, the Chinese language is based on ideograms so there is no correspondence between the written symbol and the sound it represents. Phonological recoding is therefore less likely for native Chinese speakers (Schmitt, Pan, and Tavassoli, 1994).

Native English speakers are therefore habituated to a two step process in encoding and recalling information. Native Chinese speakers by contrast are more likely to be one step processors because of the style they developed in childhood language learning. Presented with a visual array of brand names and prices, Chinese should be more efficient at absorbing and using this information when asked to recall it to render a price perception. The Chinese should therefore be less dependent on mnemonic cues (such as frequency) and more likely recognize the true underlying situation (in the Alba et al experiment, that is no price difference between Clarks and Taylors). Thus it is hypothesized that the domination of magnitude by frequency observed by Alba et al will be attenuated or removed with Chinese native speakers because they will process the visual information more efficiently. This attenuation would be most pronounced while operating in the native language (Chinese), but as a habit of mind, is likely to persist in other scenarios.

The second cultural characteristic has to do with the sociology of Hong Kong versus Canada.

Hong Kong is a city of vibrant commerce where deal making is the norm. The Chinese are keen bargain shoppers (Executive Briefing, 1997). Koreans, a contiguous Asian community, have also been found to have an additional consumer style characteristic compared to US consumers (Hafstrom, Chae, and Chung, 1992). High scorers on this "Price-Value Conscious" dimension believe that they usually pay lower prices and watch for sale items. Even the Japanese, in the past quality fixated, have "become a nation of bargain-hunters and yen-pinchers" (Sakamaki, 1994). So it is expected that Chinese subjects will have the motivation to use their hypothesized advantage in processing visual information to form more accurate store price perceptions.

H1:  Frequency will not dominate magnitude for Chinese native language subjects to the extent that it does English native language subjects.

The research uses the natural laboratory of cross-cultural effects to test this hypothesis. As such the test is subject to confounds with the other characteristics included in the culture bundle. Even though the income levels of Canada and Hong Kong are similar, other potential influences exist such as different retail environments. The hypothesis gives reason to examine for an effect. If an effect is demonstrated, the result is managerially useful but requires further theoretical probing. This caveat always applies when one cultural characteristic is singled out as the basis for research.

REPLICATION WITHOUT CHANGE

Method

The basic Alba et al (1994) experiment described above was replicated in a college town in Ontario Canada and in Hong Kong. The Ontario replication although in a different country to the original, was in a similar cultural zone, so no differences were expected based on cultural factors. The Ontario replication served to test whether the original experiment would or would not replicate in the hands of another experimenter. This issue is not trivial. For example Kellaris and Cox (1989) could not replicate the results of Gorn (1982).

There were minor modifications to procedures employed in the replications. The hypothetical stores were changed from "outside the Southeast" in the Alba et al introduction to "Florida".

Florida was chosen as a plausible holiday destination in the USA, thus allowing the original US grocery items and prices to be used. To minimize the imposition of the experimental process on the students, the original 60 items were reduced to a set of 30 of the original items. These 30 items were split half rotated, that is half the respondents received questionnaires with the brand order switched from the original. The original price range set for answers to the first question was reduced from $100-$130 to $50-$65 (to correspond to the halving of the items). Question two was changed from a base of $500 for a month to $125 for a week (to correspond to a holiday visit). In both Canada and Hong Kong the subjects were undergraduate business students, as in the Alba et al original. Although of Chinese race and mother tongue (Cantonese) the Hong Kong students were proficient in English and used the same untranslated research instruments.

Results

The results (with the results for Alba et al scaled to the revised procedure) are presented as in Alba et al. The figures shown are the mean differences between the "frequency" and the "magnitude" conditions (Table 1).

The Canada replication means are statistically different from zero at .05, using the Z test for observations within respondents.The results show that the original study replicated in Canada, with different experimenters.The Canadian results could therefore be used as a valid base for comparing with Hong Kong.

The Hong Kong study yielded a similar, and statistically significant frequency advantage for the first question, cost of the items shown. For the second measure, the cost at Clarks of a $125 basket bought at Taylors, the frequency advantage was small and non-significant.The reasons given in response to Question three, how did you assign prices, were dominated by frequency, as in the Alba et al studies, for both Canada and Hong Kong.

REPLICATION WITH CHANGE

In the experiments described above, prices for the two stores were shown adjacent to each other for all the grocery items. This display represents those shoppig and pre-shopping situations in which consumers are exposed to prices side by. It does not represent market situations in which prices are seen sequentially. For example, when a shopper is in a store, only the prices of that particular store are seen. The prices of the various brands can be compared intra store, but not inter store. Even when grocery stores advertise in the same edition of a newspaper, the prices of one store are not aligned with those of another. Similarly, flyers print the prices of a particular store, so sequential processing is likely. Price perceptions gained from sequential, on-at-a-time processing may not be the same as those generated by side-by-side evaluation.

One-at-a-time evaluation of information leads to quite different product preferences than two-at-a-time evaluation (Nowlis and Simonson, 1997). In addition to natural differences between sequential and side-by-side processing, the Alba et al experimental procedures may create a demand effect.The subjects may feel obliged to report a difference and therefore seek some reason to ascribe a difference to. This would be similar to the strong effect generated by the procedures of some of the early price-quality research (Rao and Monroe, 1989).

By contrast, in a one-at-a-time evaluation, respondents are not set the task of finding a difference. They assess each store on the set of information provided for it. Neither Canadian nor Hong Kong subjects would benefit from any processing cue such as frequency. The frequency effect is likely to disappear for both. Since this "evens the playing field" between the two cultures, it is also expected that there will no longer be a difference between the two nationalities.

H2: Frequency will not dominate magnitude in one-at-a-time stimulus presentation.

Corollary

H2a:There will be no difference between Canadian and Hong Kong subjects when stimuli are presented one-at-a-time.

TABLE 1

TABLE 2

Method

The method was adjusted to permit one-at-a-time stimulus presentation. The introduction referred to one store only. The price information page contained the same 30 grocery items. For each three pieces of information corresponding to feature pricing were shown, regular price, discount, and Taylors (Clarks) price. In the first question respondents were told the regular price was $54.84 and asked to estimate the Taylors (Clarks) price. In question two subjects were told that they could buy all the groceries they needed for $125 and were asked to price the same basket at Taylors (Clarks).Each respondent saw materials for one store only. This was necessary because if subjects had first gone though the display and questions for one store, they would have been sensitized by the questions in their consideration of the second store.The design correspondingly changed from the original within subjects to between subjects. Because subjects received all materials in one self contained experimental booklet, it was possible to conduct both conditions simultaneously in one class, obviating inter class confounds. The revised procedures were run in the same places, but with new subjects.

Results

The results are shown in Table 2.

The statistical analysis for differences between the means of different samples gave null results.

There was no statistically significant frequency advantage for either measure for either country.

Hypothesis H2 that one-at-a-time price display would eliminate the frequency effect, and its corollary H2a that there would be no difference between the contries, were supported. In response to the third question asking for reasons, magnitude was mentioned about twice as often as frequency.

Although H2 and H2a are supported, they are not proved because they are supported with a null result, that is the absence of proved effect. However the negligible differences obtained, besides being statistically null, are unlikely to be of management significance.

DISCUSSION

Findings

The research used cross-cultural replication to examine store price perceptions under two exposure conditions, side-by-side and sequential. The first cross-cultural hypothesis suggested that under side-by-side exposure to price information, the frequency advantage found in North America would not be found in Hong Kong. The results were mixed. For one measure, the difference in price perception of the items shown to subjects, Chinese results mirrored Canada and Hong Kong so the hypothesis was not supported. For the other measure, price of a basket of goods, the frequency advantage found in Canada and USA was not found in Hong Kong. Further replications would be useful to delimit the effect, for example with uni-lingual Chinese speakers.

The second hypothesis was that the advantage to frequency would disappear with sequential presentation of price information, and by corollary there would be no difference between the cultures with respect to price perceptions. This was supported.

Managerial Implications

The implications of the research for management differ by market context. In markets where consumers are likely to form their store price impressions with information for competing stores presented side by side, frequency has a definite advantage in North America. It is less clear that frequency has an advantage in Hong Kong. An adaptation strategy might be appropriate in which many small discounts are promoted in North America, but not in Hong Kong.

In markets where consumers are likely to form their price impressions sequentially, frequency lacks advantage in both Canada and Hong Kong. A strategy standardization on magnitude might be appropriate because of the secondary advantages of magnitude. Magnitude is likely to be more attention getting (Dickson and Sawyer, 1990), and is also superior in sales generation for the individual brand that is promoted (Gupta and Cooper,1992), and thereby more likely to yield manufacturer support.

REFERENCES

Alba, Joseph W., Susan M. Broniarczyk, Terence A. Shimp, and Joel E. Urbany (1994), "The Influence of Prior Beliefs, Frequency Cues, and Magnitude Cues on Consumers’ Perceptions of Comparative Price data," Journal of Consumer Research, 21 (September), 219-235.

Dickson, Peter and Alan G. Sawyer (1990), "The Price Knowledge and Search of Supermarket Shoppers," Journal of Marketing, 54 (July), 42-53.

Executive Briefing (1997), East Asian Executive Reports, 19 (February 15), 4-5.

Gorn, Gerald J. (1982), "The Effects of Music in Advertising on Choice Behavior: A Classical Conditioning Approach," Journal of Marketing, 46 (Winter), 94-101.

Gupta, Sunil and Lee G. Cooper (1992), "The Discounting of Discounts and Promotion Thresholds," Journal of Consumer Research, 19 (December), 41-411.

Hafstrom, Jeanne L., Jung S. Chae, and Young S. Chung (1992), "Consumer Decision-Making Styles: Comparison Between United States and Korean Young Consumers," The Journal of Consumer Affairs, 26 (Summer), 146-157.

Hofstede, Geert (1980), Culture’s Consequences: International Differences in Work-related Values, Beverley Hills, CA: Sage.

Kellaris, James J. and Anthony D. Cox (1989). "The Effect of Background Music in Advertising: A Reassessment," Journal of Consumer Research, 16 (June), 113-118.

Leong, Siew M., Swee H. Ang, and Lai L. Tham (1996), "Increasing Brand Name Recall in Print Advertising Among Asian Consumers," Journal of Advertising, 25 (Summer), 65-81.

Levitt, Theodore (1983), "The Globalization of Markets," Harvard Business Review, 40 (July-August), 113-122.

Madden, Charles S., Richard W. Easley, and Mark G. Dunn (1995), "How Journal Editors View Replication Research," Journal of Advertising, 24 (Winter), 77-87.

Monroe, Kent B. (1992), "On Replications in Consumer Research: Part 2," Journal of Consumer Research, 19 (September), Editorial.

Nowlis, Stephen M. and Itmar Simonson (1997), "Attribute-Task Compatibility as a Determinant of Consumer Preference Reversals," Journal of Marketing Research, 34 (May), 205-218.

Rao, Akshay R. and Kent B. Munroe (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), 351-357.

Sakamaki, Sachiko (1994), "Speak for Yourself," Far Eastern Economic Review, 157 (May 5), 66-67.

Schmitt, Bernd H., Yigang Pan, and Nader T. Tavassoli (1994), "Language and Consumer Memory: The Impact of Linguistic Differences between Chinese and English," Journal of Consumer Research, 21 (December), 419-431.

Wells, William D. (1993), "Discovery-oriented Consumer Research," Journal of Consumer Research, 19 (March), 489-504.

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