The Effects of Featured Brand Quality on Price Valuations of the Product Portfolio

Barbara Bickart, Rutgers UniversityBCamden
Carolyn J. Simmons, Virginia Polytechnic and State University
Chirag Vyas, TNS Intersearch
ABSTRACT - We examine how the quality and price positioning of a featured brand in a product display affects valuations of other brands in the store’s portfolio. In an experiment, we find that both high-tier and unfamiliar portfolio brands benefit from the presence of a high-tier featured brand (versus a medium-tier featured brand). Further, store perceptions mediate the effect of the featured brand on valuations of the high-tier portfolio brand, but not on valuations of the unfamiliar brand. We discuss the implications of these results for understanding the impact of featured brands on product judgments and suggest directions for future research.
[ to cite ]:
Barbara Bickart, Carolyn J. Simmons, and Chirag Vyas (2003) ,"The Effects of Featured Brand Quality on Price Valuations of the Product Portfolio", in NA - Advances in Consumer Research Volume 30, eds. Punam Anand Keller and Dennis W. Rook, Valdosta, GA : Association for Consumer Research, Pages: 264-269.

Advances in Consumer Research Volume 30, 2003     Pages 264-269

THE EFFECTS OF FEATURED BRAND QUALITY ON PRICE VALUATIONS OF THE PRODUCT PORTFOLIO

Barbara Bickart, Rutgers UniversityBCamden

Carolyn J. Simmons, Virginia Polytechnic and State University

Chirag Vyas, TNS Intersearch

ABSTRACT -

We examine how the quality and price positioning of a featured brand in a product display affects valuations of other brands in the store’s portfolio. In an experiment, we find that both high-tier and unfamiliar portfolio brands benefit from the presence of a high-tier featured brand (versus a medium-tier featured brand). Further, store perceptions mediate the effect of the featured brand on valuations of the high-tier portfolio brand, but not on valuations of the unfamiliar brand. We discuss the implications of these results for understanding the impact of featured brands on product judgments and suggest directions for future research.

Product display plays an important role in maintaining brand equity and the overall equity of the retail brand portfolio. The goal of this paper is to examine how one aspect of display, brand precedence, affects evaluations of other brands in the portfolio. Specifically, we compare the effects of giving precedence o, or featuring, a high-tier brand (a well known high quality, high priced brand) versus a medium-tier brand (a well known medium quality, lower priced brand) on valuations of the other brands offered in the store and on perceptions of the store’s quality.

The arrangement of a product assortment is an important factor in shaping preferences and influencing the choice of specific brands (Simonson 1999). Previous research has demonstrated effects of display format on attention (Dreze, Hoch and Purk 1994; Janiszewski 1998), brand evaluations and valuations (Buchanan, Simmons and Bickart 1999; Simmons, Bickart and Buchanan 2000), and brand choice (e.g., Nowlis and Simonson 1997; Simonson, Nowlis, and Lemon 1993; Simonson and Winer 1992). This body of work confirms the importance of managing product display for brand identity, market share, and profitability.

The primary focus of research on display format has been on display structure (i.e., whether products are organized by brand or by quality level). In contrast, we examine the effects of giving a particular brand precedence in the display. Specifically, we examine how display precedence affects store perceptions as well as valuations of both well-known and unfamiliar brands. Display precedence may affect store perceptions because consumers believe that retailers give precedence to their most important brands (Buchanan et al. 1999). Thus, a featured brand may provide cues to consumers about the general quality or value of the other merchandise carried by the outlet.

Display precedence may affect evaluations of specific brands either directly or as mediated by store perceptions. Simmons et al. (2000) show that the presence of high equity brands in a retail department is associated with higher profitability for lower-equity brands. Specifically, the presence of high equity brands in the department results in fewer markdowns on lower-equity brands and therefore higher profit margins. We examine whether the same benefit can be attained by simply featuring a higher equity brand (holding the portfolio constant), as well as whether higher-equity portfolio brands receive the same benefits as lower-equity portfolio brands. Finally, we investigate whether these effects are mediated by store perceptions.

We first describe relevant literature and develop predictions. We then discuss the results of an experiment and explore some directions for future research.

THEORETICAL BACKGROUND

Display variables have been shown to affect consumer information processing in several ways. For example, consumers tend to use the search strategy that requires the least effort given the information display format (e.g., Bettman and Kakkar 1977). In addition, display structure affects the ease of comparing brands, with a mixed (or side-by-side) display facilitating brand comparisons (e.g., Buchanan et al. 1999, Nowlis and Simonson 1997). Nowlis and Simonson (1997) find that because of the increased ability to compare, mixed displays benefit lower quality, lower price products in terms of brand choice. Finally, display structure can heighten the salience of specific product attributes (e.g., Nowlis and Simonson 1997) or make accessible a standard of comparison for subsequent judgments (e.g., Schwarz and Bless 1992).

Like display structure, brand precedence in a display has also been shown to alter information processing. Buchanan et al. (1999) show that giving precedence to a brand creates expectations about the relationship between the featured brand and other brands in the retailer’s portfolio. When the portfolio brands violate those expectations, even high-equity brands may be negatively reevaluated. Areni, Duhan, and Keicker (1999) show that featuring a brand on the basis of a low-salience attribute (e.g., the region in which a wine is produced) can alter choice by encouraging brand comparisons on that low-salience attribute.

A common theme in these findings is that display variables make particular contextual information salient. In particular, featuring a brand is likely to make salient the characteristics of the featured brand (Areni et al. 1999). Hence, we expect that featuring a high-tier brand brings to mind quality and higher price associations, while featuring a medium-tier brand brings to mind value or low price associations. Because consumers are motivated to minimize judgment effort (Simon 1978), they may rely on these accessible associations when evaluating other brands in the context or the store itself. Whether a featured brand affects judgments should depend on two factors: (a) whether the consumer constructs a judgment or retrieves an existing judgment from memory, and (b) whether the consumer categorizes the entity to be judged with the featured brand.

For a judgment to be affected by contextually accessible associations, the judgment must be constructed on the spot. Hence, judgments of unfamiliar brands are likely to be affected by the featured brand. However, Buchanan et al. (1999) show that consumers also construct judgments for familiar brands when a display is inconsistent with expectations about the relationship between brands. Because consumers expect stores to feature their most important brands (i.e., higher tier brands), consumers may be especially likely to reconstruct judgments about familiar brands when a medium-tier brand is featured.

Similar considerations apply in determining whether a store evaluation is constructed on the spot. Judgments of unfamiliar stores should be constructed. However, we expect that evaluations of a familiar store will be constructed if the featured brand is inconsistent with what consumers expect from the store.

When a judgment of a portfolio brand or the store is constructed, the impact of the featured brand on this judgment should depend upon whether the entity to be judged is categorized with the featured brand. If it is categorized with the featured brand, assimilation to the featured brand should occur, whereas if it is categorized separately from the featured brand, contrast from the featured brand should occur (Schwarz and Bless 1992). Because they belong to the same retail portfolio, portfolio brands are likely to be categorized with the featured brand as long as they are not highly dissimilar from it. Hence, when portfolio brand judgments are constructed, both unfamiliar brandsCfor which little is knownCand familiar brands that are moderately similar to the featured brand are likely to be assimilated to the featured brand. Only brands that are known to be highly dissimilar to the featured brand are likely to be contrasted from it.

When a store evaluation is constructed, the store should be categorized with the featured brand for two reasons. First, store is superordinate to the brands it carries. Second, because consumers expect that retailers feature their most important brands (Buchanan et al. 1999), characteristics of the featured brand should be diagnostic for a store-level judgment.

Consider the implications of these arguments for the case at hand. We examine a situation in which the featured brand is either a high-tier brand or a medium-tier brand; the portfolio includes a high-tier brand, a medium-tier brand, and an unfamiliar brand; and the retail store is also unfamiliar. Our primary dependent variable is the price that consumers are willing to pay for each brand. Respondents are likely to construct these judgments for the unfamiliar portfolio brand because they have no previously formed judgment. They are also likely to construct these judgments for the high-tier and medium-tier portfolio brands when a medium-tier brand is featured because featuring a medium-tier brand over a high-tier brand violates their expectations. Further, these portfolio brands are expected to be categorized with the featured brand and therefore assimilated to it because they belong to the same retail portfolio and are not known to be highly dissimilar from the featured brands. Hence, we predict:

H1: Valuations of all portfolio brands will be higher when a high-tier brand is featured than when a medium-tier brand is featured.

Consumers are also likely to construct judgments about the store because it is unfamiliar. Further, they are likely to assimilate their store judgments to the featured brand because the store is superordinate to its brands and because they expect retailers to feature their most important brands. Therefore, we predict:

H2: Store perceptions will be more favorable when a high-tier brand is featured than when a medium-tier brand is featured.

Thus far, we have ignored the possibility that the effects of the featured brand on valuations of portfolio brands may be mediated by store perceptions, rather than reflecting a direct influence of the featured brand. Estimates of willingness to pay would seem to reflect more than simply the product itself. Consumers are willing to pay for ways in which a store adds value, be it better service, a nicer shopping ambience, or a lessened sense of risk that comes with the assurance that the retailer understands and appreciates quality. Therefore, we explore whether the effect of a featured brand on portfolio brand valuations is direct or indirectCthat is whether these judgments are mediated by store perceptions.

METHOD

Participants

Eighty students participated in this study in a classroom setting. Participation was voluntary. The participants were split fairly evenly by gender. Ages ranged between 18 and 46, with a median age of 24.

Procedure

Participants were told that an online retailer needed consumers’ perceptions of a new format for an online product display. The retailer’s name was not provided. Each participant was shown two pages, each with a "display" from an online store. One display contained four jean brands and the other display contained four watch brands. The displays included only a picture of the products and the brand names. Participants were asked to examine the displays carefully. Then they were asked to write down the price they were willing to pay for each of the displayed items. The next task asked for participants’ impressions of the online store. After completing these tasks, participants were given a separate questionnaire to complete, which included manipulation checks on the familiarity and quality of the featured brands.

Design and Materials

Participants were shown mock-ups of web pages for two product categories. Displays for each product category included four brands. Three of these brands were the portfolio brands and remained constant across conditions. One of the portfolio brands was a well-known high-tier brand, one was a well-known medium-tier brand, and one was a less familiar brand. The fourth brand was the featured brand, which was either a well-known high-tier brand or a well-known medium-tier brand.

The displays included photographs of the products, but no price information. The featured brand was shown on the top half of the page, accompanied by the label "Our Featured Brand." The portfolio brands were displayed in the lower half of the page, accompanied by the label "We also carry the following brands." The photograph of the featured brand was larger than the photographs of each portfolio brand. Each participant evaluated two layouts, one for each product category. We varied the order in which the product categories were evaluated as well as whether the high-tier or medium-tier brand was featured for the product category across four conditions:

1. Jeans 1st/High-Tier FeaturedBWatch 2nd/Medium-Tier Featured

2. Jeans 1st/Medium Tier Featured- Watch 2nd/High Tier Featured

3. Watch 1st/High Tier Featured- Jeans 2nd/Medium Tier Featured

4. Watch 1st/Medium Tier Featured- Jeans 2nd/High Tier Featured

The product categories and brand names used in the experiment were selected via two pretests. In the first pretest, participants were asked to rate their familiarity with brands in several product categories, including cameras, beer, watches, jeans, microwave ovens, and cordless phones. Based on this pretest, we selected jeans and watches as the product categories for the study. Most respondents were familiar with a number of brands in these categories and subjective knowledge was high and did not vary by gender.

The second pretest was used to select high-tier, medium-tier and unfamiliar brands within these two categories. Participants were asked to rate brands in each category on familiarity and quality. The high-tier and medium-tier brands were rated similarly on familiarity, but varied in quality ratings. The unfamiliar brand was rated lower than these brands on familiarity. Table 1 shows the brands used in the experiment.

Finally, we varied the order in which respondents provided a price for each brand. This procedure allows us to determine if participants use the price of the first brand as an anchor for determining the price of the other brands. Using a Latin Square design, there are four combinations within each display condition, as shown in Table 2.

TABLE 1

BRAND NAMES USED IN PRODUCT DISPLAY PAGES

TABLE 2

SEQUENCE MANIPULATION FOR EVALUATING BRAND PRICES

Dependent Measures

After viewing the displays, participants were asked what price they were willing to pay for each of the brands presented in the display (i.e., "What price would you be willing to pay for the Brand X shown in the display?"). As described above, we varied the order in which these prices were elicited. We then asked participants for their perceptions of the online store on three seven-point semantic differential scales (poor/superior quality store; ordinary/exceptional merchandise; better/worse than average store). The first two items are highly correlated (rjeans=.75; rwatches=.63). The third item, however, is not highly correlated with either of the other items (rs>.12 for both product categories). Therefore, we used the average of the first two items as the measure of store perceptions. We also measured product category knowledge and purchase frequency (both on seven-point scales). For both product categories, knowledge and purchase frequency did not vary across conditions (ps>41).

In the follow-up questionnaire (administered after the display pages were removed), participants were asked to select the featured brand in each product category from a list of brands. For jeans, correct recall of the featured brand is 65% for the high-tier brand and 58% for the medium-tier brand. For watches, correct recall of the featured brand is 65% for the high-tier brand and 53% for the low-tier brand. Recall does not vary by type of brand (p’s>.25). Participants were then asked to rate their familiarity with a number of brands in each product category, as well as the quality of these brands, both on seven-point scales (very/not familiar; very high/very low quality). Finally, using seven-point scales, we obtained measures of interest (M=3.34, SD=1.49) and involvement with the study (M=3.69, SD=1.56), sex, age, and the perceived purpose of the study.

ANALYSIS AND RESULTS

Manipulation Checks

Two assumptions are important to our interpretation of these results. First, the featured high-tier and medium-tier brands should vary on quality, but not on familiarity. Second, the unfamiliar portfolio brand should be low in familiarity and less familiar than the high-tier and medium-tier portfolio brands. As shown in Table 3, familiarity and quality ratings obtained in the follow-up questionnaire are consistent with these assumptions. The featured high-tier and medium-tier brands do not differ on familiarity (F1,78<1 for both jeans and watches), while the featured high-tier brands are seen as better quality than the featured medium-tier brands (F1,75=33.95, p<.0001 for jeans and F1,76=161.70, p<.0001 for watches). Further, the high-tier portfolio brands are seen as better quality than the medium-tier portfolio brands (F1,76=72.76, p<.0001 for jeans and F1,76=26.50, p<.0001 for watches). Finally, the unfamiliar portfolio brand is low in familiarity and less familiar than either the high-tier portfolio brand (F1,77=206.42, p<.0001 for jeans and F1,78=44.88, p<.0001 for watches) or the medium-tier portfolio brand (F1,76=125.43, p<.0001 for jeans and F1,78=77.25, p<.0001 for watches). [There are two unexpected outcomes in Table 3. First, participants are less familiar with Cartier (the high tier portfolio watch brand) than Rolex (the high tier featured watch brand). Cartier, however, is rated more highly on familiarity than Tissot (the unfamiliar portfolio watch brand), which is the key comparison. In addition, while participants were not familiar with Tissot, they did rate it high in terms of quality. There is no reason to believe, however, that high quality perceptions associated with the Tissot name would affect the interpretation of the results.]

TABLE 3

MANIPULATION CHECKS ON BRAND FAMILIARITY AND QUALITY

TABLE 4

MEAN WILLINGNESS TO PAY FOR PORTFOLIO BRANDS BY QUALITY OF FEATURED PRODUCT

Valuations of Portfolio Brands

The key dependent measure is the price participants are willing to pay for each of the portfolio brands. We expected that valuations of the portfolio brands would be higher when a high-tier brand was featured. The results in both product categories are consistent with these predictions.

We first determine whether our within-subjects manipulations resulted in differential transfer (position x treatment interactions) that would contaminate the effects of interest. Recall that the order in which brand valuations were measured is varied within subjects, while the order of categories and the order of featured brands is partially varied within subjects; that is, each subject saw two of the four category x featured brand conditions. This design offers efficiencies in data collection and increased power for statistical tests, while avoiding the sensitization to our manipulations that would occur if subjects each saw all levels of our independent variables. However, a tradeoff in this design is that we cannot separate one component of differential transfer (the order of category x order of featured brand interaction) from the category x featured brand interaction. Because we expect the category x featured brand interaction to be significant (the price range for watches is much greater than the price range for jeans), we rely on our other tests of differential transfer (order of categories, order of featured brands, and measurement order), and assume that variance in the order of category x order of featured brands interaction is attributable to the category x featured brand interaction.

We find no evidence of differential transfer for high-tier portfolio brands (ps>.12) or for unfamiliar portfolio brands (ps>.13). However, we find evidence of differential transfer for the medium-tier portfolio brands, attributable to both order of category presentation (F1,63=3.05, p<.08) and measurement order (F3,63=2.27, p<.08). (We set a at .10 for these tests because we wished to conclude a null effect.) When there is differential transfer, the normal procedure is to discard data from all conditions after the first. Because our sample size does not allow adequate power for statistical tests if we follow this procedure, we eliminate the medium-tier portfolio brand from the analyses that follow. However, means for all portfolio brands are reported in Table 4.

As predicted by H1, valuations of both high-tier and unfamiliar portfolio brands are greater when the featured brand is a high-tier brand vs. a medium-tier brand (Ms=339.62 and 104.86, F1,59=5.08, p<.02 for high-tier portfolio brands and Ms=98.57 and 44.07, F1,54=6.99, p<.01 for unfamiliar portfolio brands). This effect is qualified by a category x featured brand interaction (F1,64=5.03, p<.02 for high-tier portfolio brands and F1,61=7.25, p<.009 for unfamiliar portfolio brands). Simple effects tests show that the effect is greater for watches than for jeans. Specifically, for high-tier portfolio brands, the effect is marginally significant for jeans (F1,70=2.81, p<.09) and highly significant for watches (F1,69=5.51, p<.02), while for unfamiliar portfolio brands, the effect is significant for both categories, but larger for watches (F1,67=4.98, p<.02 for jeans and F1,64=7.38, p<.008). As can be seen in Table 4, a similar pattern is apparent for medium-tier portfolio brands.

Also as expected, there is main effect of category with valuations being lower for jeans than for watches (Ms=39.75 and 404.71, F1,59=12.27, p<.0009 for high-tier portfolio brands and Ms=19.49 and 123.15, F1,54=25.30, p<.0001 for unfamiliar portfolio brands), and a similar pattern is apparent for medium-tier brands.

Store Perceptions

As predicted by H2, perceptions of store quality are higher when a high-tier brand vs. a medium-tier brand is featured (F1,73=12.12, p<.0008) and this effect does not vary across categories (F<1; Ms=4.00 and 3.57 for jeans and 4.32 and 3.80 for watches). Because the overall store portfolios are not the same in the high-tier-featured versus low-tier-featured conditions, the effect cannot be entirely attributed to giving the featured brand precedence; that is, effects could be due partly or in total to presence of the featured brand, independent of how it is displayed.

We next examined whether the effects of the featured brand on brand valuations were mediated by store perceptions. Our analysis suggests that store perceptions mediate the effect of the featured brand on ratings of high-tier portfolio brands, but not on ratings of unfamiliar brands. According to Baron and Kenny (1986), three pieces of evidence are required to establish mediation: (a) the independent variable must significantly affect the proposed mediator, (b) the proposed mediator must significantly affect the dependent variable, and (c) when both the independent variable and the proposed mediator are regressed on the dependent variable, the proposed mediator must remain significant while the independent variable becomes nonsignificant (for complete mediation) or is reduced in effect size (for partial mediation).We have already shown that the independent variable, featured brand, significantly affects the proposed mediator, store perceptions. Further, the proposed mediator, store perceptions, also significantly affects the dependent variable, ratings of portfolio brands (F1,150=11.19, p<.001 and F1,143=9.76, p<.002 for the high-tier and unfamiliar portfolio brands, respectively). Finally, when both featured brand and store perception are regressed on ratings of portfolio brands, store perceptions remain significant for both high-tier and unfamiliar portfolio brands (F1,149=8.49, p<.004 and F1,142=6.73, p<.01), while the effect of the featured brand becomes nonsignificant for high-tier portfolio brands (F1,149=2.05, p>.15), but remains marginally significant for unfamiliar portfolio brands (F1,142=3.47, p<.06).

GENERAL DISCUSSION

This experiment is the first step in showing how a featured product affects valuations of other products in the portfolio. We show that when a well-known brand is featured, the price and quality positioning of the brand is used as a cue to evaluate other products in the portfolio. Featuring brands of higher quality and price (high-tier brands) increased the willingness to pay for both high-tier and unfamiliar brands in the portfolio. In this study, participants were willing to pay between 15% and 257% more for brands when a high-tier brand was featured. These findings were robust across two product categories. While the results were less consistent for medium-tier portfolio brands, the findings were in a similar direction. For the medium-tier brands we observed differential transfer. Valuations of medium-tier brands may be most susceptible to judgment order because of a less clear natural anchor for making such judgments. In contrast, it is possible that store perception anchors the judgments of unfamiliar brands, while existing brand perceptions anchor perceptions of the higher-tier brands.

The findings replicate and extend earlier research by Buchanan et al. (1999) and Simmons et al. (2000), who provided evidence that featuring a well known brand can result in higher valuations of another brand in the portfolio. In their research, the featured brand is given precedence by displaying it first in a layout and in a headline. We show similar effects when a brand is given precedence via a larger display. In addition, we extend their findings to a situation where more than two brands are included in the product portfolio.

Our findings also suggest that at least for high-tier brands, the effect of the featured brand is mediated by perceptions of store quality. In other words, featuring a high-tier brand has a positive effect on perceptions of overall store quality, which increases willingness to pay for the high-tier brand. The featured brand appears to have a direct effect on valuations of the unfamiliar brand. Our examination of the mediating role of store perceptions was exploratory in nature, but suggests that the way in which display precedence operates on value judgments may depend in part on brand familiarity and whether the consumer is constructing a judgment for the first time or updating an existing judgment.

In this experiment, we examined the effects of featured brands in a relatively impoverished information environment. Respondents saw only a photo and a brand name. An important issue for future research is how additional product information may moderate these effects. When more information is provided, there may be situations in which featuring a high-tier brand (versus a medium-tier brand) results in lower valuations of the portfolio brands. Providing additional information could affect perceptions of similarity, depending on the level of detail and concreteness of the information. Thus, portfolio brands could be categorized separately from the featured brand, resulting in contrast effects. Future research could also examine the impact of different forms of display. Featuring brands via certain types of displays (i.e., end-of-aisle displays, versus unique signage) may hold specific meanings for consumers, resulting in different effects on valuations of portfolio brands. In addition, we did not provide the name of the retail outlet, which is likely equivalent to providing an unfamiliar name. While this scenario does not seen unrealistic in the current online environment, future research should address the role of existing store image and whether it enhances, diminishes or changes the nature of the featured brand effects observed here. Additional information about either the products or the retailer may lead to expectations about the outlet and the types of products that might be featured in the outlet. Consumers’ expectations about brand relationships (or strategic equivalence) may affect the likelihood that a featured brand serves as a standard of comparison against which other brands are evaluated (Simmons et al., 2000).

Another possibility is that consumers’ search goals (exploratory versus goal-directed) may moderate the effects of display precedence on brand valuations. Search goals moderate the effects of display characteristics on attention (Janiszewski 1998) and thus may also affect perceptual processes such as those reported in this study. In particular, goal-directed search may increase consumers’ likelihood of making comparisons on salient attributes, resulting in contrast effects on valuations.

We focused on the effects of a featured brand on brand valuations. The effects on perceptions of other product attributes or brand choice may differ, however. For example, featuring a high-tier brand may increase the favorability of store perceptions and brand valuations, but reduce the likelihood of choosing high-margin brands in the portfolio. In the long run, managers want to build equity in a store’s image or perceived quality, while in the short run, they hope to increase sales and profitability (via consumer choice of high margin brands). Future research should address how display structure affects these sometimes competing goals, particularly since the effects of context are not always the same on choice and purchase likelihood measures (Nowlis and Simonson 1997).

Finally, an issue of growing importance to retailers is how to build equity in the store by featuring private label brands. For example, Macy’s is prominently featuring its own I.N.C. brand in hopes of building a unique store identity (White 2002). Future research may provide guidance to retailers in using display structure to create more value for their own products and for the outlet itself.

REFERENCES

Areni, Charles S., Dale F. Duhan, and Pamela Kiecker (1999), "Point-of-Purchase Displays, Product Organization, and Brand Purchase Likelihoods," Journal of the Academy of Marketing Science, 27, (Fall), 428-441.

Baron, Reuben M. and David A. Kenny (1986), "The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations," Journal of Personality and Social Psychology, 51 (December), 1173-1182.

Bettman, James R. and Pradeep Kakkar (1977), "Effects of Information Presentation Format on Consumer Information Acquisition Strategies," Journal of Consumer Research, 3 (March), 233-240.

Buchanan, Lauranne, Carolyn J. Simmons and Barbara A. Bickart (1999), "Brand Equity Dilution: Retailer Display and Context Brand Effects," Journal of Marketing Research, 36 (August), 345-355.

Dreze, Xavier, Stephen J. Hoch and Mary E. Purk (1994), "Shelf Management and Space Elasticity," Journal of Retailing, 70 (Winter), 301-327.

Janiszewski, Chris (1998), "The Influence of Display Characteristics on Visual Exploratory Search Behavior," Journal of Consumer Research, 25 (December), 290-301.

Nowlis, Stephen M. and Itamar Simonson (1997), "Attribute-Task Compatibility as a Determinant of Consumer Preference Reversals," Journal of Marketing Research, 34 (May), 205-218.

Schwarz, Norbert and Herbert Bless (1992), "Constructing Reality and Its Alternatives: An Inclusion/Exclusion Model of Assimilation and Contrast Effects in Social Judgment," in Leonard L. Martin and Abraham Tesser (eds.), The Construction of Social Judgments. Hillsdale, NJ: Lawrence Erlbaum Associates, pp.217-248.

Simmons, Carolyn J., Barbara A. Bickart and Lauranne Buchanan (2000) "Leveraging Equity Across the Brand Portfolio," Marketing Letters, 11 (3), 210-220.

Simon, Herbert A. (1978), "Rationality as a Process and Product of Thought," American Economic Review, 68, 1-16.

Simonson, Itamar (1999), "The Effect of Product Assortment on Buyer Preferences," Journal of Retailing, 75 (3), 347-370.

Simonson, Itamar, Stephen Nowlis, and Katherine Lemon (1993), "The Effect of Local Consideration Sets on Global Choice Between Lower Price and Higher Quality," Marketing Science, 12 (Fall), 357-377.

Simonson, Itamar and Russell S. Winer (1992), "The Influence of Purchase Quantity and Display Format on Consumer Preference for Variety," Journal of Consumer Research, 19 (June), 133-138.

White, Erin (2002), "Federated Gives I.N.C. More Ink," The Wall Street Journal, March 12th, B1.

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