Effects of Prior Belief on Feature-Based Price Estimates

ABSTRACT - When advertising models in a product line, sellers sometimes describe the model features but do not provide the specific price of each model. This article examines how buyers, in such situations, estimate the price of an individual model based on the features it offers. We find that price estimates depend on the features offered as well as on buyer characteristics. Specifically, we find that buyers who strongly believe that quality and price are positively related in the product category, provide higher price estimates of the models than those who are less encumbered by this belief. The study also finds that the difficulty in assessing the importance of the features lowers the price estimates. Implications of these findings are discussed.

Citation:

Tridib Mazumdar and Cheoul Ryon Kim (1993) ,"Effects of Prior Belief on Feature-Based Price Estimates", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 586-590.

Advances in Consumer Research Volume 20, 1993      Pages 586-590

EFFECTS OF PRIOR BELIEF ON FEATURE-BASED PRICE ESTIMATES

Tridib Mazumdar, Syracuse University

Cheoul Ryon Kim, Syracuse University

ABSTRACT -

When advertising models in a product line, sellers sometimes describe the model features but do not provide the specific price of each model. This article examines how buyers, in such situations, estimate the price of an individual model based on the features it offers. We find that price estimates depend on the features offered as well as on buyer characteristics. Specifically, we find that buyers who strongly believe that quality and price are positively related in the product category, provide higher price estimates of the models than those who are less encumbered by this belief. The study also finds that the difficulty in assessing the importance of the features lowers the price estimates. Implications of these findings are discussed.

INTRODUCTION

Stores sometimes advertise the attributes or features of several models of a product line but do not provide specific price information for each model. For example, consumer electronics stores may advertise several models of VCR with different features and indicate that prices range from $200 to $500. Real-estate firms may advertise "Homes from $150,000" and then provide descriptions (or pictures) of the models (e.g., number of bedrooms, wooded lot, cul-de-sac etc.). While these types of price ads are being used by retailers, it is unclear whether these practices are more effective than providing buyers with specific price information.

When specific price of a model (e.g., a preferred model) is unknown, buyers may either initiate external search for the price information or they may try to estimate its price based on the feature descriptions provided in the ad. These estimates may have important consequences for buyers' subsequent store patronage as well as their choice decisions. For example, if the price estimate of an otherwise preferred model exceeds the acceptable price limit, buyers may exclude the brand from their consideration set and decide not to visit the store at all. Also, if buyers decide to actually visit the store, the initial price estimate may serve as a reference point for judging the actual purchase price. Clearly, the price judgment and choice decision of the buyer will depend upon whether the actual price compares favorably or unfavorably with their original estimate (Winer 1986).

The objective of this article is to report the results of an exploratory study that examines how feature-based price estimates are influenced by: (1) the degree to which buyers believe that there is a positive association between quality and price in the product category (hereafter referred to as either "prior belief" or simply "belief") and (2) the difficulty experienced by buyers in assessing the importance (or usefulness) of the features offered by different models. Conceptual arguments leading to specific hypotheses are presented next. We then discuss the procedures and results to test the hypotheses. Finally, we discuss the managerial implications of the findings and highlight the limitations of the study.

CONCEPTUAL BACKGROUND

Prior Belief

Buyers, through a variety of direct or indirect experiences, detect relationships among marketing phenomena. When reinforced, these associations develop into beliefs such as, Japanese cars are more reliable than domestic cars, convenience stores are more expensive than supermarkets, specialty outlets carry better quality products than discounters, higher-priced brands are of better quality than lower-priced brands. Beliefs such as these are considered efficient ways for consumers to organize past experiences in memory (Crocker 1981; Alloy and Tabachnik 1984) and use them for subsequent evaluation and categorization of stores and merchandise (Peterson and Wilson 1985; Zeithaml 1988). However, strongly held beliefs, once formed, are often resistant to changes even in the face of conflicting evidence, resulting in biased consumer judgments (Bettman, John, and Scott 1986) and search behavior (John, Scott, and Bettman 1986).

In this paper we focus on one such belief that involves a positive association between product price and quality. Considerable research has been done to examine the validity of a positive price-perceived quality relationship (Monroe and Krishnan 1985; Zeithaml 1988; Rao and Monroe 1989). From a normative standpoint, this relationship is considered natural because products with superior performance may require deployment of greater resources and therefore, firms should receive higher prices to recover additional costs (Scitovsky 1945; Lancaster 1966; Ratchford 1975). However, several studies have shown that product prices are not always ordered according to their "objective" (e.g., Consumer Reports) qualities, causing the price-"objective" quality relationship to vary in its strength as well as direction across product class (Oxenfeldt 1950; Morris and Bronson 1969; Riesz 1978; 1979; Geistfeld 1982; Gerstner 1985; Tellis and Wernerfelt 1987; Kamakura, Ratchford, and Agrawal 1988). Evidence also suggests that buyers' reliance on price for judging product quality depends on whether or not there is a positive relationship between price and "objective" quality (Lichtenstein and Burton 1989). In summary, buyers may be "aschematic" or "schematic" (Peterson and Wilson 1985 p. 263), depending upon whether the quality-price schema in buyers' memory is weak or strong respectively in a given evaluative context.

That the strength of belief about a positive quality-price relationship varies implies that the price estimate of a given model based on the features it offers may also vary as a result. [Product quality is a multidimensional construct (Zeithaml 1988). Buyers may define quality on abstract dimension such as durability, reliability, trouble-free performance, or superior workmanship. However, in many product categories, the basic or the core product is the same across models or brands, but additional features discriminate higher quality models from lower quality ones. In this paper, we examine the latter situation where buyers equate product quality with intrinsic attributes or features offered by a given model or brand. These types of situation have been identified in past research (e.g. Alba and Marmorstein 1987; Zeithaml 1988).] That is, when estimating price of the model, buyers not only take into account the feature information, but their prior belief also plays an important role in the estimation process (Chapman and Chapman 1967; Jennings, Amabile, and Ross 1980). Researchers have shown that buyers who strongly believe that price and product quality is positively related have a higher level of acceptable prices than those who do not believe that such a relationship exists (Lichtenstein, Bloch, and Black 1988). Extending this finding to the current context suggests that when a model contains a number of features, a strong belief may lead buyers to infer that the model should also cost more. In contrast, buyers who are relatively unencumbered by this belief may consider only the objective feature information and weigh these features according to their relative importance weights to arrive at a price estimate. Thus, given a set of feature descriptions of several models, a strong belief about a positive quality-price association may serve to raise the price estimates of the models. The following hypothesis is therefore proposed:

H1: Given a set of feature descriptions of several models, the stronger the belief about a positive quality-price association in the product category, the higher will be the price estimates for the models.

Difficulty in Assessing Feature Usefulness

When advertising models of a product, retailers often provide a rather long list of features each model contains, and buyers may experience difficulty in assessing the usefulness of these features. Researchers have noted that buyers experience uncertainty not only about the existence of certain features in a given brand, but also about the importance of these features (Urbany, Dickson, and Wilkie 1989). Inability to assess the importance of features may be caused by buyers' lack of familiarity with the product class or because they do not possess the necessary knowledge to determine what functions the features perform (Brucks 1985; Alba and Hutchinson 1987). Even in instances where buyers are capable of comprehending the functional property of a feature, they may still experience uncertainty about its importance because they are unable to forecast accurately when and how frequently they will use these features (Kahn and Meyer 1991).

Kahn and Meyer (1991) show that buyers perceive greater value for an extra feature offered in a product when there is less uncertainty surrounding the importance weight they should assign to the feature. In light of this evidence, we may argue that buyers who can predict the usefulness of the features offered in a model may estimate a higher price for the features than buyers who experience difficulty in assessing the utilities of the features. We propose the following hypothesis:

H2: The more difficulty buyers experience in predicting the usefulness of the features contained in a model, the lower will be the estimated price of the model, ceteris paribus.

Interactive Effects

The relative difficulty experienced by buyers in predicting the usefulness of the features may also mediate the effect of buyers' prior belief about a quality-price relationship on their price estimates (proposed in H1). Buyers who can not discriminate the important features from the unimportant ones may assign equal weights to each feature (Park 1976) and simply count the number of features to make an assessment of the overall value of a product (Brucks 1985; Alba and Marmorstein 1987). Less knowledgeable buyers may also rely on extrinsic cues (i.e., cues not related to the physical product itself) such as, brand name or product price to evaluate product quality (Park and Lessig 1981; Rao and Monroe 1988). Similarly, when buyers try to estimate prices of different models based on the features these models offer but are relatively uncertain about the potential usefulness of the features, they are more likely to invoke their belief about the quality-price relationship than buyers who are more able to assess the importance of the features.

H3: The greater the difficulty in assessing the usefulness of the features offered by a model, the stronger will be the positive effect of prior belief on buyers' price estimates (proposed in H1).

RESEARCH METHOD

Stimulus

As noted earlier, buyers may evaluate a model from the physical features as well as on dimensions such as durability, reliability, and workmanship, which are not readily observable. Since the effects of unobservable dimensions on price estimates of study participants are difficult to control and assess, we decided to use a product class where product quality could be judged on the basis of the presence or absence of features (see also footnote).

The product chosen was "electric iron." Informal discussions with salespersons of several appliance stores revealed that all models of electric irons perform the same basic functions, but the quality and price levels of the models vary because of the additional features each model offers. From the catalog of a local merchandiser, we picked eleven models and described them on six features using a matrix format. The number of features in each model ranged from one to six (see Table 1). To eliminate the brand name effect, the real brand names were concealed and the models were placed in the rows in random order (with identifications A through K). The six features namely, temperature light (TEMP), auto shut-off (AUTO), self-clean (CLEAN), extra steam spray (STEAM), non-stick coating (NOSTICK), and water window (WW), were presented in the column, again in random order. The cells provided the feature values (present or absent). Many store catalogs use this format to present feature information of different brands or models, thus making our stimulus presentation format ecologically valid.

Study Participants

Eighty undergraduate students (40 males and 40 females) enrolled in two sections of introductory marketing management course were selected for the study. The participants were told that the study was aimed at assisting a local appliance store in setting prices of different models of irons.

Measures of Independent Variables

Quality-Price Belief. Following the procedure used by previous researchers (e.g., Bettman et. al. 1986; John et. al. 1986), participants' belief concerning a positive quality-price relationship was measured rather than manipulated. On a seven point scale, participants responded to what extent they agreed that "the better the quality of an electric iron, the higher will be its price".

Difficulty in Assessing Feature Usefulness. Participants were told that the models of electric irons available in the local store may contain one or more of the six features described earlier. They were then asked to rate on a seven point scale the difficulty they would experience in predicting the usefulness of these features, relative to that experienced by the average student population (significantly lower - significantly greater). This type of self-perceived relative knowledge measure has also been used in past research (e.g., Brucks 1985).

Quality Ratings and Price Estimation Tasks

The participants then received the stimulus information and were asked to provide quality ratings for each model using a 0-10 point ["substantially below average quality(0) - substantially above average quality (10)"] scale. They were free to use any criteria for making quality judgments. This quality measure was needed to ascertain if the participants indeed perceived the brands with many features as better quality models than those with fewer features.

TABLE 1

MEAN QUALITY RATINGS AND ESTIMATED PRICES

After rating the qualities, the participants were told that prices of the eleven models range from $13 to $39 in the local store but the prices of individual models are unknown. The participants were then asked to provide an estimate of the actual price that they would be expected to pay in the store for each of the models. They were then asked to conjecture what the true purpose of the study was. None related the price estimation with quality-price belief or perceived feature usefulness issues. They were then debriefed and thanked for their participation.

ANALYSIS AND RESULTS

The number and types of features offered by each of the eleven models and the respective mean quality ratings and the price estimates are presented in Table 1. Model J and I, each with one feature, obtained the lowest quality ratings. At the high end, Model C contained all the features and Model G had five features. These two models were judged the best and the second best quality models. A high correlation of 0.86 (p<0.0001) between the number of features and the overall quality rating ensured that as intended, the participants did use the provided feature information to infer the quality of respective models.

Tests of Hypotheses

The belief about a positive quality-price association (H1) and the difficulty in assessing the usefulness of the features (H2) were respectively hypothesized to be positively and negatively related with buyers' price estimates based on feature information. It was also postulated (H3) that greater difficulty in assessing the feature usefulness would enhance the effect of quality-price belief on buyers' price estimates. Thus, a positive interaction effect between difficulty in assess feature usefulness and prior belief is expected. To test these hypotheses, the following regression model was tested:

PRICEj =f(BELIEF, DIFF, BELIEF xDIFF,DUMj)(1)

PRICE_j =the estimated price for model j, j=1,2,..... 11;

BELIEF = respondents' self-reported belief scores (1-7),

DIFF = respondents' self-reported relative difficulty in assessing feature usefulness (1-7),

DUM_j = model-specific dummies taking values of 1 if model=j and 0 otherwise.

Note that the structure of the model is similar to that of Analysis of Covariance, where the dummies represent the discrete levels. The coefficients of the dummies reflect the effects of the feature combinations contained in the models on the price estimates. The coefficients associated with BELIEF, DIFF, and the interaction term will capture the incremental effects of the respective variables and the interaction term on the price estimates for all the eleven models combined.

The parameter estimates and the corresponding t-statistics are reported in Table 2. Since the model dummies are ordered according to increasing price estimates, all dummy coefficients are positive. An adjusted R² of 0.73 indicates a good overall fit of the model. In addition, the coefficient for the BELIEF variable is positive and significant, supporting H1. The DIFF coefficient is in the hypothesized direction (i.e., negative) but is significant only at p<0.10, providing a moderate support for H2. Finally, as predicted we have a positive BELIEF x DIFF interaction, but the effect is not significant at the conventional level.

TABLE 2

REGRESSION RESULTS

SUMMARY AND DISCUSSION

Buyers' price estimates of a model are found to be a function of the features it offers, weighted by the perceived importance of the features. [To assess the importance weights of the respective features, an analysis of variance was performed with price estimate only as a function of the features, each with two levels (present or absent). The t-values for each feature are statistically significant (p<0.0001). The respective t-values in increasing order are: STEAM=8.29, AUTO=8.73, NOSTICK=9.46, TEMP=9.70, WW=11.83, CLEAN=16.24.] The price estimates are also found to be influenced by two characteristics of the buyers, namely (1) their belief concerning a positive quality-price relationship in the product category and, (2) the relative difficulty they experience in assessing the usefulness of different features.

First, the study finds that a strongly held belief about a positive quality-price relationship serves to increase buyers' price estimates of a given model. This finding has important implications in managerial decisions involving whether to provide potential buyers a price range or specific prices for models being advertised. Past research indicate that the belief about a positive quality-price relationship vary across product classes (e.g., durable versus non-durables) (Zeithaml 1988; Lichtenstein and Burton 1989). In product categories where the belief is strong, our result suggests that buyers' price estimates of a model based on its feature may be biased upwards. Since these estimates may serve as reference points for buyers' future price judgments, firms may capitalize on this bias by advertising models in these product categories by presenting feature descriptions and a price range rather than specific prices. However, it may be important to note that based on a strong quality-price belief, buyers may estimate prices of certain models that exceed their upper thresholds of acceptable prices. Buyers, in such situations, may judge the brands as "too expensive" and decide not to visit the store (Monroe and Petroshius 1981).

The second finding of our study is that buyers' price estimate of a model based on its feature(s) is negatively related to the difficulty in assessing the usefulness of these features. This finding implies that in instances where the advertised brands or models belong to a new product class and where the benefits of the features are unclear and ambiguous to potential buyers, the price estimates based on feature descriptions are likely to be biased downwards. In such cases, firms may wish to advertise specific prices of each model, which will eliminate the need for estimating prices of models based on feature information. Alternatively, firms could try to provide clear explanations about the benefits of the features and help potential buyers visualize the occasions and frequency of using these features.

Limitations and Future Research Directions

As noted in the beginning of this article, the research reported here was carried out as an exploratory research. There are several ways the research can be improved. From a theoretical standpoint, since the price estimation is used as a judgment task, the research needs to incorporate the preference reversal and anchoring and adjustment literature to provide a stronger theoretical foundation and propose additional hypotheses. From a methodological perspective, there is a clear need for using multiple and reliable measures of quality-price belief and difficulty in assessing feature importance so that the study results can be unambiguously interpreted. We also did not rotate the orders across subjects when presenting the model descriptions. This may have resulted in anchoring and adjustment biases in subjects' price estimates. Finally, we realize in retrospect that the product class used here (i.e., iron) may not have been relevant for student subjects. To improve the generalizability of the findings, the study must be replicated using other product categories.

REFERENCE

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

Alba, Joseph W. and Howard Marmorstein (1987), "The Effects of Frequency Knowledge on Consumer Decision Making," Journal of Consumer Research, 5, 14 (June).

Alloy, Lauren B. and Naomi Tabachnik (1984), "Assessment of Covariation by Humans and Animals: The Joint Influence of Prior Expectations and Current Situational Information," Psychological Review, 91(January), 112-149.

Bettman, James R., Deborah Roedder John, and Carol A. Scott (1986), "Covariation Assessment by Consumers," Journal of Consumer Research,13 (December), 316-326.

Brucks, Merrie (1985), "The Effects of Product Class Knowledge on Information Search Behavior," Journal of Consumer Research, 12, 1-16.

Chapman, Loren J. and Jean P. Chapman (1967), "Genesis of Popular but Erroneous Psychodiagnostic Predictions," Journal of Abnormal Psychology, 72(3), 193-204.

Crocker, Jennifer (1981), "Judgment of Covariation by Social Perceivers," Psychological Bulletin, 90 (September), 272-292.

Geistfeld, Loren V. (1982), "The Price-Quality Relationship - Revisited," Journal of Consumer Affairs, 16(Winter), 334-335.

Gerstner, Eitan (1985), "Do Higher Prices Signal Higher Quality?," Journal of Marketing Research, 22, 209-215.

Jennings, Dennis L., Teresa M. Amabile, and Lee Ross (1980), "Informal Covariation Assessment: Data-based versus Theory-based Judgments," in Judgment Under Uncertainty: Heuristics and Biases, eds. Daniel Kahneman, Paul Slovic, and Amos Tversky, Cambridge, England: Cambridge University Press, 211-230.

John, Deborah Roedder, Carol A. Scott, and James R. Bettman (1986), "Sampling Data for Covariation Assessment: The Effect of Prior Beliefs on Search Patterns," Journal of Consumer Research, 13 (June), 38-47.

Kamakura, Wagner A., Brian T.Ratchford, and Jagdish Agrawal (1988), "Measuring Market Efficiency and Welfare Loss," Journal of Consumer Research, 15 (December), 289-302.

Kahn, Barbara E. and Robert J. Meyer (1991), "Consumer Multiattribute Judgments under Attribute-Weight Uncertainty, Journal of Consumer Research, 17 (March), 508-522.

Lancaster, Kelvin J. (1966), "A New Approach to Consumer Theory," Journal of Political Economy, 74(April), 132-157.

Lichtenstein, Donald R. and Scot Burton (1989), "The Relationship Between Perceived and Objective Price-Quality," Journal of Marketing Research, 26, 429-443.

Lichtenstein, Donald R., Bloch, Peter H., and William C. Black (1988), "Correlates of Price Acceptability," Journal of Consumer Research, 15, 2(September), 243-252.

Monroe, Kent B. and R. Krishnan (1985), "The Effect of Price on Subjective Product Evaluations," in Perceived Quality, J.Jacoby and J.Olson, eds. Lexington, MA: Lexington Books, 209-232.

Monroe, Kent B. and S. Petroshius (1981), "Buyers' Perceptions of Prices: An Update of the Evidence," in T. Robertson and H. Kassarjian (eds.), Perspectives in Consumer Behavior, 3rd ed. Glenview, Il.: Scott Foresman, 23-42.

Morris, Ruby T. and Claire S. Bronson (1969), "The Chaos in Competition Indicated by Consumer Reports," Journal of Marketing, 34, 103-112.

Oxenfeldt, Alfred N.R. (1950), "Consumer Knowledge: Its Measurement and Extent," Review of Economics and Statistics, 32, 300-314.

Park, C. Whan (1976), "The Effect of Individual and Situation Related Factors on Consumer Selection of Judgment Methods," Journal of Marketing Research, 13 (May), 144-151.

Park, C. Whan and V. Parker Lessig (1981), "Familiarity and Its Impact on Consumer Biases and Heuristics," Journal of Consumer Research, 8 (September), 223-230.

Peterson, Robert A. and William R. Wilson (1985), "Perceived Risk and Price-Reliance Schema and Price-Perceived-Quality Mediators," in Perceived Quality, J.Jacoby and J.Olson, eds. Lexington, MA: Lexington Books, 247-268.

Rao, Akshay R. and Kent B. Monroe (1988), "The Moderating Effect of Prior Knowledge on Cue Utilization in Product Evaluations," Journal of Consumer Research, 15(2), 253-264.

Rao, Akshay R. and Kent B. Monroe (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Perceptions of Product Quality: An Integrative Review," Journal of Marketing Research, 26 (August), 351-357.

Ratchford, Brian T. (1975), "The New Economic Theory of Consumer Behavior: An Interpretive Essay," Journal of Consumer Research, 2, 65-75.

Riesz, Peter C. (1978), "Price Versus Quality in the Marketplace, 1961-1975," Journal of Retailing, 54(Winter), 15-28.

Riesz, Peter C. (1979), "Price-Quality Correlations for Packaged Food Products," Journal of Consumer Affairs, 13(Winter), 236-247.

Scitovsky, T. (1945), "Some Consequences of the Habit of Judging Quality by Price," Review of Economic Studies, 12, 100-105.

Tellis, Gerard J. and Birger Wernerfelt (1987), "Competitive Price and Quality under Asymmetric Information," Marketing Science, 6 (Summer), 240-253.

Urbany, Joel E., Peter R. Dickson, and William L. Wilkie (1989), "Buyer Uncertainty and Information Search," Journal of Consumer Research, 16 (September), 208-215.

Winer, Russell (1986), "A Reference Price Model of Brand Choice for Frequently Purchased Products," Journal of Consumer Research, 13, 250-256.

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence," Journal of Marketing, 52, 2-22.

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