The Effect of Negative Information on the Evaluations of Brand Extensions and the Family Brand

Jean B. Romeo, Boston College
ABSTRACT - While the number of brand extensions has proliferated, there is little theory or methodology to help managers understand how extensions may affect a family brand name. The purpose of this research is to explore the effect that negative information about extensions may have on evaluations of extensions and the family brand's image. The study manipulated an extension's product category and attribute similarity with the family brand. The results indicated that when extensions are in the same product category as the family brand, negative information is most detrimental to extension evaluations and evaluations of the family brand image. Thus, while introducing extensions that are closely related to the family brand does increase the probability of consumer acceptance, managers should realize that this strategy can be detrimental to the family brand if the new extension is the target of negative information.
[ to cite ]:
Jean B. Romeo (1991) ,"The Effect of Negative Information on the Evaluations of Brand Extensions and the Family Brand", in NA - Advances in Consumer Research Volume 18, eds. Rebecca H. Holman and Michael R. Solomon, Provo, UT : Association for Consumer Research, Pages: 399-406.

Advances in Consumer Research Volume 18, 1991      Pages 399-406

THE EFFECT OF NEGATIVE INFORMATION ON THE EVALUATIONS OF BRAND EXTENSIONS AND THE FAMILY BRAND

Jean B. Romeo, Boston College

ABSTRACT -

While the number of brand extensions has proliferated, there is little theory or methodology to help managers understand how extensions may affect a family brand name. The purpose of this research is to explore the effect that negative information about extensions may have on evaluations of extensions and the family brand's image. The study manipulated an extension's product category and attribute similarity with the family brand. The results indicated that when extensions are in the same product category as the family brand, negative information is most detrimental to extension evaluations and evaluations of the family brand image. Thus, while introducing extensions that are closely related to the family brand does increase the probability of consumer acceptance, managers should realize that this strategy can be detrimental to the family brand if the new extension is the target of negative information.

INTRODUCTION

While the number of brand extensions in the marketplace has proliferated, managers have little insight into whether extensions affect consumers' perceptions of the family brand name. Yet, the potential influence that an extension may have on a brand name is most important because if it is negative, it could damage the family brand's image. For instance, if a known brand name gets attached to a product failure, the possibility exists for negative "ruboff" on the parent brand (Fannin 1987). The effect that brand extensions may have on the brand's image becomes even more critical when one considers that corporate licensing (one corporation linking up with another's successful brand or trademark to market new products that the trademark owning company does not produce) has become a $14 billion annual business (Norris 1987). Companies that license their brands in order to increase revenues should understand when this strategy could affect their brand. The purpose of this study is to provide some insight into how negative product information about a brand extension may affect evaluations of the extension as well as the family brand. The following questions are investigated:

- How does negative information about an extension affect the family brand's image?

- Will this effect depend on how closely associated the extension is to the family brand?

- What factors influence consumers' perceptions of how closely associated the extension is to the family brand?

BACKGROUND

Currently, there is little theory or methodology to guide managers in understanding the effect that an extension can have on the brand name (Jolley and Hawkins 1988). However, one can find opposing views in the literature on how extensions may affect a brand's image. One view is that line extensions are detrimental to a family brand's image since they blur the sharp focus of the brand in consumers' minds -- the more products positioned under a brand name, the less meaning the brand name has (Ries and Trout 1986). For instance, you cannot say Dial if you want soap (Dial makes deodorant, too, and Scott if you want paper towels (Scott also makes napkins and toilet tissue). Ries and Trout -feel that a well-known name gets well known because it stands for something specific -the more extension under one brand name, the weaker that brand name becomes.

Some managers feel that many extensions surrounding a brand will enhance its image. The Sunkist brand is licensed to a variety of products because management believes that the diverse range of extensions (e.g., Sunkist Vitamin C, Sunkist Orange Soda) increases the frequency of the Sunkist message to the consumer and strengthens the desired association with good health and vitality (Kesler 1987). Park, Jaworski, and MacInnis (1986) present a framework for managing a brand's image over time which they call Brand Concept-Image Management (BCM). In this framework, the final stage (after introduction and elaboration) of brand image management is the fortification stage where the brand image is strengthened by linking it to products produced by the firm in different product classes. Their theory is that many products with similar images reinforce one another and strengthen the image of the brand. They provide the example of how the Vaseline brand was fortified by extending the brand to beauty care products (Vaseline Intensive Care Bath Beads) and baby products (Vaseline Wipe N Dipes, Vaseline Baby Powder).

While the long-term effect that an extension may have on a family brand remains unclear, it seems likely that negative information about an extension would be detrimental to the family brand. This is evident from the broad support throughout the behavioral and marketing literature for the potency of negative information. For instance, research has found that negative adjectives are more powerful than positive adjectives (Anderson 1965) and the weights given to negative adjectives have exceeded the weights given to positive adjectives when several must be combined into one overall evaluation (Feldman 1966, Richey, McClelland, and Shimkunas 1967). In a marketing context, unfavorable product ratings tended to have a greater impact on attitudes and purchase intentions than favorable ratings (Weinberger and Dillon 1980).

One explanation for the increased influence of negative information is that negative information stands out more than positive information because there are more positive cues in the environment. Thus, negative cues, which are more infrequent, attract more attention (Kanouse and Hanson 1971). Fiske (1980) hypothesized that cues which deviated from the moderate positive norm should be more informative. She found that subjects paid more attention and gave more preferential weighing to negative and extreme cues in forming likability ratings of people. Thus, cues that are more rare may be more informative because they discriminate among similar objects.

In order to provide insight into the effect that negative information about an extension may have on a family brand, recent brand extension research is summarized. The idea of a family brand schema is then introduced to provide a framework from which the hypotheses are derived.

Brand Extension Research

Brand extensions became the guiding strategies of product planners in the 1980's (Tauber 1988). One reason for the proliferation of brand extensions is that they promote marketing efficiencies. Promotion expenses and risk associated with a new product introduction are lowered because both consumer and retailer acceptance is greater for a new product with an existing brand name than with a new name. People may be more apt to buy a new product with an existing brand name because the known brand name provides the assurance that the new product is of the same quality as the other product(s) with the brand name. Thus, consumers can relate the new product to a product with which they are already familiar. Categorization research implies that "When a consumer has little experience with a product, being able to categorize it with products that are familiar may permit a set of important inferences to be made" (Cohen and Basu 1987, p. 470). This relates to brand extensions since, "New products may be perceived as members of an existing brand 'family' (category) simply by virtue of having the brand name" (University of Minnesota Consumer Behavior Seminar 1987, p. 228).

Recent brand extension research has focused on how consumers perceive and evaluate extensions. These studies have found that the greater the similarity between the extension and the family brand, the greater the transfer of positive attitudes from the brand to the extension. One study found the transfer of affect from the brand to the new extension depends on the similarity between the new extension and the original branded product (UMCBS 1987). In fact, these researchers found that a brand's reputation for excellence in one product category may have a negative effect on consumer ratings of new products in an unrelated product area. They concluded that a consumer may reason that a brand's specialization in one product area may prevent it from being associated with a good product in an unrelated product category.

Aaker and Keller (1990) investigated how consumer knowledge about a brand may affect perceptions about the extension's quality. They found that a positive quality image for the brand influenced perceptions of the extension only when there was a basis of fit between the two products. In this study, the "basis of fit" between a brand and an extension was measured using the following dimensions: 1) extent to which consumers viewed the two product classes as complementary, 2) extent to which consumers viewed the two product classes as substitutes, and 3) the perceived credibility of firms operating in the original product category to make a product in the extension's product category.

Family Brand Schema

The brand extension research described in the preceding section indicates that consumers may have a schema for a family brand. A schema is a cognitive structure that represents organized knowledge about a certain concept (Fiske and Taylor 1984, p. 140). -A schema is developed through our experiences -- a collection of individual components becomes an integrated organization unit with strong associations among the once individuated components (Fiske and Dyer 1985). Once a schema is activated, it functions as an organized whole and serves as a perspective for attending to and interpreting events. Thus, when consumers become familiar with a brand name, they may form a type of schema representing the category of branded products. For instance, one could have a schema for "Ivory" branded products which might include features such as pure, wholesome, white, smooth, and gentle. Since schemata focus on how we assimilate new information with existing knowledge (Fiske and Taylor 1984), we would know what information is congruent to our schema (e.g., "new Ivory Shampoo is gentle enough for children") and incongruent (e.g., "new Ivory Detergent is tough on dirt").

A family brand schema may enable consumers to determine which extensions are similar to the brand and which are not. For instance, the brand Country Time Lemonade was so strongly associated with lemonade, that consumers couldn't accept Country Time Apple Cider. The set of associations consumers had for the Campbell's brand resulted in the belief that Campbell's Spaghetti Sauce would be orangy, runny, and not authentic Italian. This evaluation was based on consumers' schema for the Campbell's brand which was strongly associated with its tomato soup. (The company therefore introduced the sauce under a new brand, Prego).

HYPOTHESES

Positive brand belie& will transfer to positive extension beliefs when the extension is perceived as similar to the family brand (Consumer Behavior Seminar 1988) and when there is a basis of fit between the brand and the extension (Aaker and Keller 1990). How will this transfer of positive brand attitudes be affected when the extension is clearly inferior to the family brand?

Research indicates that people are very attentive to information which is inconsistent with their schemas. For instance, people take longer to interpret and evaluate inconsistent information. In addition, inconsistency which is unambiguous, strong, or evaluative, is even more likely to capture attention and be remembered. The underlying principle is that people think about and remember the exceptions (Fiske and Taylor 1984, Chapter 5).

If the set of associations subjects have about a family brand name is positive, negative information about extensions that are perceived as similar to the brand is hypothesized to be distinctive since it is inconsistent with consumers' schema. Thus, subjects should pay attention to and remember this information which is strong, unambiguous, and evaluative. Negative information about products that are perceived as dissimilar to the family brand should be less distinctive since these products are not strongly associated with the family brand schema. Therefore, it is expected that negative information about an extension which is similar to the family brand will be more noticeable than negative information about an extension which is not similar to the family brand.

H1: Negative information should be more influential on evaluations of extensions that are closely related to the family brand (high perceived similarity) than on extensions that are not closely related (low perceived similarity).

Research has found that a positively evaluated brand will lead to positive extension evaluations when the extensions are similar to the original family brand. However, what influence will negative associations with extensions have on the brand's image? Since the transfer of affect from a brand to an extension (Family Brand --->Extension) is mediated by the ''fit'' between the brand and the extension, it is hypothesized that the transfer of affect from the extension to the brand (Extension --->Family Brand) is also be mediated by "fit."

H2: Negative information about extensions closely related to the family brand will have more of an effect on the family brand image than information about extensions that are not closely related to the family brand.

In this study, an extension that is considered to be closely related to the family brand (high perceived similarity) is in a similar product category and has similar attributes compared to the family brand. An extension that is not closely related (low perceived similarity) is in a dissimilar product category and has dissimilar attributes compared to the family brand.

EXPERIMENT

For this study, a decision had to be made whether to use a known brand or a hypothetical brand. Since subjects' prior experience and knowledge of the brand could affect evaluations, one brand extension study (UMCBS 1987) used a hypothetical brand in order to control for a priori information or impressions that are associated with a known brand name. However, using a hypothetical brand has a disadvantage in that it is unknown to consumers. When a company pursues a brand extension strategy, the brand is one which is familiar to consumers, thus consumers have most likely formed some sort-of schema about that brand.

Following guidelines provided by Aaker and Keller (1990) in their study of consumer response to brand extensions, a known brand name should: 1) have a favorable overall quality image, 2) elicit relatively specific associations, and 3) not have already been broadly extended. For this study, a family brand name that was familiar to subjects and did not elicit a strong negative affective response was used. The extensions, however, were new products for the brand name. The family brand selected for this study was Tropicana. Tropicana was chosen since its association with orange juice would be equally relevant to men and women, and at the time of the study, it was not broadly extended. In order to check that Tropicana would elicit an overall positive affect, twenty six undergraduate business students (who did not participate in the main experiment) rated their attitude toward the brand on a series of semantic differential scales. The results indicated that subjects had a positive attitude towards the Tropicana brand. (Ratings on the 7-point evaluation scale where a rating of 1 was associated with a low evaluation and a rating of 7 was associated with a high evaluation were: Good=6.7, Pleasant=6.4, Superior=6.2, Interesting=4.8, Nice=5.9, Important=5.4).

The extensions' similarity to the brand name was manipulated in terms of their product category and attribute similarity to the family brand. Product category similarity was varied at two levels by selecting extensions in a similar product category (juice) to the family brand and in a different product category (sherbet). Attribute similarity was also varied at two levels by having extensions with many similar attributes (citrus-related) as the family brand and extensions with many different attributes (raspberry). Thus, four extensions under one family brand name were investigated (see Figure for extension manipulations). While similarity in terms of product category has been investigated in brand extension studies, similarity in terms of attributes has not. Research in categorization has found that if an object's attributes are consistent with the schema it evokes (e.g., category of branded products), category-based processing will result; if the object's attributes are perceived as inconsistent with the category, piecemeal processing will result (Fiske and Pavelchak 1986, Sujan 1985). For instance, the failure of Country Time Apple Cider, which is in a similar product category to the family brand (juice), yet has different attributes (apple cider instead of lemonade) indicates that attribute similarity as well as product category similarity may determine the basis of fit between an extension and a family brand. Thus, the purpose of manipulating the extensions' attributes as well as product category was to provide insight into the factors that may affect similarity perceptions.

FIGURE

TROPICANA EXTENSION MANIPULATIONS

A pre-test was conducted to check that the extension in a similar product category and with similar attributes was perceived as closely associated to the family brand compared to the extension with dissimilar product category and dissimilar attributes. This check was a modification of Rosch's (1975) rating tasks. Twenty subjects who did not participate in the main experiment were asked to rate how well each of ten products fit their image of a brand name. They were instructed that a 1 means that the product is a good example of their image of the brand name and a 7 means that the product fits poorly with their image of the brand name. The mean ratings indicated that the citrus juice extension was closely associated with the brand name (mean=1.65) and the raspberry sherbet was not closely associated with the brand (mean=5.75) As expected, the two remaining extensions were rated in between with citrus sherbet (similar attributes/dissimilar product category) receiving a rating of 4.30 and raspberry juice (dissimilar attributes/similar product category) receiving a rating of 4.80.

Experimental Procedure

Subjects were undergraduate business students recruited from a large northeastern university. The data was collected in two sessions. The first session w&s held during class time of four undergraduate marketing courses. The purpose of this session was to collect subjects' image of the Tropicana brand.

Brand image was assessed by having subjects respond to a series of seven-point semantic differential scales. The bipolar adjective pairs used to measure brand image were developed by integrating some of the bipolar adjective pairs used by Jacoby and Mazursky (1984) into the framework suggested by Osgood, Suci, and Tannenbaum (1957).

Three weeks later, students in the same marketing classes were told that they could receive extra credit points by participating in a research study. They were given the opportunity to sign-up for one of several sessions outside of class time. At the sessions, students read a case study that they were told was currently being developed for a textbook of case studies. Case study stimuli were used because they provided a realistic format to experimentally manipulate negative information about the brand extensions. The cases consisted of information about a new Tropicana product that had been introduced but where sales were disappointing. They were also shown an evaluation of the product similar to those in Consumer Reports magazine. Six product characteristics were listed in the first column, followed by five columns of evaluations ranging from excellent to poor. The product characteristics were determined from actual Consumer Reports evaluations of juice and sherbet products. The characteristics used were: pleasantly sweet, agreeable aftertaste, flavor of fresh fruit, natural taste, freshness of flavor, and natural coloring. Evaluations were all in the fair and poor categories. In addition, under the evaluation grid there was a comment section to reinforce the negative evaluation.

At each session, one of the case studies was distributed (case studies were randomly assigned to sessions). After reading the case, students answered questions about the case, evaluated the Tropicana extension on a series of seven-point scales and evaluated the Tropicana brand image (on the same set of scales used in the first study). Subjects also responded to three-item involvement scales to assess their involvement with orange juice and the extension's product category (either fruit juice or sherbet). The specific items used were product category interest, time spent thinking about the product category, and average importance of the product category (adapted from Bloch, Sherrel, and Ridgway 1986). In addition, subjects were asked if they ever had Tropicana Orange Juice or any other Tropicana products and whether they had seen any advertisements for Tropicana products.

RESULTS

A total of 80 subjects completed the initial survey and the follow-up survey three weeks later (treatment sizes ranged from 19 to 21). Almost all of the subjects (98%) indicated that they have had Tropicana Orange Juice (this confirms the pre-test which found students to be familiar with the Tropicana brand).

TABLE 1

POST-EXPOSURE EVALUATIONS OF EXTENSIONS

The family brand image measures taken at time 1 and time 2 were factor analyzed. As expected, the brand image measures loaded on three dimensions -- evaluation, potency, and activity. For the purpose of this study, only the evaluation dimension was of interest in order to investigate the effect of negative information on the evaluation of the family brand image. Family brand image was operationalized as the summed response across eight semantic differential scales that loaded on the evaluation factor: quality, taste, superior, fresh, natural, unique, high class (Chronbach's alpha = .9073 for time 1 and .8961 for time 2). Subjects who evaluated the Tropicana image as either neutral or negative were eliminated from the analysis (4% of respondents fell into this category).

In order to check that the four extensions were evaluatively equivalent, some subjects received case studies with positive information about the extension. A one-way ANOVA was performed across the four extensions using the overall extension evaluation as the dependent measure. The results indicated that there was not a significant difference in evaluations of the four extensions when subjects were not presented with negative information (F(3,75) = .7999, P< 5)

The first hypothesis concerns the expected effect that negative information may have on subjects' ratings of the extensions. It was expected that negative information about an extension closely associated with the brand would be most distinctive since it was inconsistent with the positive brand schema. Thus, the similar brand extension (similar product category and attributes) should be rated lower than the dissimilar extension (dissimilar product category and dissimilar attributes).

A 2 X 2 analysis of covariance was performed to investigate this hypothesis. The independent variables were product category (similar/dissimilar) and attribute (similar/dissimilar). The covariate was subjects' involvement with the product category (either juice or sherbet) and the independent variable was the summated product rating score (the rating scales were combined into an overall extension evaluation; Chronbach's alpha =.9171). The results indicated that the covariate was significant (F(1,73 = 7.525, p<.01).

Product category was the only significant main effect (F (1,73) = 10.654, p<.003). For both low and high involvement subjects, extensions that were similar to the brand in terms of product category (see Table 1) received lower evaluations than extensions that were dissimilar to the brand.

Subjects' schema for Tropicana was strongly associated with juice (similar product category), so that negative information about Tropicana juice products had more impact on evaluations than negative information about Tropicana sherbet products. Thus, subjects paid more attention to this information because it was distinctive, and lower extension evaluations resulted. There was no difference in extension evaluations across the attribute manipulations, thus it seems that product category was more important than attributes when evaluating the extension. Hypothesis 1 was supported when similarity is defined in terms of product category only.

The second hypothesis concerns the effect that negative information about extensions may have on the overall family brand image. It was expected that negative information about an extension closely associated with the brand (similar product category and similar attributes) would be more detrimental to evaluations of the family brand image than information about an extension not closely associated (dissimilar product category and dissimilar attributes) with the brand schema.

TABLE 2

FAMILY BRAND IMAGE EVALUATIONS

A 2 X 2 analysis of variance was performed with the product category (similar/dissimilar) and attributes (similar/dissimilar) as the independent variables. The dependent measure was the change in brand image (difference in brand image before and after negative information about the extension was encountered). Involvement with the product category was not found to be a significant covariate, and thus dropped from the analysis.

The results indicated a marginally significant main effect for product category only ( F (1,71) = 3.003, p<.08). When the extension was in a similar product category to the brand (juice), the negative information led to an average decrease in family brand image of .56. However, when the extension was in a dissimilar product category (sherbet), the negative information led to an average increase in family brand image of 1.49. Table 2 shows the change in brand image that resulted after subjects encountered negative information about an extension.

Post hoc contrasts were performed to determine if the decrease in brand image is significant when the product category is similar to the family brand. The one-tailed tests were not significant for the two juice extensions, thus there is not adequate support for Hypothesis 2. Two additional post-hoc tests were conducted to determine if the change in image is significantly positive when the extension is in a dissimilar product category to the family brand. The results indicated that when the attributes are similar to the family brand, the change in brand image is positive (p<.05). The increase in brand image was not expected for subjects who encountered negative information about an extension in a dissimilar product category. One explanation is that people discounted sherbet as being a product associated with the family brand, and therefore its evaluation did not affect their overall schema for the brand. This explanation is developed further in the following section.

DISCUSSION

One purpose of this paper was to provide insight into the factors that influence consumers' perceptions of how closely associated the extension is to the family brand. While previous studies have manipulated extensions' product categories in order to vary similarity to the family brand, they have not manipulated extensions' attribute similarity. The results in this study indicated that the extension's product category was more important than its attributes when assessing similarity to the family brand.

The findings suggest the following relationship between negative information and the evaluation of brand extensions:

negative information may lead to more negative extension evaluations when the extensions are in a similar product category than when they are in a dissimilar product category to the family brand.

This effect might be due to the fact that negative information about extensions closely associated with a family brand is incongruous with the brand schema. This inconsistency is distinctive and results in lower evaluations.

The findings were not conclusive as to the relationship between negative information about an extension and its effect on brand image. Future research is needed to investigate whether negative information may be more detrimental to the brand's image when it is targeted to extensions in a similar product category than when it is targeted to extensions in a dissimilar product category to the family brand.

Brand extension studies have found that the transfer of positive attitudes from a brand to an extension will be facilitated when there is a basis of fit between the brand and the extension. This study indicated that this basis of fit (in terms of product category) might also facilitate the transfer of affect from the extension to the family brand. Thus, while introducing a new brand in a similar product category has benefits, it could be a detrimental strategy for the family brand image if the extension is the target of negative information (i.e., does not perform well).

The increase in brand image when the extension was in the dissimilar product category was an unexpected result. Fiske and Taylor (1984) note that if behavior is inconsistent with a schema, it can be dismissed as due to temporary situational factors, and may not be remembered. In this case, subjects might resolve the inconsistency by attributing the negative extension to be the result of the brand's inability to produce an extension in a different product category. Thus, the negative evaluations were more reflective of the product and not the original brand image. In determining the fit of the extension to the family brand, Aaker and Keller (1990) investigated the perceived ability of the firm operating in the family brand product class to make the product in the second product class. They concluded that if consumers feel that the skills needed to make the original product would not transfer to the product extension, then the perceived quality of the brand or beliefs about the brand may not transfer to the extension. This study provides some insight into the fact that maybe these extensions will not affect a brand's image when the target of negative information.

CONCLUSION

Brand extensions have become attractive marketing strategies since they save companies millions of dollars promoting a new brand name to consumers. However, critics of extension strategies are concerned about the effect that unsuccessful extensions may have on a family brand's image. This study was designed to provide insight into the effect that unsuccessful extensions may have on a family brand's image. The results found here indicated that negative information about extensions that are closely related to the family brand may be more detrimental to the family brand image than information about extensions that are not closely related to the brand name.

This study's limitations include data collection in a laboratory setting and the use of student subjects. In addition, only one brand is investigated. Another limitation is the possibility that the two manipulations (product category similarity and attribute similarity) were not independent of each other. The pre-test results indicated that the perceived fit between the extension and the family brand was very high for the extension with the similar product category and similar attributes. The perceived fit for the remaining three extensions were all very similar. Thus, a "medium fit" may have not been achieved.

However, this study does provide some initial insight into how consumers use family brand schemas to assess negative information about extensions and the effect this information may have on the family brand image. Future studies might investigate the effect of negative information on a wider range of extensions for several family brands.

REFERENCES

Aaker, David A. and Kevin Lane Keller (1990), "Consumer Evaluations of Brand Extensions," Journal of Marketing, 54 (January), 2741.

Anderson, Norman H. (1965), "Averaging Versus Adding as a Stimulus Combination Role in Impression Formation, Journal of Personality and Social Psychology, February, 1-9.

Bloch, Peter H., Daniel L. Sherrel, and Nancy M. Ridgway (1986), "Consumer Search: An Extended Framework," Journal of Consumer Research, 13 (June), 119-126.

Cohen, Joel and Kunal Basu (1987), "Alternative Models for Categorization: Toward a Contingent Processing Framework," Journal of Consumer Research, 13 (March), 455-472.

Fannin, Rebecca (1987), "Stretch," Marketing and Media Decisions, January, 22-28.

Feldman, Sheldon (1966), "Motivational Aspects of Attitudinal Elements and Their Place in Cognitive Interaction," Cognitive Consistency, S. Feldman, Ed., New York: Academic Press.

Fiske, Susan T. (1980), "Attention and Weight in Person Perception: The Impact of Negative and Extreme Behavior Journal of Personality and Social Psychology, 38 (6), 889-906.

Fiske, Susan T. and Linda M. Dyer (1985), "Structure and Development of Social Schemata: Evidence from Positive and Negative Transfer Effects," Journal of Personality and Social Psychology, 48 (4), 839-852.

Fiske, Susan T. and Mark Pavelchak, "Category-Based versus Piecemeal-Based Affective Responses," in Handbook of Motivation and Cognition, R.M. Sorrentino and E.T. Higgins, Eds., New York: The Guilford Press, 167-203.

Fiske, Susan T. and Shelley Taylor (1984), Social Cognition, Reading, MA: Addison-Wesley Publishing Company.

Jacoby, Jacob and David Mazursky (1984), "Linking Brand and Retailer Images: Do the Potential Risks Outweigh the Potential Benefits?" Journal of Retailing, 60 (2), 1-5-122.

Jolley, Bill and Dell Hawkins (1988), "Brand Extensions," in Defining, Measuring, and Managing Brand Equity, Lance Leuthesser, Ed., Report No. 88-104, Cambridge, MA: Marketing Science Institute.

Kanouse, David and Reid L. Hanson, Jr. (1972), "Negativity in Evaluations," in Attribution: Perceiving the Causes of Behavior, E.E. Jones et al. Eds., Morristown, NJ: General Learning Press.

Kesler, Lori (1987), "Extensions leave brand in new areas," Advertising Age, June 1, s1-s4.

Norris, Eileen (1986), "Companies bring logos and profits full circle," Advertising Age, June 9, s-l.

Osgood, C.E., G.J. Suci, and P.H. Tannenbaum (1957), The Measurement of Meaning, Urban: University of Illinois Press.

Park, Whan C., Bernard J. Jaworski, and Deborah J. MacInnis (1986), Strategic Brand Concept-Image Management," Journal of Marketing, 50 (October), 135-145.

Richey, M.H., L. McClelland, and A. Shimkunas (1967), "Relative Influence of Positive and Negative Information," Journal of Social Psychology, 97, 233-241.

Ries, Al and Jack Trout (1986), Positioning: The Battle for Your Mind, New York: McGraw-Hill Book Company.

Rosch, Eleanor (1973), "On the Internal Structure of Perceptual and Semantic Categories, in Cognitive Developments and the Acquisition of Language, T. E. Moore, Ed., New York: Academic Press.

Sujan, Mita (1985), "Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments," Journal of Consumer Research, 12 (June), 31-46.

Tauber, Edward (1988), "Brand Leverage: Strategy for Growth in a Cost-Control World," Journal of Advertising Research, 28 (August/September), 26-30.

University of Minnesota Consumer Behavior Seminar (1987), "Affect Generalizations to Similar and Dissimilar Brand Extensions," Psychology and Marketing, 4 (3), 225-237.

Weinberger, Marc G., Chris T. Allen, and William R. Dillon (1981), "Negative Information: Perspectives and Research Directions," in Advances in Consumer Research, Kent B. Monroe, Ed., 8, Ann Arbor: MI, 398-404.

Weinberger, Marc G. and William R. Dillon (1980), "The Effects of Unfavorable Product Information," in Advances in Consumer Research, J. Olson, Ed., 7, 528-532.

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