Price Effects on Choice and Perceptions Under Varying Conditions of Experience, Information, and Beliefs in Quality Differences

Carl Obermiller, University of Washington
John J. Wheatley, University of Washington
ABSTRACT - Subjects with both strong and weak beliefs about quality differences among brands were receptive to information and the opportunity to sample two products. They changed beliefs, attitudes, and behavior in a consistent manner, but only from the perspective of their own subjective perceptions of the evidence rather than on the basis of the objective evidence presented to them.
[ to cite ]:
Carl Obermiller and John J. Wheatley (1984) ,"Price Effects on Choice and Perceptions Under Varying Conditions of Experience, Information, and Beliefs in Quality Differences", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 453-458.

Advances in Consumer Research Volume 11, 1984      Pages 453-458

PRICE EFFECTS ON CHOICE AND PERCEPTIONS UNDER VARYING CONDITIONS OF EXPERIENCE, INFORMATION, AND BELIEFS IN QUALITY DIFFERENCES

Carl Obermiller, University of Washington

John J. Wheatley, University of Washington

ABSTRACT -

Subjects with both strong and weak beliefs about quality differences among brands were receptive to information and the opportunity to sample two products. They changed beliefs, attitudes, and behavior in a consistent manner, but only from the perspective of their own subjective perceptions of the evidence rather than on the basis of the objective evidence presented to them.

INTRODUCTION

Current economic conditions have resulted in consumers' increased concerns with prices. Sellers have met these concerns with increased emphasis on pricing as a marketing strategy, as evidenced by increased use of mark-downs by retailers, increased market share for off-price retailers, and the emergence of generic brands as a market force. Despite this very visible use of pricing as a competitive tactic, the effects of price on perceptions and choice are still poorly understood

Of specific interest here are the effects of price differences under conditions of choice similarity. Casual observation reveals that many product categories are characterized by little variation in quality among alternatives. Furthermore, price is not always a dependable guide to those differences that do exist Sproles (1977) found that 49% of the products he studied showed either no price-quality relationship or a negative price-quality relationship Most of us would agree that consumers are increasingly concerned with prices and yet many product categories do not show clear, positive price-quality relationships How do consumers use price when a price-quality relationship is ambiguous?

A postulate for many consumer behaviorists is that consumers are rational, utility maximizers; i.e., they choose alternatives that offer the greatest value. Rao (1972) presented value as a perceived rather than an absolute characteristic, which may be modeled as:

value =        perceived quality             per unit

                    perceived price

This model works only if quality and price are broadly defined to include benefits such as image enhancement and risk reduction and costs such as information search and uncertainty Thus, if we postulate that consumers are utility maximizers, a preference among alternative brands reflects either higher perceived quality or lower perceived price for the chosen alternative.

When consumers are unsure of brand similarity, because evidence is conflicting or ambiguous, price may have both direct and indirect effects on value maximization. One possibility is that a range of prices will be interpreted as an indication of a range of qualities When confronted with a range of qualities, consumers may employ a heuristic of buying from the high end of the price range (or the low end or the middle) on the presumption of greater value.

A second possibility is that price may act as a perceptual cue to quality Thus, if other cues to quality are ambiguous, consumers may base their quality judgments on price differences As a result, consumers may perceive the higher-priced brand as higher in quality. Choice would then be made on an assessment of marginal differences in quality and price as in the case of evident quality differences.

It should be noted that the question is not whether consumers behave according to an economic model or a psychological model. The consumer is postulated as a utility maximizer But psychological processes allow price to have indirect economic effects The issue to be investigated is how price affects consumers' assessments of brands' values when quality differences are uncertain. If the uncertainty is addressed with evidence from experience with and information about a product, will consumers choose the lower priced alternative?

Wheatley, Yalch, and Chiu (1980) provided subjects with several sources of information relevant to quality assessment: price differences, objective information from an expert source, personal experience with the product, and a cost constraint The information cited evidence of the similarity of alternatives; likewise, since the "brands" were actually identical, experience should also have provided evidence of similarity Subjects selected six cans of high and tow priced colas in any combination. The findings of the study suggested that cost constraints and information had equal, small effects Combined, the cost constraint and information resulted in average choices of 2.1 cans of the $1.59 cola and 3.9 cans of the S 99 cola. As reported in the study, 50% of the 72 subjects opted for all six cans of the lower-priced cola and received the maximum savings Sensible half The puzzle is the remaining 50%. Given 72 subjects and a mean of 3.9 cans of inexpensive cola, 281 cans of the lower-priced cola were selected. If 36 subjects accounted for 216 of those cans (6 cans each), then the remaining 36 subjects selected only 1.8 cans of the lower-priced brand and 4.2 cans of higher-priced cola This, is spite of what objectively appears to be very strong motivation to do otherwise--the evidence of their senses, a report testifying to the similarity of colas, the possibility of $.60 in cash money, and, one would guess, a notion that choosing the lower-priced "brand" was the expected thing to do.

One explanation for the weak effects of the cost, information, and experience factors in the Wheatley et al study is the resistance of prior beliefs to change As Wheatley reported, cola is a product characterized by "widespread belief in. quality differences among brands." The strength of these beliefs in a price-quality relationship may have been sufficient to resist one taste experience, exposure to information, and the attraction of a t. 60 cash inducement.

In general, consumers prefer more to less for their money, but they are frequently ill-equipped to judge accurately which is more and which is less. Physical attributes are often similar or differences ambiguous, and objective information difficult to obtain. In such cases, consumers may rely on past experiences, observations, and information, which may indicate that prices and quality are related in such a way that selecting the higher-priced alternative would maximize value. The importance of these prior beliefs about price-quality relationships has not been addressed The issue has relevance for researchers in consumer behavior. When consumers have strong prior beliefs about quality differences, are they resistant to information or experience that would lead them to get more for less? Are they, in consequence, inefficient or wasteful? The issue is also relevant to managers: If a firm introduces a brand of similar quality at a lower price, will information (advertising) or experience (sampling) be effective in the face of strong PriOr beliefs?

Numerous studies have shown that price is positively related to perceptions of quality of non-durables (Monroe, 1973; Monroe and Petroshius, 1981; Olson, 1977), particularly when price is the only available information. The use of price as a cue to quality has also been shown to result from individual difference variables: education --Shapiro (1972), experience-- Jacoby (1971), Lambert (1972), and Wheatley and Chiu (1979), and confidence-Bauer (1960), Cox (1962), Della Bitta (1971), and Olson (1972) This study contributes to the price-quality research stream by examining another individual difference variable--prior beliefs about quality differences among brands in a product class

Evidence on the changes in prior beliefs as a result of experience or information is limited and equivocal. One reason for this is that while experience, information, and observation are among the determinants of belief, beliefs may also be based on feelings or even a desire to conform. Some research indicates that beliefs based on past experience are very resistant to change from new information (e.g., Nisbett, Borgida, Crandall, and Reed, 1976; Ross, Lepper, and Hubbard, 1975). Behavior may, however, change more rapidly than beliefs (Berelson and Steiner, 1967) Other research suggests that people use information in making judgments, even when it contradicts their experiences (e.g., Burnkrant and Cousineau, 1975) Robertson (1976) has argued that if preconceived beliefs are weakly held, new information or experiences might easily dislodge them; new information or experiences are less likely to change strongly held beliefs.

Hypotheses

Strong prior beliefs in quality differences among brands in a product class should enhance the likelihood that consumers will employ a price-value heuristic or use price as a cue to quality. Thus, consumers will show a greater preference for higher priced brands in product categories for which they hold strong prior beliefs in quality differences than in product categories for which they hold weak prior beliefs in quality differences. Moreover, strong beliefs in quality differences should be more resistant to change than weak beliefs. As a result, in a product category for which consumers have weak beliefs in quality differences, we should expect evidence of similarity from either taste or information to result in greater shifts from high to low priced brands than in product categories for which consumers have strong beliefs in quality differences Consequently, we hypothesize a main effect of prior beliefs on initial preference and an interaction between prior beliefs and evidence of similarity with respect to change in preference.

Method

The study employed a mixed design, a 2 (product) x 2 (information) x 2 (experience) x 2 (price? design. Product and information were between subjects factors; experience and price were within subjects factors. The two products were cola and popcorn, which were chosen because they were familiar products for the subjects, and they represented products for which subjects held different degrees of prior belief in quality differences. Preliminary measures of a convenience sample indicated that the student subject population had strong, widespread beliefs about quality differences among cola brands, whereas they held weak, ambiguous beliefs about quality differences among popcorn brands. The information treatment consisted of a summary of a study attributed to Consumer Reports. In the "perfect" information condition, the report indicated that brands of cola (popcorn) were practically identical in production and processing techniques and that under controlled conditions, people could not distinguish among them. In the "irrelevant" information condition, the report gave sales figures, usage data, and user demographics for the product class (cola or popcorn) The within-subjects factors were experience--subjects gave preferences for, and made ratings of, the "brands" before and after sampling them -- and price -- subjects sampled both high and low priced "brands."

Subjects were one hundred sixty members of a university community All but seven of these were undergraduate students. Ninety-two were male. One hundred twenty-nine were from eighteen to twenty-five years old, and twenty-three were from twenty-six to thirty-four; the remaining eight were older. Subjects were recruited from a lounge area on campus and asked to participate in a market research taste test. In order to ensure the impact of the differences in prior beliefs between cola and popcorn, subjects were screened by asking them two questions:

Do you believe that, aside from prices, there are differences among brands of cola (popcorn)?

Yes         No              I don't know

Using this scale, indicate how strongly you believe that brands of cola (popcorn) are different

weak belief   1    2     3      4     5   strong belief

In the cola condition potential subjects were included only if they answered "Yes" and checked 4 or 5 for strength of belief. In the popcorn condition, subjects were included if they indicated either "Yes" or "I don't know" and checked either 1, 2. or 3.

Procedure

Potential subjects were first screened for prior beliefs, which determined their product condition Then they were randomly assigned to one of the information treatment conditions. After screening, subjects were seated at a table on which were displayed the two "brands" in identical packages (six-packs of cola, bowls of popped corn along with a clear plastic bag of unpopped corn). The "brands" were identified only by the labels Xt and Wb and the price information, $1.29 or $1.99. (These prices reflected the retail price range for these products at the time of this study. The labels were shown to be evaluatively neutral in a pretest.) The subjects began by reading and filling out a booklet that included measures of product ratings, usage, the information treatments, and "brand" ratings. At the end of the initial measurement session, subjects were informed of the cost constraint--a description of the prize of a six-pack (one pound package) of cola (popcorn) and the additional $.70 cash that went along with the lower priced "brand." Subjects were asked to indicate an initial preference for either the lower or higher priced "brand " Then the subjects were given samples of the two "brands" (2 oz. cups of chilled colas, samples of popped popcorn), after which they completed the booklets, which included more measures of "brand" ratings, demographics, several questions regarding the information treatments and a post-experience measure of preference. After each preference measure, the subjects had an opportunity in an open-ended question, to describe the basis of their decision-making process The respondents were also asked, at the end of the experiment, to indicate what they thought its purpose was. None of the subjects showed evidence of anticipating any of the research hypotheses

In all conditions, subjects had price information, and all preference measures were made under the cost constraint. All subjects expressed an initial preference, sampled both "brands" and expressed a final preference. Except for the information differences and product differences, all subjects went through the same procedure.

RESULTS

Randomization and Manipulation Checks

Subjects were randomly assigned to information conditions within each category. A comparison of product usage and demographics indicated no statistically significant differences between information groups for either product. Nor did the information groups within products differ significantly in their strength of beliefs about product differences.

Three questionnaire items referred to the information treatment. Two of these were comprehension checks; the third was a measure of agreement Only the perfect information conditions received relevant information, and they were measured for comprehension of two points: the equivalence of ingredients or process and the inability of people to distinguish among brands. For cola the percentages correct on these items were 84% and 79%; for popcorn they were 92% and 82%. The percentages of correct answers suggest that subjects read the information attentively and understood it adequately.

Agreement or disagreement with the summary of the Consumer Reports study reflected the strength of prior beliefs Subjects in the cola condition were in less agreement with the perfect information (means on the rating, with higher numbers indicating more agreement were 3.47 for cola and 3.97 for popcorn; t=2.01, p<.03) There was no difference in agreement between the product conditions when the information was irrelevant (means were 4 68 for cola and 4.87 for popcorn; t= 97, p<.25)

Choice

Behavioroid measures were taken before and after tasting the products--subjects indicated which "brand" they wanted to receive as a prize for participating in the study. Both choices were constrained by the $.70 that accompanied the lower priced "brand." The preference data are illustrated in Table 1; in order to show the combined effects of both types of evidence of similarity (taste experience and information), the data contrast subjects before tasting who had received irrelevant information with subjects after tasting who had received perfect information. (The independent effects of taste experience and information were small (not statistically significant) reductions in preference for the higher priced "brand" of each product.)

TABLE 1

PREFERRED "BRAND" BY PRODUCT, PRICE, AND INFORMATION

Table 1 illustrates the expected differential effects of price on preference: before tasting and with irrelevant information, subjects preferred the higher priced cola (60% vs. 40%) but the lower priced popcorn (63% vs 37%). The size of the effect was tested with a t test of proportions, comparing the proportions of subjects in the two product categories who preferred the higher priced "brand" (t=2. 13, p=.04). Since the subjects were screened, the finding may not, of course, generalize to all users of these products In addition, there may be other differences between products besides strength of prior beliefs. These other differences may also account for the different effects of price, e.g., frequency of purchase. Olson (1977) among others has discussed the problems of using products as representatives of points on a theoretical dimension in price-quality research. We return to this point below

The data in Table 1 show that the combination of experience and information resulted in a marked shift from higher to lower price "brands" for both products (cola: t=2.56, p= 01; popcorn: t=2.72, p=.01). It was hypothesized that the strong beliefs about quality differences that subjects held for cola "brands" would be more resistant to change than those pertaining to the popcorn "brands." The changes in preference for the higher priced "brand" were superficially identical for the two products: from 60% to 33%, a decrease of 27 percentage points for cola; and from 375 to 10%, a decrease of 27 percentage points for popcorn. However, this 27% change represents very different proportionate changes in preference for the two products that are masked by a "floor effect " (The floor effect is created by the small proportion of subjects in the popcorn condition who initially preferred the higher priced "brand.") Thus, for cola, the nominal 27% decrease represents a real decrease in preference of 45% (27%/ 60%); whereas, for popcorn, the nominal 27% decrease reflects a real decrease in preference of 73% (27%/37%) Thus, as hypothesized, a greater proportion of those who preferred the higher priced popcorn switched preference on the basis of experience and information. Moreover, given the evidence, nearly all (90%) popcorn subjects preferred the lower priced "brand"; a full one-third of the cola subjects still preferred the higher priced "brand " We now turn to the task of explaining why so many of the subjects chose the hither priced alternative.

Evaluations

Subjects evaluated the brands on ratings of overall quality before and after sampling. The mean values of these ratings are presented in Table 2.

TABLE 2

EVALUATION BY PRODUCT, PRICE, EXPERIENCE AND INFORMATION

The data in Table 2 were analyzed as a 2 (product) x 2 (information) x 2 (experience) repeated measures design using analysis of variance. The quality ratings were analyzed separately for each of the high and low price levels. For the higher priced "brand," an interaction between product and experience was discovered (F=15.3, p<.001). The overage quality rating of higher priced cola (averaged across information conditions) was 4.84 before the taste experience and dropped to 4.29 after tasting. For popcorn, on the other hand, the ratings improved from 4 44 to 4 80. The decrease in quality rating for cola is explainable as a change due to experience; but the improvement in quality rating for popcorn was something of a surprise It may reflect lower familiarity with the product; upon tasting, subjects discovered they liked popcorn more than they remembered

Changes in the ratings of the lower priced "brand" were larger and consistent with expectations. There was an information X experience interaction (F= 3.60, p=.06). Ratings of quality for the lower priced "brand" increased with experience in both information conditions, but the increase was smaller when augmented by perfect information. Thus, the mean quality rating (averaged across products) improved from 3 77 to 4.48 when information was irrelevant; and from 4 25 to 4.53 when information was perfect Those who received perfect information had a higher initial evaluation of the lower priced "brand," but, after experience had about the same evaluation as those with irrelevant information had after experience. The limit on improvement may be a ceiling effect; evaluations can only improve so much.

As was true of preferences, the effects of evidence on product evaluation can be seen more clearly by combining the effects of experience and information. In Table 3 evaluation measures are contrasted for subjects who were ignorant (before experience, irrelevant information) versus subjects who were informed (after experience, perfect information). For the former, price was the only new information available; for the latter, price was tempered by both information and experience.

TABLE 3

EVALUATION BY PRODUCT, PRICE, AND EVIDENCE

There are two interesting findings in Table 3. First, a higher price for cola had the effect of statistically significantly increasing quality rating for ignorant subjects (4.85 vs. 3.38; t=6. 70, p< 001) but not for informed subjects (4 40 vs 4.23; :=.76, p=.20). A higher price had no statistically significant effect for popcorn. Thus, as expected, when no evidence was available, a high price enhanced perceptions of quality for cola but not for popcorn However, when accompanied by experience and information, the price difference had no appreciable effect on quality evaluation for either product. Second, the state of being informed had different effects on the ratings of high versus low priced "brands." The evaluation of the higher priced brand changed inconsistently (cola: -.45; popcorn: +.35); whereas, the evaluation of the lower priced brand was consistently improved (cola: +.85; popcorn: +.64) Moreover, changes in evaluation of the lower priced "brands" were larger than the changes in evaluation of the higher priced "brands " This asymmetry of change suggests that the high priced "brands" were the standards of comparison. The finding that evidence of equality raised ratings of the suspect "brand" rather than reducing both to the lower common denominator may be the result of more prior experience with higher priced brands. Or it may be an example of the Polyanna Principle (Matlin and Stang, 1978)--a generalized tendency to think well of things. A final possibility is that subjects' evaluations were veridical--the "brands" were, in fact, high priced brands

The evaluation data are slightly inconsistent with the choice data. Although evaluations of the two price levels of cola were moved closer together by the weight of experience and information, evaluations of the higher priced cola were still more positive than evaluations of the lower priced cola This, even though preference for the higher priced cola shifted from 24/40 (for the before taste, irrelevant information subjects) to 14/43 (for the after taste, perfect information subjects). In similar fashion, evaluations of the two prices of popcorn were changed slightly by the evidence even though only 4 out of 39 subjects preferred the higher priced "brand." However, two points should be made regarding these attitude-behavior inconsistencies. First, the inconsistencies are explainable in terms of differences in objects of measurement. The attitude measures are of the "brands"; the preference measures are of a cost-constrained choice. Thus, the attitude measures ignore the utility of the $.70 that goes with the lower priced "brand " Second, although cola subjects indicated more positive evaluations of the higher priced "brand" even after evidence, this finding should not be taken as evidence for a greater resistance to change Attitude change was at least as great for cola, in spite of strong prior beliefs in quality differences Thus, the aggregate attitude data apparently do not support the hypothesized interaction effect of greater resistance to evidence when products are characterized by strong beliefs in product quality differences.

Analysis of Switching Behavior: Beliefs and Evaluations

This study was premised on the notion that those subjects who opted for the more expensive cola in the Wheatley (1980) study (and in general) did not appear to be acting sensibly. The results of this study are somewhat encouraging as they suggest that subjects did adjust their prior beliefs on the basis of new evidence Moreover, even strong beliefs can be changed--an encouraging sign for anyone that hopes for increasing efficiency in the market system. Yet, despite the evidence that subjects did use information and experience to reevaluate "brands," a full 32% of the cola subjects still opted foe the higher priced cola. Moreover, the evaluation of the higher priced cola remained higher than the lower priced one even after information and experience In an attempt to understand the persistence of price perception effects, subjects were categorized according to their pattern of preference Four groups were formed: LOW-LOW, those who preferred the lower priced "brand" on both choice measures; HIGH-HIGH, subjects who selected the higher priced "brand" on both occasions; LOW-HIGH, subjects who switched upward in price after tasting; and HIGH-LOW, subjects who switched downward. Table 4 displays the means of a number of belief and evaluation ratings for each of the four switching patterns.

TABLE 4

BELIEFS AND EVALUATIONS BY SWITCHING PATTERN

The data in Table 4 were analyzed as a 2 (product) x 2 (information) x 4 (switching pattern) design using analysis of variance. None of the dependent measures indicated a statistically significant interaction between switching pattern and either of the other two variables; therefore, for ease of illustration the data are displayed without partitioning by product or information conditions.

The data describe subjects as clearly sensible in their behavior, as seen through their own eyes LOW-LOW's most believed all brands taste the same; HIGH-HIGH's Least believed it. Conversely, subjects who opted for the lower priced "brand" (LOW-LOW and HIGH-LOW) least believed in the price-quality relationship; whereas, those who opted for a higher priced "brand" (HIGH-HIGH and LOW-HIGH) most believed it. Subjects acted consistently with their beliefs about quality differences

Even more interesting is the pattern of evaluation ratings LOW-LOW's expected little difference in brands before tasting and, accordingly, selected the lower priced "brand.' After tasting, their evaluations of the lower priced "brands" improved relative to the higher price alternatives (quality difference =-.62). The HIGH-HIGH's held the opposite evaluations--they perceived the higher priced "brand" as higher in quality before tasting, which explains their initial preferences; and, after tasting, they retained those differences in evaluations (quality difference=1.20).

An objective observer might question the rationality of the behavior of the LOW-HIGH's. They did not have strong initial beliefs about quality differences (quality difference=.36); yet, they switched to the higher priced "brand." Given their perceptions, they switched for good reason. After tasting, these subjects perceived the greatest differences in quality (quality difference=1.58)

These 14 subjects are an intriguing group All of them tasted identical products. Seven of them also received perfect information. The reasons for their switching are not apparent, but it is interesting to speculate about possible latent price-quality bias that requires some ambiguous experiential evidence before it can emerge. Future research might attempt to identify individual difference variables that account for and characterize these subjects who perceive price-quality relationships only after ambiguous (in fact, objectively contradictory) experiential evidence.

The final group of switchers, HIGH-LOW's initially preferred the higher priced "brands" but appears to have yielded to the evidence. Initial evaluations showed large differences in favor of the higher priced brands (quality difference=1.30) After experience, evaluations favored the lower priced "brands" (quality difference=-.14). (Objective rationalists and marketers may be encouraged that HIGH-LOW's outnumbered LOW-HIGH's by 27 to 14.)

Conclusions and Discussion

The research was begun with two hypotheses. The first was that subjects with strong prior beliefs in quality differences would have greater preference for a higher priced "brand" than subjects with weak prior beliefs in quality differences. This hypothesis was supported by measures of choice and evaluation. The second hypothesis was that the effects of prior belief on the price-preference relationship would interact with evidence of similarity of brands, with stronger beliefs more resistant to change. The second hypothesis was partially supported. The inclusion of a taste experience and information resulted in fewer choices of the higher priced "brand" of both products, but the proportionate change for cola was less than for popcorn Evaluation measures, however, indicated that evidence resulted in more change for cola "brands" than for popcorn "brands " A strong "floor effect" may have prevented larger effects on both choice and evaluation measures for popcorn. Since subjects initially liked the lower priced popcorn "brand," they had little room for improvement

Post hoc analysis contrasted subjects on the basis of preference switching patterns. This analysis indicated clear consistency among preferences, beliefs about quality differences, and evaluations of the "brands "

The results of the study are encouraging to all who look for enough consistency and sensibility in consumers to offer hope for ultimate understanding. First, on average, subjects were receptive to evidence and changed beliefs, attitudes, and behavior consistent with their experience and information, even those with strong prior beliefs. Second, although objectively some subJects opted for less when they could have had more, when viewed from the perspective of subjective perceptions, all subjects appear to have opted for more. Some individuals' perceptions of quality were sufficiently enhanced to offset the marginal utility of price differences despite evidence of information and experience Future research should investigate individual difference explanations of these complex effects of price on perception

Limitations

The principle limitation of this study was the confounding of prior belief strength and product condition. The combination was used in order to strengthen the control over the prior belief strength construct, but a consequence is an uncertainty about the cause of the difference between product groups. Similar uncertainty occurs whenever researchers attempt to operationalize a theoretical construct by selecting mundane examples that are intended to represent different levels of the construct Any pair of products will have numerous differences in addition to levels of the construct of interest. Since prior belief strength, like many product-related variables, is difficult to manipulate, researchers may have to continue to rely on operationalization by example. Confidence in relating effects to the construct of interest can be increased by operationalizing the construct in more than one way, i.e., using more products in order to reduce the likelihood of rival hypotheses that can account for the effects Future research might use other products that are judged to differ in terms of prior beliefs. Corroborative evidence could also result from internal analyses, where measures of belief strength are used to relate subjects' prior beliefs and price-preference relationships for single products.

The study also has the limited generalizability that characterizes much of the work that has been done on price effects. Cost constraints, information, and experience represented additional complexities to the bare-bones price-quality paradigm, but other important complexities were ignored: forced choice, student subjects, non-durable products, and lack of any information other than price (such os brand or store image). The roles of price in both perceptual processing and economic assessment are complex, and it is unlikely that a study can simultaneously isolate interrelationships among constructs and demonstrate effects in mundane settings Additional research is needed and being done with an aim to adding to our understanding of the role of psychological processes in economic decisions and their significances in "real-world" activities.

REFERENCES

Bauer, Raymond (June, 1960), "Consumer Behavior as Risk Tasking," Dynamic Marketing for a Changing World, American Marketing Association, 389-398.

Burnkrant, R E. and Cousineau, A (1975), "Informational and Normative Social Influence in Buyer Behavior," Journal of Consumer Research, 2, 206-215

Cox, Donald (December, 1962), "The Measurement of Information Value: A Study in Consumer Decision-Making," Emerging Concepts in Marketing, American Marketing Association, 413-421.

Della Bitta, Albert (1971), "An Experimental Examination of Conditions Which May Foster the Use of Price as an Indicator of Relative Product Attractiveness." unpublished doctoral dissertation, University of Massachusetts.

Jacoby, Jacob, Olson, J. C. and Haddock, R. A. (1971), "Price, Brand Name, and Product Composition Characteristics as Determinants of Perceived Quality," Journal of Applied Psychology, 55, 6, 570-579.

Lambert, Zarrell V. (February, 1972), "Price and Choice Behavior," Journal of Marketing Research, 9, 35-40.

Matlin, Margaret and Stang, David (1978), The Polyanna Principle, Cambridge, MA: Schenken Publishing Co.

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

Monroe, Kent B. and Petroshius, S. M. (1980), "Buyer-s Perceptions of Price: An Update of the Evidence," in Perspectives in Consumer Behavior, Harold H. Kassarjian and Thomas S. Robertson, eds., Glenview, Ill.: Scott Foresman and Co., 43-55.

Nisbett, R. E., Borgida, E., Crandall, R., and Reed, H. (1976), "Popular Induction: Information Is Not Always Informative," in Cognition and Social Behavior, edited by J. Carroll and J. Payne. Hillsdale. NJ: Lawrence Erlbaum.

Olson, J. C. (1977), "Price as an Informational Cue: Effects on Product Evaluations," in Consumer and Industrial Buying Behavior, A. G. Woodside, J. N. Sheth, and P. D. Bennett, eds., New York: North Holland, 267-286.

Olson, J. C. and Jacoby, J. (1972), "Cue Utilization in the Quality Perception Process," Proceedings of the Third Annual Conference of the Association for Consumer Research, 167-179.

Rao, Vithala (1972), "A Model for Brand Choice Under Price Quality Hypothesis," American Marketing Association Proceedings, 34, 366-371.

Robertson, Thomas S. (April, 1976), "Low Commitment Consumer Behavior," Journal of Advertising Research, 16, 19-24.

Ross, L., Lepper, M., and Hubbard, M. (1975), "Perseverance, Self-Perception, and Social Perception: Biases and Attributional Processes in the Debriefing Paradigm," Journal of Personality and Social Psychology, 32, 245-254.

Shapiro, Benson P. (August, 1973), "Price Reliance: Existence and Sources," Journal of Marketing Research, 10, 289-294.

Sproles, George (1977), "New Evidence on Price and Product Quality," Journal of Consumer Affairs, 11, 63-77.

Wheatley, John and Chiu, J. (August, 1979), "The Influence of Intrinsic and Extrinsic Cues on Product Quality Evaluations of Experts and Non Experts," Proceedings of the American Marketing Association, Minneapolis, 205-209.

Wheatley, John, Yalch, R., and Chiu, J. (1980), "In Search of the Economists' Consumer: The Effects of Product Information, Money, and Prices on Choice Behavior," Advances in Consumer Research, VIII, 583-587.

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