The Influence of Price on Product Perceptions and Product Choice

Kent B. Monroe, Virginia Polytechnic Institute and State University
ABSTRACT - This paper discusses the contributions and limitations of three research papers investigating the influence of price on consumer behavior. Two of the papers investigate the influence of price on product choice dynamically. Although the methodologies of the two papers were vastly different, each research effort is in the direction needed for future behavioral price research. The third paper renews some concerns about previous behavioral price research and corroborates earlier findings that the order of price presentation influences individuals' price perceptions.
[ to cite ]:
Kent B. Monroe (1982) ,"The Influence of Price on Product Perceptions and Product Choice", in NA - Advances in Consumer Research Volume 09, eds. Andrew Mitchell, Ann Abor, MI : Association for Consumer Research, Pages: 206-209.

Advances in Consumer Research Volume 9, 1982      Pages 206-209


Kent B. Monroe, Virginia Polytechnic Institute and State University


This paper discusses the contributions and limitations of three research papers investigating the influence of price on consumer behavior. Two of the papers investigate the influence of price on product choice dynamically. Although the methodologies of the two papers were vastly different, each research effort is in the direction needed for future behavioral price research. The third paper renews some concerns about previous behavioral price research and corroborates earlier findings that the order of price presentation influences individuals' price perceptions.


The research reported by Park, Lessig, and Merrill (1982) is an important step forward in price research. This research recognizes, as suggested earlier by Monroe and Petroshius (1981), that price plays a multi-dimensional role in influencing purchase decisions. Further, this research provides another approach for discovering the dynamics of the purchase decision process. However, as to be expected with a pioneering effort to investigate the dynamics of the choice process, there are some difficulties with the reported research. These problems exist because the authors use ambiguous technical terms, do not provide a substantive review of relevant price research, nor do they provide complete details about the research methodology

Ambiguous Terms

Initially, the authors write of price as a stimulus dimension; later, price is referred to as a product dimension, and, at least once, as a brand attribute. But, what is a dimension? And what is a "category of stimulus dimension"? It would have been simpler to refer to price as a stimulus, and to each separate price stimulus as a cue, which is consistent with the price research literature. Further, the authors write of the categorization process of perception, yet call each individual price stimulus (cue) a category. A little more care in the selection and use of the technical language would have improved the paper.

The authors place emphasis on their categorization of product-related stimuli as continuous or non-continuous. Further, they offer the proposition that people can assign utility to non-continuous stimuli easier and with greater stability than with continuous stimuli such as prices. However, the operational dichotomy of stimuli appears to be whether a stimulus magnitude can be quantified, for example, price or gas mileage ratings. In reality, any price stimulus is presented as a discrete entity, not as a part of a continuum. Hence, the continuous - noncontinuous categorization is artificial and lacks conceptuAl meaning for price research. The issue of stability is important and will be discussed below.

Previous Price Research

The authors report an inability to find descriptions of the way consumers use price in choice situations (purchase?). However, in earlier information processing research, Haines (1974) and Bettman (1979) report on the use of price as a means of initially screening purchase alternatives into acceptable or unacceptable choices.

Moreover, Jacoby and Olson (1977), and Olson (1980) offer propositions on the role of price in the purchase decision process. Further, the notion of acceptable and unacceptable prices has been experimentally verified (Monroe 1971, 1973).

The authors suggest that the process of categorizing prices into ranges produces a high degree of stimulus ambiguity due to the lack of an objective external reference. First, what do they mean by stimulus ambiguity? Second, the earlier conceptualizations and empirical work by Gabor and Granger (1966), Kamen and Toman (1970), Monroe (1971) and Wheatley (19813 consistently establish that buyers to have reference prices that serve as evaluative standards. Further, these reference prices may be external (objective), such as last price paid, or price of the leading brand.

The issue of stability in the price categorization process is raised in this paper, particularly when an individual is making a choice from among acceptable alternatives. However, the stimuli set in the second stage of the choice process is different than in the first alternative screening stage. Hence, there is likely to be a reassessment of the now salient stimuli. Second, Della Bitta and Monroe (1974), and in this session both Rexeisen (1982) and Wheatley et al. (1982) have shown that there are effects due to the order of presenting price stimuli. Theoretically, this phenomenon is explained and predicted by adaptation-level theory (Helson 1964). Thus, it would be more appropriate to recognize the dynamic role of price in the choice process, rather than expecting a single, static role of categorizing purchase alternatives into acceptable or unacceptable groups.

Research Method

A number of questions arise relative to the data collection phase. There is considerable similarity between this research effort and the study reported earlier by Park (1978). Since the stimuli came from the 1976 price book, were the data collected during the 1976-1977 period? If not, then the price stimuli would be unrealistic since they range from $2,911 to $5,526. Another question arises if this study is a direct result of Park's (1978) earlier inquiry. If it is, then what was learned in the previous study that prompted the seemingly research design changes? It would be useful to know this learning process by the researchers to enable other investigators to improve their research designs.

The description of the third phase of data collection, satisfaction ratings, is unclear whether each subject rated all 45 stimuli or only those stimuli from the acceptable alternatives obtained in the second phase. If all 45 stimuli were rated, then it is not clear that the importance the researchers place on these satisfaction ratings in their interpretation of the protocols is correct. For example, the authors attach much significance as to whether subjects used price differences as a choice criterion even though the prices had equal satisfaction ratings. Less significance must be placed on the satisfaction ratings if all 12 prices were rated on the 11-point scales than if a more limited set of acceptable prices, say six, were rated on the satisfaction scale. Further, if 12 prices were rated on an 11-point scale, and only two, three, or four stimuli cues from other categories were rated on the same 11-point scale. then it is possible there was a scale x stimulus interaction confounding the results.

The authors suggest that having subjects make final choice from a set of acceptable alternatives was done for several reasons. Yet, they never tell us what these reasons are. Also, it seems that subjects' verbalizations were recorded for the final choice, but not during the process of categorizing the alternatives as acceptable or not. If so, then valuable information about the subjects' initial decision process has been lost and we to not know what stimuli were used in this initial screening process.

One of the objectives of the fourth stage protocol development was "to identify the impact of price differences as compared to that of differences on other product dimensions." First, what does impact mean in this context? Second, how can one determine stimuli differences when they are described as average, good, very good, or small, medium, and large. Whereas, the differences in the price stimuli and gas mileage were quantified, the differences among each of the other eight stimulus categories were verbal descriptions. Is a $200 difference in price similar to a difference between average and good for frequency of repairs? Indeed, as mentioned earlier, there is probably more of a lack of an external reference for these unquantified stimuli. We are not told the magnitude of the price differences in the results reported, nor are we told how differences between verbal categories were determined.


One of the difficulties in understanding the results of this study is how to interpret what is meant by stimulus differences as suggested above. Also, whether a satisfaction scale can be used to suggest differences in perceived utility is an important issue. However, one key to understanding the results is whether all prices were rated on the satisfaction scale or only the prices of acceptable product choices. A second concern is whether a price difference is perceived to be noticeably different for the purpose of a choice. That is, two prices may be perceived as providing different degrees of satisfaction, yet in the choice process not be perceived to be sufficiently different to warrant a choice solely on the basis of price differences. This issue relates to the concept of jnd or just noticeably different stimuli (Monroe 1979, p. 43).

If the five groups developed by the authors are cross-classified according to whether the prices were rated equally satisfactory or whether the price differences were used in the choice process, the results are as given in the table below. As the table suggests, 27 subjects used price differences as a choice criterion while 23 subjects did not. But, interpreting the lack of use of price differences is impossible without knowing the magnitude of the price differences. For example, as the price differences increased in magnitude, did more subjects use price as a choice criterion? When price was a choice criterion, did the subjects always select the lowest-price alternative? Similarly, how unequal were the satisfaction rating for the 29 subjects? And, as the differences in satisfaction ratings increased, did more subjects use price as a choice criterion?


The role of price in the choice process is dynamic, not un stable in the sense it is unpredictable. The phenomenon isolated in this study has been reported earlier by Haines (1974) and Bettman (1979). That is, price may be used as an early screening device to eliminate certain alternatives. Then, once deliberation has proceeded using other product attributes, price may again become a part of the decision process, particularly if price differences between the final alternatives are perceived as significant to the decision maker. Despite the detailed criticisms of this research, it remains an important step forward in behavioral price research. More research on the role of price in the dynamics of the purchase decision process is needed.



The paper by Rexeisen (1982) attempts to assess whether individuals use price as an indicator of product quality, while overcoming some methodological limitations of previous research. Before commenting on the reported research, we should first address the question "What do we mean by valid price-quality relationship?" Essentially, it is necessary to specify whether the relationship of interest is either

product quality ---------+---------> price (1)


price ---------------+-----------> perceived quality (2)

The first situation postulates that it costs more to produce higher quality (in an objective, technical sense) products, and therefore, products of higher quality are priced higher. The second situation postulates that buyers perceive (believe) that higher-priced products are also higher in quality. While Rexeisen is concerned with the latter proposition, it is important to specify exactly what the domain of the research is. Finally, the issue of validity in the context of this research ought to raise the question of whether the relationship conforms to law, logic, or facts under specified conditions.

Despite methodological limitations of previous price-perceived quality research, it is clear there is a positive relationship as specified above. However, this relationship has not always been found to be statistically significant. Hence, the controversy over whether there is a valid price-perceived quality ought to be centered on determining when the relationship is a determinant of consumer choice, not whether the relationship, per se. exists.

Conceptual Framework

The author sidesteps the issue of whether sufficient theoretical constructs have been identified that we may "formulate a meaningful conceptual framework for price cue research" by proposing "to test, rather than explain" the price-perceived quality relationship. Yet, the author does specify four research hypotheses while disclaiming a conceptual framework. On the contrary, there are several conceptual frameworks proposed and used for price cue research: perceived risk (Olson 1973, 1977), adaptation-level (Monroe 1973, 1977), or assimilation-contrast (Monroe 1971, 1973, 1977). The real research issue, as investigated by Park et al., (1982) is what role does price play in the buyer's choice process? Or, what is the information value of price (or prices) within the choice process?

Perhaps, due to the lack of using a conceptual framework, the hypotheses tested in this research lack clarity and specificity. The first and third hypotheses suggest that quality ratings will be "differentially affected" by price or place of purchase. What is meant by "differentially affected"? Similarly, what does the second hypothesis mean by "moderate the price due effect"? How would this hypothesized moderation take place? The fourth hypothesis suggests that order of price presentation will not influence quality ratings. Yet, as reported earlier by Della Bitta and Monroe (1974) and reiterated by Monroe (1977, 1979), there is a good conceptual argument for expecting significant effects due to order of price presentation. Moreover, Wheatley (1981) applies this conceptual argument in his research on the influence of generic brands on consumer choice.


Even though the research was designed to overcome previously cited methodological weaknesses, many questions remain. Was a manipulation check performed on whether the pretest anticipated price ranges were consistent with the subjects' anticipated price ranges? Indeed, it is possible that the $29 per square yard price might have been outside the subjects' acceptable price range. If so, then this price unacceptability could weaken the observed price-perceived quality relationship.

Three dependent variables were used (quality rating, value for the money, and perceived worth), yet the hypothesis relate only to the quality rating. Consequently, it is questioned as to why the other two dependent variables are in the study. The author used three measuring scales, yet they are similar methods and not useful for testing convergent validity as claimed by the author. Convergent validity is the degree that two or more attempts to measure the same concept through maximally different methods are in agreement (Bagozzi 1980, p. 129). The high inter-method correlation reported in the paper only indicates that the similar methods produced a similar response on the three dependent variables.

Analysis and Results

The correlation of 0.41 between two of the dependent variables suggests non-independence between these two variables. A correlation of this magnitude implies that the two variables might be tapping the same underlying construct. Although the two variables are not identified, it would seem most likely that the two variables are value for the money and perceived worth. Thus, it might have been more appropriate to have used a MANOVA with these two correlated dependent variables.

In the reported ANOVA, there are a number of low F values, substantially below 1.00. For example, F values of .03, .11, .16 likely are significantly less than treatment error. The implication is there was a lack of treatment effect due to the manipulation not working.


It is doubtful that this study has provided an answer to the question of the validity of the price-perceived quality relationship. The main finding that order of presentation had a significant effect corroborates earlier findings (Delta Bitta and Monroe 1974). However, the author's prescription of randomizing price treatments will not remove this effect in future studies. A randomized presentation of prices simply produces a different, perhaps unknown order effect. It would better to recognize that when high prices are presented first, subsequent lower prices are judged more. favorably than if the presentation order is reversed


The paper by Wheatley, Chiu, and Allen (1982) is a well-designed study covering the important issue of product choice. There is a good review of the literature; the conceptualization is sound; and the hypotheses are explained very well. However, as with the preceding papers, some parts of the methodology were not fully explained.


A pretest was conducted to determine that consumers perceive large differences between brands of catsup, but only small differences are perceived between brands of canned tomatoes. Unfortunately, a manipulation check was not used to confirm that these perceptions prevailed among all respondents.

As Rexeisen (1982) noted, there is a need to use multiple indicators for each construct, if we are ever going to be able to check on the reliability of our scales and the validity of our constructs. Unfortunately, this paper follows the tradition of a single scale for the product quality construct. It would be methodologically better to measure perceived quality by utilizing several measurement techniques or scales.

The authors do not indicate what prices were actually used. Instead, we are told the price treatment consisted of a wide price spread and a narrow price spread. Therefore, it is not possible to interpret the significance of the manipulation.


The product treatment for the private brand and the price treatment for the national brand both appear to be significantly less than the treatment error. As noted above, such a result implies the lack of the treatment taking. In particular, it is difficult to understand how the price treatment in Table 6 had a zero sum of squares.

Attempting to understand the resulting choice behavior in this study, the data provided in Wheatley et al., Figures 2 and 3 were converted to the four d graphs below. The numbers represent the proportion of respondents who switched brand choices or remained "loyal" during the experiment. For example, in Figure 1, 48% of those individuals who chose the national brand of canned tomatoes in the first interview, also chose the same brand during the second interview. Similarly, 38, of the individuals switched from the private brand to the generic brand of tomatoes. Figure 1 represents one sample of respondents, and Figure 2 represents the second sample of respondents.

Examining the two figures suggests that each sample behaved similarly whether the choice was canned tomatoes or catsup. As shown in Figure 2, respondents in that sample were more prone to shift their choice from the private to the generic brand. Similarly, the other respondents were more prone to shift from the national to the generic brand (Figure 1). Therefore, was there a confound in this study in that composition of the samples provide a possible explanation of the observed choice behavior?

Alternatively, the results may imply that purchase shifts from private _ generic brands are more apt to occur when (1) there are wide price spreads for brands perceived to be relatively homogeneous, or (2) there are narrow price spreads for brands perceived to be relatively heterogeneous. Obviously, more research using larger samples of products and respondents is necessary to obtain more definitive results. Nevertheless, Wheatley et al's results are intriguing.



Overall, this study is a useful attempt to understand choice behavior. Employing the within subject design permitted an analysis of how individuals might respond over time to a generic brand. Despite the problems inherent in a within subject design, we need to do more re search on an individual basis instead of the heavy reliance on cross-sectional designs.


Bagozzi, Richard P. (1980), Causal models in Marketing, New York: John Wiley and Sons.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, M: Addison-Wesley Publishing Co.

Della Bitta, Albert J., and Monroe, Kent B. (1974), "The Influence of Adaptation Levels on Subjective Price Perceptions," in Scott Wart and Peter Wright (eds.) Advances in Consumer Research Vol. 1, Urbana, IL: Association for Consumer Research, pp. 359-369.

Gabor, Andre, and Granger, Clive W. J. (1966), "Price as An Indicator of Quality: Report on an Enquiry," Economica, 46, pp. 43-70.

Haines, George El. (1974), "Process Models of Consumer Decision Making," in G. Davit Hughes and Michael L. Ray (eds.), Buyer/Consumer Information Processing, Chapel Hill, NC: University of North Carolina Press pp. 89-107.

Helson, Harry (1964), Adaptation-Level Theory, New York: Harper and Row.

Jacoby, Jacob, and Olson, Jerry (1977), "Consumer Response to Price: Attitudinal, Information Processing Perspective," in Yoram Wind and Marshall Greenberg, (eds.) Moving Ahead with Attitude Research, Chicago: American Marketing Association. pp. 73-86.

Kamen, Joseph, and Toman, Robert (1970), "Psychophysics of Prices," Journal of Marketing Research, 7, pp. 27-35.

Monroe, Kent B. (1971), "Measuring Price Thresholds by Psychophysics and Latitudes of Acceptance," Journal of Marketing Research, 8, 460-464.

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

Monroe, Kent B. (1977), "Objective and Subjective Contextual Influences on Price Perception," in Arch G. Woodside, Jagdish N. Sheth, and Peter D. Bennett (eds.), Consumer and Industrial Buying Behavior, New York: Elsevier North Holland, Inc., pp. 287-296.

Monroe, Kent B. (1979), Pricing: Making Profitable Decisions, new York: McGraw-Hill Book Company.

Monroe, Kent B. and Petroshius, Susan M. (1981), "Buyers' Perceptions of Price: An Update of the Evidence," in Harold R. Kassarjian and Thomas S. Robertson (eds.), Perspectives In Consumer Behavior, 3rd ed., pp. 43-55.

Olson, Jerry C. (August 1973), "Cue Properties of Price: Literature Review and Theoretical Considerations," paper presented at the 33rd Annual Convention of the American Psychological Association, Montreal.

Olson, Jerry C. (1977), "Price as an Informational Cue: Effects on Product Evaluations,t' in Arch G. Woodside, Jagdish N. Sheth, and Peter D. Bennett (eds.) Consumer and Industrial Buying Behavior, New York: Elsevier North Holland, Inc., pp. 267-286.

Olson, Jerry C. (1980), "Implications of an Information Processing Approach to Pricing Research," in Charles W. Lamb, Jr., and Patrick M. Dunne (eds.) Theoretical Developments in Marketing, Chicago: American Marketing Association, pp. 13-16.

Park, C. Whan (1978), "A Conflict Resolution Choice Model," Journal of Consumer Research, 5, pp. 124-137.

Park, C. Whan, Lessig, V. Parker, and Merrill, James R. (1982), "The Elusive Role of Price in Brand Choice Behavior," in Andrew A. Mitchell (ed.), Advances in Consumer Research, Vol. 9.

Rexeisen, Richard J. (1982),"Is There a Valid Price Quality Relationship?" in Andrew A. Mitchell (ed.), Advances in Consumer Research, Vol. 9.

Wheatley, John J. (1981), "The Effect of Generic Products on Consumer Perceptions and Brand Choice," in Kent B. Monroe (ed.) Advances in Consumer Research, Vol. 8, pp. 166-169.