Consumer Perceptions of Multi-Dimensional Prices

Hooman Estelami, Fordham University
ABSTRACT - Multi-dimensional prices are prices which consist of multiple components-such as prices quoted in terms of the combination of monthly payments and number of payments-rather than a single lump-sum dollar amount. This paper investigates how consumers form their price perceptions of multi-dimensional prices. Using an information integration approach, results from a laboratory study indicate that under conditions typical of the market place, consumers do not evaluate multi-dimensional prices rationally. Instead they utilize a simplified model, resulting in inaccuracies in their price perceptions. Increasing the dimensionality of price is also found to result in price perception inaccuracies. Moreover, consumers are found to place a larger weight on the monthly payment amount than on the number of payments. These results have implications from both a marketing, and a policy-making point of view.
[ to cite ]:
Hooman Estelami (1997) ,"Consumer Perceptions of Multi-Dimensional Prices", in NA - Advances in Consumer Research Volume 24, eds. Merrie Brucks and Deborah J. MacInnis, Provo, UT : Association for Consumer Research, Pages: 392-399.

Advances in Consumer Research Volume 24, 1997      Pages 392-399


Hooman Estelami, Fordham University

[The author would like to thank Prof. Donald R. Lehmann for his helpful comments on earlier drafts of this paper.]


Multi-dimensional prices are prices which consist of multiple components-such as prices quoted in terms of the combination of monthly payments and number of payments-rather than a single lump-sum dollar amount. This paper investigates how consumers form their price perceptions of multi-dimensional prices. Using an information integration approach, results from a laboratory study indicate that under conditions typical of the market place, consumers do not evaluate multi-dimensional prices rationally. Instead they utilize a simplified model, resulting in inaccuracies in their price perceptions. Increasing the dimensionality of price is also found to result in price perception inaccuracies. Moreover, consumers are found to place a larger weight on the monthly payment amount than on the number of payments. These results have implications from both a marketing, and a policy-making point of view.

"1995 Acura Integra ... $189 a month ... 36 months ... No down payment!"

(Advertisement in the November 11, 1994 edition of the New York Times).

Offers like the one mentioned above are hard to avoid now a days. The drive to finance the ownership of products rather than to pay a single lump-sum payment has caused the media to literally become flooded with complex pricing schemes for products ranging from luxury automobiles to sneakers. For example, the automobile leasing business alone is expected to double in the next five years, and the purchase of 3 out of every 4 new cars sold today is financed (Business Week 1994). What is unique about prices such as the one mentioned above is their multi-dimensional nature-the fact that the price no longer consists of a single dollar amount (e.g., $15,395), but rather consists of multiple dimensions (e.g., monthly payment of $189, 36 monthly payments, and $0 down). As such, we will refer to these kinds of prices as "multi-dimensional prices" (MDP) in the balance of this paper.

One of the fundamental questions brought about by the wide presence of multi-dimensional prices is regarding the consumers’ ability to appropriately evaluate them. Consumers are often bombarded with a large number of multi-dimensional prices through newspaper, TV, and radio ads, such as the one shown in Figure 1. For example, each Sunday issue of the New York Times (for the first 12 weeks of 1995) on the average featured over 65 MDP ads. Moreover, multi-dimensional prices communicated through TV and radio ads are typically included in a 15 to 30 second presentation, in conjunction with a large amount of non-price information, and for many product categories-such as household appliances, new automobiles, and telecommunication services-the dominant form of price quotation is MDP. In the case of new automobiles for example, over 70% of the advertised prices are in an MDP format.

Interestingly, the existing research on consumer price perceptions has had limited coverage of MDPs. Little is known about how consumers integrate the various components of an MDP. Meanwhile, MDPs may offer the firm the ability to increase consumer demand without dropping the effective price-for example by reducing the magnitude of one price dimension (e.g., monthly payments) and compensating for it in another dimension (e.g., number of months). MDPs are also relevant from a policy-making perspective, as they make inaccurate consumer price perceptions a very likely possibility. In this paper, we investigate how consumer price perceptions of multi-dimensional prices are formed. In specific, using an information integration approach, the paper studies the integration model used by consumers in consolidating the various dimensions of a multi-dimensional price. Findings from a laboratory study indicate that under conditions typical of the market place, consumers do not evaluate MDPs rationally. Instead they use a simplified integration model, resulting in inaccuracies in their price perceptions.


Pricing is perhaps the oldest research area in marketing. Interestingly, a review of three classic pricing books (Nagle and Holden 1995, Monroe 1990, DeVinney 1991), and Rao’s (1993) pricing review indicates no specific references relevant to the topic of multi-dimensional pricing. However, what seems to exist in the current literature is evidence on consumer difficulty in evaluating complex prices. This evidence can be found in the works of Greenleaf, Morwitz and Johnson (1994), Gourville (1994) and the unit pricing literature, to be discussed below. Greenleaf, Morwitz and Johnson (1994) have proposed that breaking up the price of a product (for example by charging the tip as a separate item rather than incorporating it into the menu prices of a restaurant) reduces the consumer’s price sensitivity. Since the objective value of the price is not affected by its division into different parts, from the perspective of classical economic theory, consumers should be indifferent to divided prices. Using a series of simulted shopping experiments, the authors demonstrated that divided prices are perceived as significantly less expensive, and produce stronger purchase intentions. Gourville (1994) has also found evidence for consumers’ inability in dealing with complex prices. Gourville’s work has primarily concentrated on the concept of pennies-a-day (PAD) pricing, whereby prices are framed in terms of the much smaller daily amounts rather than a large lump-sum dollar amount (e.g., "Support a starving third world child for 50 cents a day" vs. "$183 a year"). Based on the concept of categorization (Mervis and Rosch 1981) and mental accounts (Thaler 1985), Gourville argues and demonstrates that in spite of objectively large dollar values associated with some transactions (e.g, $183 per year), PAD framing (e.g., 50 cents a day) can significantly reduce consumers’ price sensitivity.

Both the works cited above seem to suggest that consumers have difficulty in evaluating prices when the presentation of the price is made more complex. Additional evidence on consumers’ difficulty in coping with price complexity can be found in the unit pricing literature. For example, Russo (1977) studied consumers’ utilization of unit price information in an actual field setting. He found that unless unit price information is made explicitly available, consumers are unlikely to estimate and utilize unit prices in their purchase decisions, thereby making sub-optimal purchases. Russo therefore concluded that information processing factors can drastically impact consumers’ ability to process price information. Capon and Kuhn (1982) have also shown that when a product is offered in various package sizes at different price levels, consumers are highly inaccurate in identifying the "best buy". Moreover, their study showed that this inaccuracy is not dependent on the level of education and training of the consumer. Other evidence relating to unit prices can be found in the widely practiced retail strategy of quantity surcharges (Nason and Della Bitta 1983, Widrick 1979). Quantity surcharges occur when a larger package of the same brand is more expensively priced (on a per unit basis) than a smaller package (Agrawal, Grimm and Srinivasan 1993). The frequent occurrence of quantity surcharges across many product categories (Cude and Walker 1984) seems to suggest that consumers are unable to conduct the mental arithmetic required to recognize that they may be paying a higher per-unit price for the larger size packages.



Parallel evidence to the above can be found in the area of bundling, which is primarily concerned with how consumers evaluate a product which itself is made of multiple sub-products. For example, Yadav (1994) has shown that in assessing the quality of a bundle, consumers utilize a process similar to anchoring and adjustment (Tversky and Kahneman 1974). They assess the quality of one component first (anchoring) and adjust their initial assessment based on subsequent assessments of the remaining components (adjustment). As a result of using the anchoring and adjustment heuristic, an additive integration rule best represents how consumers integrate the values of the individual components of a bundle (Gaeth et al. 1990). Therefore, as will be discussed below, similar heuristics and simplification strategies may be used by consumers when encountering multi-dimensional prices.

While the works cited above indicate that consumers have difficulty in dealing with complexity in prices, none of the existing research in pricing has investigated how consumers integrate the individual dimensions of price to form their price perceptions-the primary focus of this paper.


In this section of the paper, the characteristics of the integration model used by consumers to evaluate multi-dimensional prices is studied. From a rational point of view, certain prie dimensions in an MDP need to be configurally (interactively) integrated into the consumers’ price perceptions. For example, given a series of multi-dimensional prices made of monthly payments, number of payments and a down payment, there exists a rational pattern of price perceptions corresponding to the various MDPs. This pattern would have to be based on some mathematical principle which would arrive at the net offered price. In the case of the MDPs mentioned, the rational model, assuming no discounting, would be determined by multiplying the number of payments by the monthly payments and incrementing that by the down payment. What therefore characterizes the rational evaluation models for multi-dimensional prices is that certain price dimensions (e.g., monthly payments and number of payments) need to be configurally integrated into the consumer price perceptions. In other words, combined changes in each price dimension must create large shifts in consumer price perceptions. This configural evaluation of the price dimensions needs to take place either through direct numerical processing of the required arithmetic, or by an interactive utilization of the relevant price dimensions. In what follows, we examine literature which would provide insight on the likelihood of either of these two events taking place. It is argued that in the environments in which multi-dimensional prices are communicated to the consumer (e.g., through newspaper ads such as those shown in Figure 1, or TV and radio ads), the individual dimensions of an MDP are likely to be independently integrated into consumer price perceptions. Moreover, hypotheses on the relative weights of the price dimensions, and the effects of increasing price dimensionality on price perception accuracy are put forward.

Evidence from Cue Utilization Research

In order for a consumer to evaluate a multi-dimensional price, he/she must use the individual dimensions of the MDP as information cues in his/her judgements. Related works on human utilization of information cues seem to suggest that the most commonly used integration rule used to form human judgements is linear (Anderson 1981). Summarizing the results of the early works, Slovic (1972) concluded that when utilizing information, judges tend to use "only the information that is explicitly displayed in the stimulus object ... information that has to be stored in memory, inferred from the explicit display, or transformed tends to be discounted or ignored" (p. 14). Across a large variety of tasks, these studies indicate that the most prevalent form of cue integration by judges is linear. Complex curvilinear or configural (interactive) cue utilization strategies, although at times observed, are typically not undertaken by the majority of judges. Further work by Brehmer et al. (1980; experiments 4 and 5) has shown that linear integration in human judgements may persist even under conditions where the task instructions explicitly require the judges to configurally integrate the cues. As a result of using a linear integration model, information cues which from a normative point of view need to be configurally treated, are often independently integrated into human judgements. The primary reason for linear cue utilization is the limitations on human information processing abilities which force judges to use a simplifying heuristic when evaluating a multi-attribute alternative. Process tracing studies of human judgements have shown that when judging a multi-attribute alternative, judges tend to use a process much like the anchoring and adjustment heuristic of Tversky and Kahneman (1974; e.g., Lopes 1982, Russo and Dosher 1983). In this process, judges initially focus their attention on one particular attribute, and form an initial overall judgement based on the value of that attribute. They then proceed to examine the remaining attributes and adjust their overall judgement accordingly. The serial nature of the process therefore creates primacy in the judgements while also limiting the degree by which configural integration of attribute information can take place. As a result, there is a tendencyfor the individual attributes to be independently integrated into the overall judgement.

Birnbaum’s Subtractive Theory

Additional evidence for independent integration of stimuli dimensions can be found in Birnbaum’s (1978) Subtractive Theory of stimuli integration. For many continua, when people are asked to judge "ratios" and "differences" of stimuli, they appear to use only a subtractive operation regardless of the instructions. This phenomenon has been shown to exist for stimuli such as the loudness and pitch of tones (Elmasian and Birnbaum 1984), heaviness of weights (Mellers, Davis, and Birnbaum 1984), and darkness of dot patterns. Birnbaum’s works suggests that when instructed to evaluate either the ratio or the difference between stimuli, in both cases subjects evaluate only the algebraic difference (i.e., subtraction) between subjective stimuli values. Therefore, their responses reflect the independent integration of the stimuli scale values into their judgements. The implication of Subtractive Theory on multi-dimensional price perceptions is that in cases where individual price dimensions need to be configurally treated, they may in fact be independently integrated into consumer price perceptions.

Evidence from Studies on Mental Arithmetic

If the consumer decides not to process the individual price dimensions of an MDP as qualitative judgement cues, he/she may instead proceed to numerically process the necessary arithmetics. Studies of human mental arithmetic processes have established that conducting mental arithmetic requires the retrieval of both declarative knowledge (e.g., multiplication tables) and procedural knowledge (e.g., multiplication algorithms) from long term memory (Groen and Parkman 1972). In conducting mental arithmetic, humans break down the original problem into a series of sub-problems, which are then serially solved. The execution of these sub-problems involves a series of steps which requires the temporary storage of the intermediate results, and as a consequence, the capacity for conducting multi-digit mental arithmetic problems required to evaluate most MDPs tends to be severely limited by the capacity to process and store information in short-term memory. Studies focusing specifically on the timing and effort required to conduct mental arithmetic have for example shown that conducting the mental arithmetic needed to evaluate a multi-dimensional price (e.g., a 3-digit by 2-digit multiplication) is both time consuming and demanding, at times taking subjects well over 120 seconds to complete (Dansereau and Gregg 1966, Hitch 1978). The primary reason for the slow processing speed is the large number of arithmetic operations (e.g., hold, carry, add, etc.) associated with multi-digit arithmetics, which need to take place mentally, using the limited capacity of short-term memory. The timing results of these studies indicate that the length of time required to conduct the mental arithmetic associated with most MDPs may in fact far exceed the limited time period available to consumers (e.g., 15 to 30 seconds) when faced with the large number of MDPs they are bombarded with in the media on a daily basis. Moreover, studies focusing on the physiological effects of mental arithmetic processes have also shown that conducting mental arithmetic is an effective producer of physiological stress (e.g., Linden 1991, Caroll et al. 1986). In fact, in studies where physiological tension manipulations are required, mental arithmetic has been found to be one of the most reliable ways of creating cardiovascular change in subjects, increasing their oxygen consumption and raspatory rate (Turner and Caroll 1985, Seraganian et al. 1985, Levenson 1979). As a result, these studies suggest that conducting the mental arithmetic required to evaluate multi-dimensional prices may be a stressful task, which consumers may seek to avoid in order to avoid the associated physiological stress.

The above indicates that under circumstances typica of the environments in which MDPs are encountered, consumers may fail to numerically carry out the necessary mental arithmetics. Both Dawes et al. (1989) and the findings of Hoffman et al. (1987) indicate that in conducting many judgement tasks, judges tend to use the judgement cues as qualitative rather than quantitative (numerical) attributes. By doing so, judges ignore the necessary mental computations which need to take place. Dawes et al. (1989) believe this phenomenon to be due to the judges’ overconfidence in their accuracy. Hammond et al. (1987) on the other hand, attribute judges’ reluctance to explicitly use formulas to their fear of making mistakes when carrying out the necessary computations. In fact the later authors show that with increased time pressure and stimuli complexity, reliance on linear integration models increases. Based on the above arguments, it is therefore quite possible that consumers fail to account for the configural relationship between the price dimensions. We therefore hypothesize that:

H1: Consumer perceptions of MDPs are formed by independently integrating each price dimension into the overall price perception.

Dominance of the Monthly Payments

If the consumers’ perceptions of multi-dimensional prices are formed by integrating the price dimensions independently, as suggested by Hypothesis 1, an interesting question is whether consumers place more weight on one price dimension or another. For example, are consumers more sensitive to increases in the number of payments or the monthly payments? A budgetary perspective would suggest that the monthly payment dimension should be more critical than the number of payments in driving consumer perceptions of a multi-dimensional price. The monthly payments will in essence determine the amount of monthly budget that the consumer will have to allocate to the purchase. As a result, the monthly payment would directly impact the consumer’s living standards throughout the payment period, and is therefore highly influential in a consumer’s perception of an MDP. Thaler (1985), summarizing an unpublished 1982 field study concluded that: "the most relevant time horizon is the month since many regular bills tend to be monthly. Thus the budgeting process, either implicit or explicit, tends to occur on a month-to-month basis" (p. 207). We therefore conclude that:

H2: The monthly payment dimension has a higher weight than the number of payments in driving consumer price perceptions of MDPs.

Number of Dimensions in an MDP and Consumer Accuracy

One of the key concerns regarding multi-dimensional prices is that the existence of multiple price dimensions may have a negative influence on the 'processability’ of the price and the resulting accuracy in consumer price perceptions. It is likely that as the number of price dimensions increase, evaluation of the objective (rational) value of multi-dimensional prices becomes more difficult to make.

Studies of mental arithmetic processes, cited earlier indicate that with increases in the number of dimensions in a multi-dimensional price, the number of arithmetic processes that need to take place mentally (e.g., add, carry, hold) would increase. Such an increase could in turn negatively impact consumers’ price perception accuracy. Within the context of consumer decision making, a large body of lterature showing the inverse relationship between the number of judgement cues and consumer accuracy also exists. For example, Jacoby (1976) has demonstrates that increasing the amount of information available to the consumer can in fact hinder the consumer’s decision quality and force the consumer to make suboptimal choices. The work by Johnson and Payne (1985)-using simulation of choice processes-also concludes that stimuli complexity leads to choice sub-optimality. It is therefore proposed that:

H3: Increasing the number of price dimensions in a multi-dimensional price will reduce the consumers’ price perception accuracy.


In conducting this study, the approach of Anderson’s (1981, 1982) information integration theory is adopted. In dozens of studies, Anderson and his colleagues have demonstrated the capability of information integration methods in revealing the underlying integration models used in human judgement tasks. The information integration approach has been used by Bettman, Capon and Lutz (1975) in studying models of product attitude formation and by Gaeth et al. (1991) in studying consumers’ integration rule for judging the quality of product bundles, and has demonstrated its ability to detect complex integration models used by subjects (e.g., Stevenson 1986).

In the information integration approach, subjects are exposed to a factorial orientation of the stimuli under study (e.g., price dimensions), and their responses (e.g., price perceptions) to each profile are measured. The pattern of the subjects’ responses will enable the researcher to determine the structure of the underlying integration model used to consolidate the individual stimuli dimensions. Typically, studies employing the information integration methodology have two main characteristics: (1) the use of within subject designs where each subject is run through a factorial orientation of the stimuli dimensions, two or more times, and (2) the use of rating scales, most preferably graphic scales with end-anchors (Anderson 1982). Within subject designs are central to the information integration framework, since-unlike between subject designs-within subject designs prevent individual differences from contributing to the error term in the analysis of variance, enabling the study of the integration model at the individual level. They also are representative of many scenarios in which consumers are exposed to numerous offers in the market place, as in the case of multi-dimensional price ads shown in Figure 1. Graphic scales, although not required, are also commonly utilized for measuring the subjects’ external response. Typically, stimuli end-anchors-representing profiles slightly outside the range of the factorial design-are used to define the two ends of the graphic scale, and the experimental administration starts with stimuli representing these two end-anchors.

Once the appropriate measures have been obtained through the method described above, one is able to detect the form of the underlying price perception model by examining the form of interactions among the price dimensions [Anderson (1981, 1982) has developed two theorems which he terms the Parallelism and Linear Fan theorems of information integration theory to describe what is to follow. In order to avoid unnecessary terminology, the essence of these two theorems are briefly described instead.]. The sign of independent integration of the price dimensions is that the factorial plot of the subjects’ responses (i.e., MDP perceptions) exhibit a pattern of parallelism. The statistical test of parallelism is that the bi-linear interaction between the individual price dimensions must be statistically nonsignificant (Anderson 1982, p. 58). On the other hand, should the price perception model be multiplicative-as the rational model would require-the factorial plot of the consumers’ responses must exhibit a linear fan pattern, and the bi-linear interaction between specific price dimensions (e.g. between monthly payments and the numbr of payments) must be statistically significant. As such, satisfying this condition will imply that the underlying price perception model is multiplicative (Anderson 1982, p. 73).

Stimuli and Design

A within subject design was utilized. The design was intended to represent the real-life scenario where the consumer has to evaluate a series of MDPs in a limited time period. A typical example is when a consumer scans car dealership advertisements in a newspaper. Subjects were told that they were to evaluate the prices offered at different dealerships for a new Hyundai Excel. They were told that the prices vary in the monthly payments and the number of payments to be made, and that the potential buyer would obtain full ownership of the car at the end of the payment period. Multi-dimensional prices consisting of a full factorial combination of monthly payments M at three levels ($189, $229, $279) and number of payments N at three levels (24, 30, 36) were administered to each subject. The order of presentation of the price dimensions was counter-balanced across subjects. The subjects were asked to rate the expensiveness of each MDP profile on a 10 centimeter long graphic scale. The first two profiles corresponded to the end-anchors, representing stimuli slightly outside the range of the factorial design, and were intended to familiarize the subjects with the scale and to establish the scale’s end-anchors in their minds. The remaining profiles were the profiles from the factorial design and were randomized for each individual subject. After responding to the first set of profiles, subjects were administered a filler task, after which the same profiles were administered for a second time, but in a different random order. At the end, subjects were asked questions regarding their perception of interest rates, their experience level with financing packages, and also asked an open-ended question on how they evaluate multi-dimensional prices. Half of the subjects responded to MDP profiles with prices consisting of monthly payments (at 3 levels) and number of payments (at 3 levels), referred to as the MxN format. The other half of subjects responded to MDP profiles with prices consisting of down payment at 2 levels ($799, $1,199), monthly payments at 2 levels ($159, $239), and number of payments at 2 levels (24, 36), referred to as the DxMxN format. The primary reason for the inclusion of the second MDP format was to enable the study of the effects of increasing price complexity. Moreover, the levels of the individual price dimensions were chosen based on a survey of market prices for this class of cars, thereby providing ecologically valid stimuli for the experiment.




The subjects were graduate business students at an east coast educational institution who participated in this study in return for a monetary compensation ($5). Twenty subjects were administered profiles of the MxN format and 20 subjects were administered the profiles of the DxMxN format. Data from 4 additional subjects was obtained but could not be utilized, due to incomplete responses. To ensure the involvement of the subjects, they were told that the consistency of their responses will determine their chances of winning a small prize. Subsequently, 3 prizes valued at about $20 each were distributed to the three subjects which exhibited responses most consistent with the rational model (i.e., highest response correlation). The presence of the monetary rewards helped motivate the subjects to be more attentive and to make an effort to follow the rational model.


Table 1 shows the analysis of variance (ANOVA) table for the MxN profiles across all subjects. The dependent variable is perceived expensiveness, and the independent variables are the rice dimensions: monthly payments, and the number of payments. As can be seen from the ANOVA table, the main effects of the monthly payment dimension (M) and the number of payments dimension (N) are both significant. However, the bi-linear MxN interaction is not (p-0.972, h2=0.0004). Figure 2, showing the factorial plot of the response measure, clearly displays a pattern of parallelism, and no signs of a linear fan can be found.

A similar ANOVA was conducted for the MDPs in the DxMxN format and factorial plots were generated. As can be seen in Table 2, the main effects for monthly payment, number of payments, and down-payment are significant. However, as in the MxN case, the bi-linear interaction between monthly payments and number of payments is not statistically significant (p<0.737). Moreover, the factorial plots in the DxMxN format also exhibit a pattern of parallelism. Based on the parallelism theorem of information integration theory (Anderson 1982, p. 58), the above two indications, namely a non-significant interaction between the price dimensions and the observed parallelism of the factorial plots of the subjects’ responses, imply that the underlying integration model is not multiplicative, as the rational model would require. Instead the subjects seem to be forming their MDP price perceptions by independently integrating the monthly payments and the number of payments as suggested by Hypothesis 1.

In order to test Hypothesis 2, the marginal means in the DxMxN format were utilized. The DxMxN format was specifically designed such that the high level for each price dimension reflected a 50% increase over the low level (e.g. monthly payments at high=$239, and at low=$159). Moreover, as discussed earlier, the levels were chosen based on observations in the marketplace, providing for a reasonable level of ecological validity. This therefore enabled us to assess the relative impact of the various price dimensions on the overall price perceptions. As predicted by Hypothesis 2, the main effect for monthly payments, at 2.38 was found to be larger than the main effect for the number of payments, at 1.60 with a significance level of p<0.05. This indicates that the monthly payments have a larger impact on MDP price perceptions than do the number of payments. Additional support is provided by the h2 measure, which is an index of the amount of variation in the subjects’ responses explained by a given price dimension. The h2 for monthly payments was found to be 0.403, while for the number of payments it was only 0.221-thereby suggesting that the monthly payments explain more of the variance in price perceptions than do the number of payments. The same pattern was also observed in the DxMxN format where the h2’s for M and N were found to be 0.310 and 0.177, respectively, providing additional support for Hypothesis 2.

In order to test Hypothesis 3 regarding the accuracy of price perceptions when increasing the number of price dimensions, the correlation between the subjects’ perceptions of the MDPs and the objective values determined by the net present value of the offered MDPs was obtained. To calculate the objective net present values, each subject’s self-reported interest rate for consumer loans under $10,000 was elicited and used. The correlation was found to be 77.4% in the simpler MxN format. However, in the more complex DxMxN format, the correlation dropped to 71.5%. This difference, assessed using the Fisher z transformations (Morrison 1990, pp. 104-105), was found to be statistically significant at the p<0.01 level. Therefore Hypothesis 3-stating that increasing the number of price dimensions results in less accurate price perceptions-was supported.






The key results of the present study are as follows: (1) consumers form their price perceptions of MDPs by integrating the price dimensions independently, and (2) their price perceptions become less accurate as the number of price dimensions increases. Moreover, (3) they place a larger weight on the monthly payment amount than on the number of payments.

Clearly, the external validity of the above findings is limited by the use of a laboratory setting, and subjects that are not necessarily shoppers in the market for a given product. Therefore, replication of the results in field settings may be required. The design utilized in the study encouraged the subjects to provide their first impression of the price. This in fact is very similar to the common scenario where-as was shown in Figure 1-in the early stages of the decision making process, the consumer is bombarded with many MDP’s, for example through various dealership/retailer ads in a newspaper, the TV, or radio. Nevertheless, these initial price perceptions can largely influence the consumer’s subsequent search and purchase processes. As such, the findings are not representative of the detailed level of analysis one would expect consumers to undergo toward the final stages of the purchase process. Many questions remain to be answered. For example, do consumers use the same price perception model to evaluate other complex prices (e.g., service prices) in the market place? What is the role of individual differences (e.g, expertise, familiarity, involvement, etc.) in consumer price perceptions of multi-dimensional prices? How is the price/quality relationship affected in a multi-dimensional pricing context? Moreover, a more generalized framework for conceptualizing price dimensions may be possible. For example, each price dimension, in addition to its numerical value may vary in terms of its importance level and its salience in the consumer’s mind. These may further be influenced by factors such as the consumer’s level of experience and information search costs. The effect of the above factors on the MDP price perceptions model may therefore provide an interesting area for future research.

The findings of the study do however have certain implications for both marketing managers and policy makers. Marketing managers may need to reconsider their MDP communication strategies, by advertising MDPs that yield the lowest perceived level of expensiveness, especially in the early stages of the consumer decision making process. At the same time, policy makers need to pay careful attention such that MDP communication practices do not become deceptive and misleading in nature. As a minimum, it is hoped that this paper will inspire additional research in a relatively unstudied, yet substantive area of behavioral pricing research.


Agrawal, J., P. Grimm, and N. Srinivasan (1993), "Quantity Surcharge on Groceries", Journal of Consumer Affairs, 27, 2, 335-356.

Anderson, N.H (1981), Foundations of Information Integration Theory, Academic Press.

Anderson, N.H. (1982), Methods of Information Integration Theory, Academic Press.

Bettman, J. (1975), "Issues in Designing Consumer Information Environments," Journal of Consumer Research, 2, 169-177.

Bettman, J., N. Capon and H. Lutz (1975), "Cognitive Algebra in Multi-attribute Attitude Models," Journal of Marketing Research, v12 n2, May, 151-164.

Birnbaum, M. (1978), "Differences and Ratios in Psychological Measurement", i N. Castellan and F. Restle (eds.), Cognitive Theory (vol. 3, pp. 330-374), Hillsdale, NJ: Erlbaum.

Birnbaum, M., and R. Elmasian (1977), "Loudness Ratios and Differences Involve the Same Psychophysical Operation," Perception and Psychophysics, 23, 403-408.

Brehmer, B., R. Hagfors, and R. Johansson (1980), "Cognitive Skills in Judgment: Subjects’ Ability to Use Information about Weights, Function Forms, and Organizing Principles," Organizational Behavior and Human Performance, 26, 373-385.

Business Week (1994), "Leasing Fever," February 7, 1994, 92-96.

Capon, N. and D. Kuhn (1982), "Can Consumers Calculate Best Buys?", Journal of Consumer Research, 8, 4, 449-453.

Caroll, D., J. Turner and R. Prassad (1986), "The Effects of Level of Difficulty of Mental Arithmetic Challenge on Heart Rate and Oxygen Consumption," International Journal of Psychophysiology, 4, 3, 1670173.

Caroll, D., R. Turner, and S. Rogers (1987), "Heart Rate and Oxygen Consumption During Mental Arithmetic, a Video Game, and Graded Static Exercise," Psychophysiology, 24, 1, 112-118.

Cude, B, and R. Walker (1984), "Quantity Surcharges: Are They Important in Choosing a Shopping Strategy?", Journal of Consumer Affairs, 18, 2, 287-295.

Dansereau, D.F. and L.W. Gregg (1966), "An Information Processing Analysis of Mental Multiplication," Psychonomic Science, 6, 71-72.

Dawes, R.M., D Faust, and P. Meehl (1989), "Clinical Versus Actuarial Judgement," Science, 243, 1668-1674.

DeVinney, T. (ed.) (1991), Issues in Pricing: Theory and Research, Lexington Books, Lexington, MA.

Elmasian, R. and M. Birnbaum (1984), "A Harmonious Note on Pitch," unpublished manuscript.

Gaeth, G., I. Levin, G. Chakroborty and A. Levin (1991), "Consumer Evaluation of Multi-Product Bundles: An Information Integration Analysis," Marketing Letters, v2 n1,47-57.

Gourville, J. (1994), "Pennies-a-Day: Increasing Consumer Compliance through Temporal Re-Framing," Working Paper, Graduate School of Business, University of Chicago.

Greenleaf, E., V. Morwitz, and E. Johnson (1994), "Divide and Prosper: When Breaking Up is Good to Do," Marketing Science Institute working paper, July 15, 1994.

Groen, G., and J. Parkman (1972), "A Chronometric Analysis of Simple Addition," Psychological Review, 79, 329-343.

Hitch, G. (1978), "The Role of Short-term Memory in Mental Arithmetic," Cognitive Psychology, 10, 302-323.

Jacoby, J. (1976), "Consumer Psychology: An Octennium," Annual Review of Psychology, 27, 331-358.

Jacoby, J. and D. Speller (1974), "Brand Choice Behavior as a Function of Information Load-Replication and Extension," Journal of Consumer Research, June, 33-41.

Johnson, E., and J. Payne (1985), "Effort and Accuracy in Choice," Management Science, v31 n4, 395-414.

Levenson, R. (1979), "Effects of Thematically Relevant and General Stressors on Specificity of Responding in Asthmatic and Nonasthmatic Subjects, " Psychosomatic Medicine, 41, 28-39.

Linden, W. (1991), "What do Arithmetic Stree Tests Measure? Protocol Variations and Cardiovascular Responses," Psychophysiology, 28, 1, 91-102.

Lopes, L. (1982), "Toward A Procedural Theory of Judgement," Wisconsin Human Information Processing Program, WHIPP 17, Department of Psychology, University of Wisconsin, Madison, Wisconsin 53706.

Mellers, B., D. Davis, and M. Birnbaum (1984), "Weight of Evidence Supports One Operation for 'Ratios’ and 'Differences’ of Heaviness," Journal of Experimental Psychology, 10, 2, 216-230.

Mervis, C.B. and E. Rosch (1981), "Categorization of Natural Objecs," Annual Review of Psychology, 32, 82-115.

Monroe, K. (1990), Pricing: Making Profitable Decisions, 2nd Edition, McGraw-Hill.

Morrison, D.F. (1990), Multivariate Statistical Methods, 3rd edition, McGraw-Hill Series in Probability and Statistics.

Nagle, T. (1995), The Strategy and Tactics of Pricing: A Guide to Profitable Decision Making, Prentice Hall.

Nason, Robert, and Della Bitta, Albert (1983), "The Incidence and Consumer Perceptions of Quantity Surcharges," Journal of Retailing, 59, 2, 40-54.

Rao, V.R. (1993),"Pricing Models in Marketing," in J. Eliasberg and G. Lilien (eds): Marketing, North Holland.

Russo, E. (1977), "The Value of Unit Price Information," Journal of Marketing Research, May, 193-201.

Russo, J. and B. Dosher (1983). "Strategies for Multi-attribute Binary Choice," Journal of Experimental Psychology: Learning, Memory and Cognition, 9, 4, 676-696.

Seraganian P., J. Hanley, B. Hollander, and E. Smilgaet al. (1985), "Exaggerated Psychophysiological Reactivity: Issues in Quantification and Reliability," Journal of Psychosomatic Research, "4, 393-405.

Slovic, P. (1972), "From Shakespeare to Simon: Speculations and Some Evidence About Man’s Ability to Process Information," Oregon Research Institute Research Monograph, v12 n2.

Stevenson, M., "A Discounting Model for Decisions With Delayed Positive and Negative Outcomes," Journal of Experimental Psychology:General, 115,2,131-154.

Thaler, R. (1985), "Mental Accounting and Consumer Choice," Marketing Science, v4 n3, 199-214.

Turner, J., and D. Caroll (1985), "Heart Rate and Oxygen Consumption During Mental Arithmetic, a Video Game and Graded Exercise: Further Evidence of Metabolically Exaggerated Cardiac Adjustments?," Psychophysiology, 22, 261-267.

Tversky, A. and D. Kahneman (1974), "Judgement Under Uncertainty: Heuristics and Biases", Science, 185, 1124-1131.

Yadav, M. (1994), "How Buyers Evaluate Product Bundles: A Model of Anchoring and Adjustment," Journal of Consumer Research, 21, 2, September, 342-353.