Advances in Consumer Research Volume 12, 1985 Pages 85-90
THE EFFECT OF BRAND AND PRICE INFORMATION ON SUBJECTIVE PRODUCT EVALUATIONS
William B. Dodds, Virginia Polytechnic Institute and State University
Kent B. Monroe, Virginia Polytechnic Institute and State University
This paper reports an experiment on the effect of price and b.and information on perceptions of quality and value, and on willingness to buy. The research also investigated wi:ether perceptions differed when prices were odd or even. The results indicate that price positively influences the perception of quality, and inversely influences the perception of value and willingness to buy. Brand information enhanced the price effect, but there were no differences due to odd vs. even prices.
Do buyers use price as an indicator of product quality? Over the past thirty years researchers have attempted to answer that question. Yet, perhaps because there are some problems in how previous research has investigated the price-perceived quality relationship, there is conflicting evidence on the issue. Moreover, the effect of brand information on the perception of quality has received limited attention, again with inconsistent results. A third enigma on how price may influence buyers' evaluations of products is whether buyers respond more favorably to odd prices than to even prices. Despite the limited evidence indicating buyers to not respond differently between odd and even prices, basic marketing textbooks and sellers continue to believe in the odd-even phenomenon.
This paper reports an experiment that studied the influence of price and brand information, and the influence of odd and even prices on subjective product evaluations. After a brief review of the relevant literature, specific hypotheses are presented, and the research design, measures, and results are described. The conclusion discusses the results in terms of the influence of price and brand information as well as the odd-even price phenomenon on subjective product evaluations
The Effect of Odd-Even Pricing on Quality Perceptions
Do consumers perceive a difference in product quality and value based on small price differences? Do consumers react differently to a price of $39.95 as compared to a price of $40.00 for a given product? McCarthy and Perreault (1984, p. 619) feel marketers use odd prices because of the belief that consumers respond more favorably to these prices, perhaps because consumers perceive an odd price as substantially lower than the next highest even price. A random perusal of other introductory marketing texts indicates that other writers share this opinion. For example, Kotler (1980, p. 410) writes that many sellers believe that buyers favor odd prices over even prices. Instead of pricing a stereo amplifier at $300, the seller will price it at $299.95 or $295. Presumably, the customer perceives this odd price as a $200 price rather than a $300 price, or perceives it as a discount from the full price. Evans and Berman (1982, p. 538) point out that the evidence justifying the use of odd and even prices has been founded on "feelings" rather than research data. Stanton (1981, p. 264), while outlining the seller's basic belief in odd pricing, indicates that there is little concrete evidence to support sellers' beliefs in the value of odd prices.
While there is little empirical evidence that odd prices produce a more favorable buyer response than even prices, two studies in the past 15 years have addressed this question. Georgoff (1969) using a quasi-experimental design in a field setting examined ten products in a six store chain of department stores. Retail price endings were manipulated over a four week period with the dependent measures being sales. Results showed only random variations between price policy and sales. Lambert (1975) experimentally showed inconsistent findings on the impact of odd pricing. When paired with an even price at five different price levels, the odd price was perceived by subjects as statistically significantly lower in two situations, statistically significantly higher in one, and not statistically significantly different in the other two pairs. In a review of the price perception literature, Monroe (1973) indicated that the odd-even phenomenon assumes that the consumer is perceptually sensitive to certain prices (odd prices), and a departure from these prices (co an even price) results in a decrease in demand. Monroe concluded that there was no significant evidence supporting the psychological explanation of increased perceptual sensitivity. Moreover, research has shown that consumers have differential price thresholds such that a small difference in price will not likely produce a perception that the prices are different (Monroe and Petroshius 1981). Lambert's results seem to support this conclusion, yet the practice of using odd prices exist.
The Effect of Price on Quality Perception
Previous research on the relationship between price and perceived quality can be examined in two ways. First, single-cue studies generally have found a statistically significant price-perceived quality relationship. However, Olson (1977) has documented the limitations of single-cue studies in that they are overly simplified and the results concerning price effects have doubtful external validity, and limited internal validity.
Second, the multi-cue studies have manipulated other cues such as brand name, store image, and other information in addition to price. While attempting to overcome the limitations of the single cue studies, these multiple cue studies have typically found positive price-perceived quality relationships, although they have not always been statistically significant (Monroe and Krishnan 1984).
Potential confounds in the multi-cue studies that have led to this guarded conclusion may be similar to the role brand has played in these price-perceived quality studies. One key concern is whether the price differences in the price manipulations would likely produce perceptual discriminations by the subjects. Monroe and Krishnan (1984) suggest that not finding a statistically significant price-perceived quality relationship is inconclusive, if this result could be due to indiscernible price differences. Instead of relying on statistical significance to examine the relationship, Monroe and Krishnan (1984) examined effect sizes, and concluded that although there was support for a positive price-perceived quality relationship, the limited data base warranted a more intensive research effort.
The Effect of Brand on Quality Perception
Olson's (1977) review noted that the effect of brand name generally was strong in the price cue literature, appearing both as a main effect and as an interaction effect. However, a review of the six studies that examined both price and brand effects reveals a lack of a consistent and clear relationship. Table 1 shows that differences in the types of products utilized, the price manipulations, and the brand manipulations offers plausible reasons for the mixed outcomes. Monroe and Krishnan (1984) observe that previous conclusions indicate that brand name information dominated price information in the perception of quality. However, in their assessment of this literature, they discovered that price had a more positive effect on product quality perception when brand information was present than when brand information was absent. This finding suggests that the interaction of price and brand information not only is strong, but that the influence of price on quality perception becomes stronger in the presence of brand information than by itself. Thus, since past studies did not conceptualize the relationship in this manner, the magnitude of the interaction effect between price and brand name has not been investigated.
The hypotheses to be tested in this experiment are based on the conceptualization of the price perceived quality relationship as posited by Monroe and Krishnan (1984). In this structure three constructs, perceived quality, perceived value, and willingness to buy are utilized to model the role of price and quality perceptions. Arguing that perceived value and perceived quality are distinct constructs, these authors view perceived quality purely as an evaluative measure, whereas perceived value is represented as a tradeoff between perceived quality and sacrifice. As shown in Figure 1, price plays a dual role in this tradeoff. Higher prices will lead to greater perceived quality, and consequently, to a greater willingness to purchase based on perceived quality. At the same time, the higher price represents a measure of what must be sacrificed to purchase the good and leads to a lesser willingness to buy. Perceived value represents a tradeoff between the two variables, perceived quality and sacrifice. Willingness to buy is positively related to Perceived value.
RELATIONSHIP BETWEEN PERCEIVED QUALITY AND WILLINGNESS TO BUY
Based on this conceptualization and literature review, the following hypotheses were examined:
H1: As price increases subjects' perception of product quality, ceteris paribus, will also increase.
H2: (a) As price increases, subjects' perception of value, ceteris paribus, will at first increase and then decrease.
(b) As price increases, subjects' willingness to buy, ceteris paribus, will at first increase and then decrease.
(c) As price increases, subjects' perception of value and willingness to buy will follow the same pattern of change.
As price increases beyond an acceptable upper price, it would be expected that perceived quality would continue to increase. But, perceived value and also willingness to buy would both decrease because the sacrifice demanded becomes too important in the tradeoff with perceived quality. Thus, it would be expected that perceived value and willingness to buy would decrease after first increasing.
H3: Odd and even prices of the same general magnitude are not perceived by subjects differently; therefore there will be no differences in perceived quality, perceived value, or willingness to buy for odd or even prices of similar magnitude.
This hypothesis argues that consumers do not perceive a $79.95 product as a $70 or $79 model, but instead perceive it as essentially an $80 product. If there is a psychological reaction to an odd price, then it should conform to the price-perceived quality framework summarized above.
H4: The interaction of brand name and price will cause subjects to perceive the product to be higher in quality and value, and to be more willing to purchase the product, than subjects in a brand absent condition, at any given price.
This hypothesis argues that brand does not dominate price by its strong main effect, but rather enhances the price effect. Price and brand as extrinsic cues will individually and interactively enhance the quality perception of a product.
Research Design and Procedures
As shown in Figure 2, a 2x3x3x2 between groups factorial design was used. The three independent variables were brand, price, and odd-even prices. Brand was manipulated as either present or absent. Subjects were exposed either to a description of a Sony Walkman FM stereo cassette headset player or to an FM stereo cassette headset player without brand information. The data for the study was generated in an experimental setting where subjects were asked to assume interest in buying an FM stereo cassette headset player. The subjects were a convenience sample of 368 undergraduates enrolled in introductory marketing classes randomly assigned to 12 treatment groups with equal cell size of 29.
SUMMARY OF PRICE AND BRAND EFFECTS ON QUALITY PERCEPTIONS
Three pairs of prices were used. Three prices were even ($30.00, $80.00, $130.00) and three prices were odd ($29.95, $79.95, $129.95). Although the price treatments were nested within the odd-even factor, for the analysis, the odd/even and price manipulations were viewed as independent variables when crossed with the brand treatments.
The headset player was chosen because college students have shown interest in this product and would be assumed to have sufficient knowledge to evaluate the product. Each subject read a product description for the product and the retail price. A seven point rating scale was used for fourteen dependent indicators to measure the constructs of perceived product quality, perceived value, and willingness to buy. After responding to the specific treatment, subjects provided information about their demographic characteristics and their past experience with the product. Subjects also indicated what they thought the researcher was trying to determine in the experiment. This check on possible demand artifacts revealed no knowledge of the manipulation of independent variables nor of the Price-perceived quality relationship.
RESULTS AND ANALYSIS
Multi-item measures were used to assess the three constructs, perceived quality, perceived value, and willingness to buy. To assess the reliability of the measures, a correlation analysis and a factor analysis with a varimax rotation was performed on the fourteen dependent variables. The scale used was the same used by Petroshius (1983) except, based on her reliability assessments, three items were not used in this research.
In the factor analysis two predominant factors emerged accounting for 70.9% of the variance. The two factors described the constructs of perceived value and perceived quality. The third factor with an eigenvalue of .83. accounted for an additional variance of 6%, and had only one high loading item. Two other items planned a priori to load heavily onto this factor to give a multi-item measure of willingness to buy did not give anticipated loadings and were not analyzed with the single indicator for willingness to buy. The reliability of the product quality and value of offer constructs when analyzed using Cronbach's alpha resulted in values of .926 and .917 respectively. It was concluded that these two groups of indicators provided reliable measures of the perceived quality and perceived value constructs, permitting separate MANOVAs to be conducted for these constructs. The third construct, willingness to buy, with only one indicator was analyzed using ANOVA.
Analysis of Odd and Even Prices
The process of examining for an odd-even price effect was accomplished by breaking the overall design into three independent subdesigns where the low-priced pair (cells 1, 4, 7, 10), the medium-priced pair (cells 2, 5, 8, 11), and the high-priced pair (cells 3, 6, 9, 12) of prices were independently crossed with the brand/no brand treatments (Figure 2). The summaries for these three comparisons (Table 2) show statistically insignificant odd/even effects for the low-price and high-price pairs. But, the results for the middle-priced pair indicated statistically significant differences in perceived quality between the pair of prices. Since this finding runs counter to the hypothesized outcome of no effect, an inspection of the univariate statistics was done. All results at this level were statistically insignificant. Four measures favored the odd price and one highly insignificant result (p-,5) favored the even price.
OVERALL MANOVA SUMMARY: ODD/EVEN
The middle-priced results also had significant interactions between the odd/even price and brand/no brand for perceived value and willingness to buy. Since these interactions could be masking main effects, the middle-price comparison was reduced to two one-way designs (cells 2 and 5, and cells 8 and 11) to remove the interaction of the brand treatment. Table 3 shows a mixed result where the odd-even price treatments with brand information are significantly different in perceived quality and willingness to buy. There were no statistically significant differences in perceived value for the two prices. On the other hand, when the prices were evaluated without brand information, exactly opposite results were found for each construct.
MANOVA SUMMARY: MIDDLE PRICES
To explain this difference in results, the univariate statistics were examined. For the odd-even with brand information, two dependent measures, reliability and workmanship, were statistically significant at p=.02 and .04 respectively, and are the reason for the overall significance for perceived quality. When the significance of perceived value without brand information is examined, only one dependent measure, value for the money is statistically significant at p=.001.
Analysis of Price
Strong price x brand interactions (Table 4) precludes my direct interpretation of the price effect. After breaking the design into odd and even price designs (Figure 2), the strong interaction remained, although lot in a consistent pattern. This result was expected due to the conclusion that there was no perceived difference between odd and even prices.
OVERALL MANOVA SUMMARY- PRICE
For the odd price set (Table 5), perceived value and willingness to buy was influenced by a strong price x brand interaction. But, in the even price set, the perceived quality and perceived value constructs had strong interactions. To interpret these results, the two subdesigns were further reduced by separating out the brand treatments (Figure 2). Strong evidence that the three constructs change over a range of prices is provided by three of the four summary results (Table 6). Only the odd prices/no brand model showed a statistically insignificant price effect for willingness to buy. Further examination of the contrast showed the $79.95 model to have the most favorable response, followed by the $29.95 and then the $129.95 model.
TWO WAY MANOVA: PRICE
FOUR MANOVAS FOR PRICE EFFECTS
To be more precise in testing the hypothesis, the direction and the magnitude of the individual differences has to be done through multiple comparisons. Using Student Newman-Keuls multiple comparisons, when perceived quality was analyzed across price, strong evidence supported the hypothesis. Nineteen of twenty indicators over the four designs showed that as price increased from the low-price group to the medium-price group, perceived quality increased. Nine of these nineteen indicators were statistically different. Between the medium- and high-price groups, fifteen of twenty means were in the hypothesized direction although none were statistically different. The five directional discrepancies came within the odd price, brand name treatment, were the $79.95 model was perceived better in quality than the $129.95 model.
As might be expected, some indicators were more consistent than others. The dependent indicator workmanship produced only one significant difference between the $29.95 and $129.95 model without brand information. Perceived value with six indicators accumulated over the four designs consistently decreased as price increased. Twenty one of the twenty four contrasts were statistically different between the low- to medium-price groups. The contrasts between the medium and high price groups were also consistent, but only 9 of 24 were significantly different.
The construct willingness to buy, with the inherent problem of only one indicator, gave consistent results. As price increased, the willingness to buy was shown to decrease, with low price models most favorably viewed as attractive for purchase. Between the low- to medium-price models, three of the four groups showed a decreased willingness to buy with only one difference statistically significant. The contrast between the medium-priced models and the high-priced models supported the direction of the hypothesis with three of four groups showing a decreased willingness to buy. Using the results of Table 6 and the contrasts, the evidence is strong to argue for support of the price hypothesis. Perceived quality will increase and perceived value and willingness to buy will decrease as price increases.
Analysis of Brand Information
As discussed above, the brand main effect was masked by the interaction of price and brand. Because of this interaction, the same pattern of analysis was followed as in the evaluation of price effects. The results (Table 2) indicate that at the low and high prices the treatments with the brand name were significantly better over the three constructs. In general, for each of the indicators, the brand effect favored the brand name and in most cases, the differences were statistically significant. A puzzling and contradictory result occurred in the high-price group, where the direction of the perceived value effect favored the no brand treatment for five of the six indicators, with one of the indicators statistically significant. Replication could resolve whether this is a statistical oddity or a pricing phenomenon that deserves further attention.
To examine the middle-price group the design was broken apart by odd and even prices (cells 2, 5, 8, 11). The MANOVA results showed that the brand effect for a price of $79.95 was statistically significant for perceived quality, with no effect for perceived value, and willingness to buy. An analysis of the contrasts showed all indicators favored the brand treatment for perceived value and willingness to buy. The results at $80.00 indicated statistically significant results for all constructs. An examination of the indicators of perceived value indicated that three of the five indicators favored the no brand treatment as being higher in perceived value.
DISCUSSION AND CONCLUSIONS
This experiment was conducted to examine the effect of brand and price information on how individuals subjectively evaluate products. Multiple constructs, perceived quality, perceived value, and willingness to buy, were measured in a 2x3x2 factorial design that varied brand information, price levels, and even presentations for a given price level.
The hypothesis claiming subject's perception of product quality to increase as price increases is confirmed. For the product used in this study, evidence was strong that a price-perceived quality relationship existed. The relationship appeared to be strong for a low to medium price comparison than the medium to high price contrast. One anomaly stood out alone with evidence against the hypothesis. The $79.95 Sony Walkman was perceived to be of better quality than the $129.95 Sony Walkman. This result did not arise in the other sub-designs where no brand information and/or even prices were used.
The first two parts of the second hypothesis were rejected since most contrasts showed the low-priced stereo player to be consistently highest in perceived quality and willingness to buy, while the medium and high priced players showed decreasingly lower evaluations for the two constructs. An explanation for this result would be that the hypothesis was assuming that the price points being tested ranged over the acceptable price limits whereas the low- and high-price treatments may have been outside this range. Therefore, only the decrease in perceived value and willingness to buy was shown empirically to occur. Since these two constructs did follow the same pattern, then the third part of this hypothesis is accepted.
Odd and Even Prices
The overall conclusion is to accept the third hypothesis that odd and even prices are not perceived by subjects differently. The high and low price pairs were strongly convincing with their statistically insignificant values, while the middle price group needed more extensive analysis to find evidence of no effect. For perceived quality, the significance of the odd-even effect was explained by the insignificant but opposite direction of the means of the univariate measures. For the perceived value and willingness to buy constructs, the observation is that only a few of the multiple measures influenced the significance of the odd-even effect, possibly due to chance. While not totally conclusive, the evidence seems to support no difference in perceived quality, perceived value, and willingness to buy between odd and even prices over the three price levels and the two brand treatments.
The interaction of brand name and price caused subjects to perceive the three constructs to be higher in quality and value, and to be more willing to purchase the product than when brand name is absent. Not only this, but the evidence is sufficient to argue that the brand effect did not dominate price by its strong effect but enhanced the price effect.
In conclusion, using a brand name significantly increased perceived quality and willingness to buy as compared to using no brand. This statistically significant effect held over all price levels. Perceived value showed the hypothesized effect in the low-price level but failed to act consistently for all indicators in the medium- and high-price treatments.
Andrews, I. R. and Valenzi, Enzo K. (1971), "Combining Price, Brand and Store Cues to Form an Impression of Product Quality," Proceedings, 79th Annual Convention of the American Psychological Association, 649-50.
Evans, Joel R. and Barry Berman (1982), Marketing, New York: Macmillan Publishing Co.
Gardner, David M. (1971), "Is There a Generalized Price/Quality Relationship?" Journal of Marketing Research, 8 (May), 241-243.
Georgoff, David M. (1969), "Price Illustration and the Effect of Odd-Even Retail Pricing," Southern Journal of Business, 4, (April, 95-103.
Jacoby, Jacob, Olson, Jerry C. and Haddock, Rafael A. (1971), "Price, Brand Name and Product Composition Characteristics as Determinants of Perceived Quality," Journal of Applied Psychology, 55 (no. 6), 570-579.
Kotler, Phillip (1980), Principles of Marketing, Englewood, New Jersey: Prentice-Hall.
Lambert, Zarrel V. (1975), "Perceived Prices as Related to Odd and Even Price Endings," Journal of Retailing, 51, (Fall) 13-22+.
McCarthy, Jerome E. and William D. Perreault, Jr. (1984), Basic Marketing, Homewood, Illinois: Richard D. Irwin.
Monroe, Rent B. (1973), "Buyers' Subjective Perceptions of Price," Journal of Marketing Research, 10, (February), 70-80.
Monroe, Kent B. and R. Krishnan (1984), "The Effect of Price on Subjective Product Evaluations," in J. Jacoby and J. Olson (eds.), The Perception of Merchandise and Store Quality, Lexington, MA: D. C Heath (forthcoming).
Monroe, Kent B. and Susan M. Petroshius (1981), "Buyers' Perceptions of Price: An Update of the Evidence," in Harold H. Kassarjian and Thomas S. Robertson (eds.), Perspectives in Consumer Behavior, 3rd edition, Glenview, IL: Scott, Foresman and Company, 43-55.
Olson, Jerry C. (1977), "Price as an Informational Cue: Effects on Product Evaluations," in Arch G. Woodside, Jagdish N. Sheth and Peter D. Bennett, eds., Consumer and Industrial Buying Behavior, New York: North Holland, 267-286.
Peterson, Robert A. and Jolibert, Alain J. P. (1976), "A Cross-National Investigation of Price and Brand as Determinants of Perceived Product Quality," Journal of Applied Psychology, 61, (no. 4), 533-536.
Petroshius, Susan M. (1983), An Experimental Investigation of the Product Line Pricing Characteristics on Product Evaluations, Unpublished Doctoral Dissertation, Virginia Polytechnic Institute and State University, Blacksburg, Virginia.
Raju, P. S. (1977), "Product Familiarity, Brand Name and Price Influences on Product Evaluation," in W. Perreault, ed., Advances in Consumer Research 4, Ann Arbor, MI: Association for Consumer Research, 64-71.
Stanton, William J. (1981), Fundamentals of Marketing, New York: McGraw-Hill Book Co.
Wheatley, John J., Walton, Richard G. and Chiu, John S. Y. (1977), "The Influence of Prior Product Experience, Price and Brand on Quality Perception," in W. Perreault, ed., Advances in Consumer Research, 4, Ann Arbor, Ml: Association for Consumer Research, 72-77.