When Do Price Promotions Signal Quality? the Effect of Dealing on Perceived Service Quality

Priya Raghubir, Hong Kong University of Science and Technology
Kim P. Corfman, New York University
[ to cite ]:
Priya Raghubir and Kim P. Corfman (1995) ,"When Do Price Promotions Signal Quality? the Effect of Dealing on Perceived Service Quality", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 58-61.

Advances in Consumer Research Volume 22, 1995      Pages 58-61

WHEN DO PRICE PROMOTIONS SIGNAL QUALITY? THE EFFECT OF DEALING ON PERCEIVED SERVICE QUALITY

Priya Raghubir, Hong Kong University of Science and Technology

Kim P. Corfman, New York University

In this paper it is hypothesized and demonstrated that offering a price promotion on a service can unfavorably affect brand evaluations. This is particularly true when price promotions are uncommon in the industry and consumers perceive that the quality of firms in the industry varies. Further, it is demonstrated that the effect of promotions on brand evaluations is mediated by the attribution made for the promotion. Results from a laboratory experiment are reported.

It is universally acknowledged that sales promotions account for a large and increasing percentage of the marketing budget, and that they increase brand sales (e.g., Chevalier 1975, Neslin and Shoemaker 1989). It is less universally acknowledged that offering a price promotion can lead to unfavorable brand evaluations. On one hand, advertising gurus warn about the dangers of using sales promotions at the expense of advertising as part of a complete marketing strategy (Ogilvy 1963). On the other hand, marketing academics have searched, almost in vain, for a negative effect of sales promotions on brand evaluations (e.g., Davis, Inman and McAlister 1992).

In this paper we revisit the issue of whether offering a promotion affects brand evaluations and look at the conditions under which it might do so. We explore whether the effect of a promotion is moderated by industry factorsCspecifically, perceptions of how common it is to deal in a particular industry and the perceived range of quality among firms in the industry. We also examine whether the effect of promotions on brand evaluations is mediated by whether the promotion is attributed to brand-related vs. non-brand-related reasons.

LITERATURE

The literature on the effect of promotions on brand evaluations is surprisingly sparse and equivocal. The vast majority of prior literature that has assessed the effects of sales promotions on brand evaluations has studied the effect after product trial (Scott and Yalch 1980, Scott and Tybout 1979, Tybout and Scott 1983). Scott and her colleagues examined the effect of promotions on evaluations after subjects had tried the promoted brand. They found that promotions can positively or negatively affect brand evaluations, depending on a) whether subjects had examined the brand previously (Scott and Yalch 1980), b) when they thought about their behavior (Scott and Tybout 1979), and c) whether they had previous knowledge about the brand (Tybout and Scott 1983).

In the one study that examined the effect of exposure to promotions on pre-purchase brand evaluations, rather than the effect of promotions after product trial, Davis, Inman and McAlister (1992) found no evidence that sales promotions affect brand evaluations. Davis et al. (1992) initially measured brand evaluations for selected brands in eight categories of frequently purchased products. Three brands in each of four product categories (canned pasta, pain relievers, toothpaste, and toothbrushes) were promoted in rotation for three months. None of the brands in the other four product categories (microwaveable popcorn, saline solution, cereal, and mouthwash) was promoted during this period and served as a control. After three months, one brand each from the promoted and control product categories was evaluated by another group of subjects who had observed the brands over the experimental period. Mean evaluations of promoted brands in the post-promotional period were not found to be lower than those in the pre-promotional period. Davis et al. (1992) conclude that their results provide "... evidence that promotion does not have a negative effect on brand evaluations" (p. 147).

One possible explanation for Davis et al.'s finding is that it was due to the kinds of products they studied. They looked at the effect of promotions on frequently purchased products. In the next section, we hypothesize that the effects of promotions on brand evaluations will depend on a) how common promoting is in the industry and b) the variability of brand quality in the industry. Frequently purchased packaged goods are, as a group, commonly promoted and do not have a high variation in quality among brands in each industry, which may explain why promotions in such categories do not lead to unfavorable brand evaluations. This paper examines variations in the perceived commonness of promotions and perceived variability of quality as they affect a promotion's influence on brand evaluations.

HYPOTHESES

It is theorized that the process of pre-trial attitude formation in response to a promotion is mediated by causal attributions (i.e., consumers assigning a cause) for the promotion. Attributions may be to the brand offering the promotion or to factors external to the brand, such as competition. The principles of attribution theory help predict the conditions under which the brand is implicated as the cause for a promotion. The attribution of a cause to "internal" (here, brand-specific factors) has been theorized to be a function of whether behavior conforms to what others do, traditionally termed the distinctiveness of the behavior (Einhorn and Hogarth 1986, Kelly 1967, 1972, Hilton and Slugoski 1986). In the promotions context, distinctiveness is determined by beliefs about the amount of promoting in the product category. The less common promoting is perceived to be in an industry, the more distinctive a promotion, and therefore, the greater the likelihood that a promotion offered by a specific brand in the industry will be ascribed to brand-related causes. Accordingly, we hypothesize that the locus of causality of a promotion is a function of the perceived amount of dealing within an industry:

H1: Causes of promotions are attributed to brand-related (vs. non-brand-related) factors more often when dealing is perceived to be uncommon in an industry than when it is perceived to be common.

An important question is whether a promotion is a positive or negative signal, across consumers, industries, brands, and types of promotions. There is far from unanimous agreement regarding the valence of a promotion. However, studies have either found negative or null effects of promotions on brand evaluations, rather than positive effects. The basis for theorizing that price promotions are negatively valenced derives from the price-quality literature. (For a review see Rao and Monroe 1989.) The monetary cost of a promoted brand is lower than it would be if the promotion had not been offered, because promotions reduce price or increase quantity for a given price. The price-quality literature has found that, on average, a lower price is a signal of inferior quality (Olson 1977, Monroe and Petroshius 1981, Rao and Monroe 1989). Since price promotions reduce price and lower prices are associated with lower quality, there is reason to believe that price promotions also lead to inferences of lower quality. Lichtenstein, Burton and O'Hara (1989) found that brand-specific attributions for a promotion were negatively valenced, whereas reasons unrelated to the brand were positively or neutrally valenced.

Because dealing is a negative signal, when a brand is distinctive for promoting in an industry in which promoting is uncommon, the brand-related attributions (hypothesized in H1 as resulting from distinctiveness) are negative and result in brand evaluations more negative than if promoting were common and non-brand-related attributions had been made instead. Thus, causal attributionsCwhether the cause of the promotion is attributed to brand-related factors or non-brand-related factorsCmediate the effect that the promotion has on brand evaluations.

H2a: Brands that promote less often than other brands in their industry are evaluated more favorably than the brands that promote more often than others.

H2b: The attributions for a promotion mediate the effect of promoting on brand evaluations.

There are situations in which the effect of promotions on brand evaluations is likely to be small or insignificant. If there is little perceived variation in quality among brands in an industry (e.g., airlines and many consumer packaged goods), consumers will have little need to use promotions as signals of quality. As discussed earlier, this may explain the null findings for effects of promotions in research that has studied frequently purchased consumer products, among which there may not be large perceived differences in quality.

H3: The lower the perceived variation in quality among firms in an industry, the smaller the effect of promoting on brand evaluations will be.

METHOD

Nine service industries were chosen to represent a wide range of service types. The industries chosen ranged from long distance calling to hair salons (see Table 1) and were expected to differ in terms of perceived commonness of dealing. For each industry, subjects read about a price promotion and answered the questions described below. The order in which the nine industries were presented was randomized for each subject.

Locus of Causality: Subjects were asked to rate what they thought was the single most important reason the promotion was being offered, on a 5-point scale anchored at 1 = "Due to competition," and 5 = "Due to the company."

Brand Evaluation: Subjects responded to two questions asking "If a company offers a deal MORE (LESS) OFTEN than other companies, the quality of the service is ..." The responses were provided on a 5-point scale, anchored by 1 = "Definitely better," and 5 = "Definitely Worse."

Subjects also rated how personally interested they were in each industry and how important the quality of the service was to them on 5-point scales. All but one industry was rated above the midpoint on both interest and importance of quality. The exception was hair stylists whose mean interest rating is 2.82.

After they provided the above ratings by industry, subjects were asked to estimate the overall percentage of companies in each industry that offer promotions and the percentage that are of good quality.

Subjects were 29 undergraduates at a large northeastern university, who completed the task during one of their regularly scheduled classes.

Results

Hypothesis 1. It was expected that promotions were more likely to be ascribed to external (vs. internal/brand-related) reasons, the more common subjects believed promoting was in the relevant industry. At the industry level this is the pattern observed. (See Table 1.) In all industries in which the mean locus of causality is less than 3, i.e., the promotion is rated as more likely to be due to competition than brand-specific factors (7 out of 9 cases), the overall estimated probability of dealing is greater than 40%. In the two industries in which the overall estimate of dealing is under 40% (dental services and music shows), the mean locus of causality of the promotion is greater than 3 (3.50 and 3.66, respectively).

To check whether there is a statistically significant relationship between estimated probability of dealing and locus of causality, we combined observations across industries and subjects and regressed the locus of causality (dependent variable) on the estimated dealing percent (independent variable). The results of this univariate regression are significant (R2=0.12, p<.001). The sign of the beta coefficient is negative, as hypothesized in H1, implying that the greater the estimated probability of dealing in the industry, the greater the attribution of the promotion to competitive factors.

Hypothesis 2a. H2a predicts that promoting more than others in an industry leads to less favorable evaluations. A MANOVA using ratings of the brand if it promoted more than (vs. less than) others in the industry was conducted for each of the service industries. We expected differences in ratings such that a brand would be rated worse if it promoted more often and better if it promoted less often than others in its industry. As can be seen from Table 1, the mean quality ratings of brands that promoted more often than others in their industry are consistently worse (lower numbers represent higher quality) and the differences are at least marginally significant in seven of the nine industries. Further, the absolute numbers indicate that the mean quality ratings for brands that promoted "more often" than others in their industries are above the mid-point (3) of the scale in all but one industry (health clubs). This means that on average subjects rated a brand as worse than others in its industry if it promoted more often than the other brands in the industry and rated it as better than the others if it promoted less often. This supports H2a.

Hypothesis 2b. H2b predicts that attributions mediate the effect of promoting on brand evaluations. To test this hypothesis, we treat each industry rating made by a subject as a separate observation and look only at the brand when it promoted more often than others in the industry. We ran a regression in which the dependent variable was the evaluation of the brand. The first independent variable is the promotion's locus of causality. The other is the estimated percent of firms in the industry that promote. Since we look only at brands that promote more than the others in the industry, this captures the distinctiveness of the promotionCthe higher the percent of firms promoting, the lower the distinctiveness. Because attributions were hypothesized to mediate the effect of promoting on evaluations, we expected that the coefficient of the attribution variable (locus of causality) would be significant and the coefficient of the estimated percent of firms which deal would be lower than in a regression in which it was the only independent variable. This is the pattern of results. The regression is significant (R2=.12, p<.001) and the effect of locus of causality is positive and significant (b=.24, p<.002), implying that when promotions were attributed to the competition, brand evaluations were more favorable. The coefficient for the estimated percent of firms that deal, is negative and significant (b=-.20, p<.01), indicating that the more distinctive the promotion, the more negative the evaluation. However, the estimate is lower than in the regression that had percent of firms that deal as the only independent variable (b=-.286, difference=.08, z=1.76, p<.05). As hypothesized, it appears that the attributions for the promotion mediate the effect of the promotion on brand evaluations.

TABLE 1

MEAN RATINGS OF CAUSALITY AND VALENCE OF PROMOTING

Hypothesis 3. H3 predicts that the lower the perceived variation in quality among firms in an industry, the smaller the effect of promoting on brand evaluations will be. Among the industries we examine there are none in which subjects felt quality is uniformly low. Thus, in our test of H3, the only kind of lack of variation in quality we examine is when quality is uniformly high. Our operational hypothesis is: brands that promote more often than other brands in their industry are evaluated better the larger the percentage of brands in the industry that are perceived as being of high quality. In other words, if quality in an industry is uniformly high, a brand that promotes more than others in the industry should be rated better than it would be were quality in the industry more variable. To test H3, we ran a univariate regression using only ratings of brands that promoted more often than others in their industries. The dependent variable is the brand evaluation and the independent variable is the estimated percentage of brands in the industry that are of good quality. The regression is significant (R2=.03, p<.05) and the parameter estimate is negative (lower numbers represent better quality). Consistent with H3, this indicates that the more good quality brands there are perceived to be in the industry, the better the brand that promotes more than others is rated.

CONCLUSIONS

In this study we examined the effects of promotions on brand-related attribution and on brand evaluations in nine service industries. We found evidence that a) the higher the perceived amount of dealing in an industry, the more likely it is that attributions for the promotion will be to the brand, and b) a brand that promotes more often than others in its industry is perceived to be of poorer quality than others in the industry, while a brand that promotes less often is perceived to be of better quality. This was true across a variety of industries. Further, the effect of promotions on brand evaluations is mediated by attributions made for the promotion. We also found that the effect of promoting on brand evaluations is diminished when a high percentage of brands in the industry are viewed as being of high quality. This may explain the results of prior studies in which promotions were not found to have an effect on brand evaluations.

One of the limitations of this study is that whether the experimental brands promoted more or less often than other brands was more obvious than it normally would be to consumers. Therefore, we cannot rule out a demand effect explanation for the result that promotions have a negative effect on brand evaluations. However, it is less plausible that subjects guessed the other hypotheses. Future research manipulating the information value of a promotional signal under various situations would give us a better understanding of when and how price promotions affect brand evaluations, and contribute further to the reconciliation of conventional wisdom with the results of academic research.

REFERENCES

Blattberg, Robert C., and Scott A. Neslin (1990), Sales Promotion: Concepts, Methods, and Strategies, Englewood Cliffs, NJ: Prentice Hall.

Chevalier, Michel (1975), "Increase in Sales Due to In-Store Display," Journal of Marketing Research, 12 (November), 426-31.

Davis, Scott, J., Jeffrey Inman and Leigh McAlister (1992), "Promotion Has a Negative Effect on Brand EvaluationsCOr Does It? Additional Disconfirming Evidence," Journal of Marketing Research, 29 (February), 143-8.

Dodson, Joe A., Alice M. Tybout and Brian Sternthal (1978), "Impact of Deals and Deal Retraction on Brand Switching," Journal of Marketing Research, 15 (February), 72-81.

Einhorn, Hillel J. and Robin M. Hogarth (1986), "Judging Probable Cause," Psychological Bulletin, 99 (1), 3-19.

Fiske, Susan T. (1980), "Attention and Weight in Person Perception: The Impact of Negative and Extreme Behavior", Journal of Personality and Social Psychology, 38, 889-906.

Gidron, David, Derek J. Koehler, and Amos Tversky (1993), "Implicit Quantification of Personality Traits," Personality and Social Psychology Bulletin, 19 (October), 594-604.

Hamilton, D.L. and M.P. Zanna (1972), "Differential Weighing of Favorable and Unfavorable Attributes in Impressions of Personality," Journal of Experimental Research in Personality, 6, 204-212.

Hilton, Denis J. and Ben R. Slugoski (1986), "Knowledge Based Causal Attribution: The Abnormal Conditions Focus Model," Psychological Review, 93 (1), 75-88.

Kelly, Harold H. (1967), "Attribution Theory in Social Psychology," in Nebraska Symposium on Motivation, Vol 15, ed. D. Levine, Lincoln, NE: University of Nebraska Press.

Kelly, Harold H. (1972), "Attribution in Social Interaction," in Attribution: Perceiving the Causes of Behavior, eds. E.E. Jones, D.E. Kanouse, H.H. Kelly, R.E. Nisbett, S. Valins, and B. Weiner, Morristown, NJ: General Learning Press.

Lichtenstein, Donald R., Scot Burton, and Bradley S. O'Hara (1989), "Marketplace Attributions and Consumer Evaluations of Discount Claims," Psychology and Marketing, 6, 163-180.

Monroe, Kent B., and Susan M. Petroshius (1981), "Buyer's Perception of Price: An Update of the Evidence," in Perspectives in Consumer Behavior, eds. Harold H. Kassarjian and Thomas S. Robertson, Glenview, IL: Scott, Foresman and Company, 43-55.

Neslin, Scott A. and Robert W. Shoemaker (1989), "An Alternative Explanation for Lower Repeat Rates After Promotional Purchases," Journal of Marketing Research, 26 (May), 205-13.

Ogilvy, David (1963), Confessions of an Advertising Man, New York: Atheneum.

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

Rao, Akshay R. and Kent B. Monroe (1989), "The Effect of Price, Brand Name, and Store Name on Buyer's Perceptions of Product Quality: An Integrative Review," Journal of Consumer Research, 26, 351-7.

Scott, Carol A. and Alice M. Tybout (1979), "Extending the Self-perception Explanation: The Effect of Cue Salience on Behavior," in Advances in Consumer Research, Vol. 6, ed. William L. Wilkie, Ann Arbor, MI: Association for Consumer Research, 50-54.

Scott, Carol A., and Richard F. Yalch (1980), "Consumer Response to Initial Product Trial: A Bayesian Analysis," Journal of Consumer Research, 7 (June), 32-41.

Taylor, Shelley (1991), "Asymmetrical Effects of Positive and Negative Events: The Mobilization-Minimization Hypothesis," Psychological Bulletin, 10 (July), 67-85.

Tybout, Alice M. and Carol A. Scott (1983), "Availability of Well-Defined Internal Knowledge and the Attitude Formation Process: Information Aggregation Versus Self-Perception," Journal of Personality and Social Psychology, 44 (3), 474-479.

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