Differential Roles of Brand-Name Associations in New Product Evaluations

ABSTRACT - This paper investigates the effect of brand name associations on the evaluation of new products. Specifically, the paper looks at the roles that brand-specific associations may play in product evaluation in cases of high discrepancy between the branded product and the target product category. A between subject experiment is carried out manipulating the brand name associations and the discrepancy between the new product and the product category in which it is introduced. Results show that a brand name closest to the representation of the new product will be evaluated the most favorably. Also brand name association effects will emerge when the representation of the new product present high discrepancy with the introduction category, and no effects in the case of moderate discrepancy. Moreover, in the case of extreme discrepancy, the brand name conveying associations to the lower (subordinate) level of the hierarchy gets the most positive evaluations. The brand name conveying associations to the intermediate (basic) level in the hierarchy gets moderate evaluations as compared to the one conveying associations to the higher (super-ordinate) level which has the least favorable evaluations.

Citation:

Tripat Gill and Laurette DubT (1998) ,"Differential Roles of Brand-Name Associations in New Product Evaluations", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 343-348.

Advances in Consumer Research Volume 25, 1998      Pages 343-348

DIFFERENTIAL ROLES OF BRAND-NAME ASSOCIATIONS IN NEW PRODUCT EVALUATIONS

Tripat Gill, McGill University

Laurette DubT, McGill University

ABSTRACT -

This paper investigates the effect of brand name associations on the evaluation of new products. Specifically, the paper looks at the roles that brand-specific associations may play in product evaluation in cases of high discrepancy between the branded product and the target product category. A between subject experiment is carried out manipulating the brand name associations and the discrepancy between the new product and the product category in which it is introduced. Results show that a brand name closest to the representation of the new product will be evaluated the most favorably. Also brand name association effects will emerge when the representation of the new product present high discrepancy with the introduction category, and no effects in the case of moderate discrepancy. Moreover, in the case of extreme discrepancy, the brand name conveying associations to the lower (subordinate) level of the hierarchy gets the most positive evaluations. The brand name conveying associations to the intermediate (basic) level in the hierarchy gets moderate evaluations as compared to the one conveying associations to the higher (super-ordinate) level which has the least favorable evaluations.

INTRODUCTION

Research has abundantly shown that characteristics of a brand name (i.e., associations, memorability an liking) contribute to consumer perceptions and evaluations of new and existing product categories (Broniarczyk and Alba, 1994; Cohen, 1982; Park, Lawson and Milberg,1989; Meyers-Levy,1989). Brand effects occur primarily by affect transfer and by the brand impact on the strategic positioning of the product within the target category (Robertson 1989; Keller, 1993; Broniarczyk and Alba,1994).

It is well known that the magnitude and the directionality of brand effects on product evaluation vary primarily as a function of the level of fit between the branded product and the product category in which it is positioned. The bulk of research on brand positioning strategies and on brand extension suggests that a better fit between a brand and the product category in which it is positioned induces a more positive evaluation (Robertson,1989; Park, Milberg and Lawson, 1991). The degree of fit is determined by the number of overlapping associations between the branded product and the target product category in which it is positioned. In this view, brand-specific associations are one among other associations (e.g., physical and functional features or benefits) that contribute to the image of the branded product and to its fit with the target product category.

However, recent research suggests that brand-name associations may have a privileged role in assessing the level of fit between the branded product and the target product category in which it is positioned. In the context of brand extension, Park, Milberg and Lawson (1991) showed that brands with functional/prestigious associations were most successful in extension in product categories that were also functional/ prestigious. In the context of brand positioning strategies, Leclerc, Schmitt and DubT (1994) observed that foreign brand names associated with cultural stereotypes of hedonism induced more positive attitudes for hybrid and hedonic products but not for utilitarian ones. They showed that a French brand name with hedonic associations would provide a better perceptual fit for hedonic products than an English name and result in more positive brand attitudes, and the opposite would be true for utilitarian products.

Particularly convincing of the unique role of brand-specific associations in branded product evaluations are findings reported by Broniarczyk and Alba (1994). In the context of brand extension, they showed that the degree of fit between brand-specific associations and the extension product category was more determinant of the evaluation of brand extension than the degree of fit between the original product category in which the brand was positioned and the extension product category. For instance, in study 2 they found that Froot Loops, a cold-breakfast cereal brand associated with sweet, flavor and kids, was evaluated more favorably in extension product categories that were highly dissimilar to its existing product category (lollipops and popsicles) than it was in extension product category that was closer to it (e.g., hot breakfast cereals). In fact, their results show that brand-specific association effects were stronger in cases of high incongruency between the original and the extension product categories

The above findings highlight the fact that brand names may convey associations that are not limited to the product category per se but to specific attributes featured at a more detailed level of representation in the product hierarchy as defined in the roschian framework (Rosch and Mervis 1975; seeMeyers-Levy and Tybout, 1989; and Stayman, Alden and Smith,1992 for application in the consumer domain). In the above example, it is the association of Fruit Loops with "sweetness" or with "for kids" that is most likely to contribute to its superior equity for lollipops or popsicles compared to that for hot breakfast cereals. Question arises as to the relative equity of brands conveying associations at different levels of a product hierarchy. Following up on this example, how would Fruit Loops (associated with features at the subordinate level of the product hierarchy) score compared to a brand associated with candies (superordinate level) or to a brand associated wth lollipops (basic level)? Also what are the roles that brand- specific associations may play in product evaluations in cases of high discrepancy between the branded product and the target product category. This is precisely the research question addressed in an experimental study reported in the present paper.

RESEARCH HYPOTHESES

Consider the introduction of a new product that presents features of a diet soft drink (e.g., no sugar, low calorie). The new brand can be introduced / positioned as a diet soft-drink (subordinate level), as a soft-drink (basic level), or as a non-alcoholic beverage (superordinate level). The new brand name for introducing this new product can itself be associated with one of these three levels of the product hierarchy. The first question that is addressed in this research is what level of brand associations is likely to have the most positive impact on the evaluation of this new product.

Recent research on brand perception and brand extensions (Broniarczyk and Alba, 1994; Leclerc, Schmitt and Dube, 1994; Park, Milberg and Lawson, 1991) suggests that a brand may be evaluated positively in a new category, regardless of its category of origin, if it contributes to the perception that the new product offers the benefits sought. In the above example, regardless of the category in which it is formally introduced (diet soft drink, soft drink, or non-alcoholic beverage), the new branded product does feature the specific attributes of no sugar and low calorie, which naturally are associated with the subordinate level. Past research on brand positioning suggests that these specific attributes will preserve their structuring property, even if the brand is introduced at higher levels of the hierarchy (i.e., as a soft drink or a beverage). Sujan and Bettman (1989) as well as Myers-Levy and Tybout (1989) have shown that attributes that are discrepant are those that are most useful in defining and distinguishing a product category level and thus the positioning of the brand in the category. Therefore, irrespective of the level of the product hierarchy in which the product is introduced, the brand that conveys associations that are the closest to the specific / discrepant features of the branded product (i.e., the brand associated with the subordinate levels) will consistently induce more positive product evaluation. Conversely, the brand that conveys associations that are the farthest to the representation of the new product (e.g. a brand associated with the superordinate levels) will consistently induce more negative product evaluation. Thus, we hypothesize:

H1a: Brand name that conveys associations that are the closest to the representation of the new product (i.e. subordinate level associations) will be evaluated the most favorably.

H1b: Brand name that conveys associations that are the farthest from the representation of the new product (i.e. superordinate level associations) will be evaluated the least favorably.

H1c: Brand name that conveys associations that are at an intermediate distance from the representation of the new product (i.e. basic level associations) will receive moderately favorable evaluations.

The above hypotheses suppose that brand effects are strong enough to induce consumers to engage in elaborate processes in perceiving and evaluating the new product. Research on brand perception and brand extension suggests that consumers are likely to invest such amount of cognitive effort only when the degree of discrepany between the new product representation and its introduction category is of some significance (Broniarczyk and Alba 1994; Meyers-Levy and Tybout, 1989; Sujan and Bettman, 1989). For instance, Broniarczyk and Alba (1994) observed brand-specific association effects only for extension categories that were highly dissimilar to the original product category of the brand. Coming back to the Froot Loops example, they found that brand associations induced more positive evaluation for two highly dissimilar categories (Lollipop and Popsicle) but had no effect on simple line extension (hot cereal) or more similar category extension (waffles).

Similarly, we propose that the evaluation effects of the level of associations conveyed by the brand will emerge only when the introduction category and the specific attributes featured in the new product present significant enough discrepancy. For example, in introducing a new product featuring diet soft drink attributes, brand effects are expected when the product is introduced as a non-alcoholic beverage but not when it is introduced as a soft drink. In the latter case, it is likely that consumers will simply resolve the presence of these less unusual features for a soft drink by invoking an alternate schema at the next lower level in the hierarchy with no further elaboration (Meyers-Levy and Tybout, 1989). In such a case, brand name associations would not have much influence in the process by which subjects would categorize this product. More likely, the brand name would only be used to the extent of adding one among many other associations to the memory representation of the branded product. Therefore, we hypothesize:

H2: Brand-name association effects will emerge when the representation of the new product presents high discrepancy with the introduction category. No effects are expected under conditions of moderate discrepancy.

How do the brand-name association effects operate when the introduction category and the representation of the new product present high discrepancy (e.g., introduction as a non-alcoholic beverage with diet attributes). In such a case, the discrepancy between the introduction category and the representation of the new product cannot be resolved by invoking an alternate schema at the next lower level in the hierarchy (Meyers-Levy and Tybout, 1989). Instead, it requires to move along multiple levels in the hierarchy. This traversal is deemed effortful due to the absence of an associate pathway (Anderson and Bower 1980). As a result, consumers are likely to engage in more elaborate processing and would look for additional cues to resolve this discrepancy between the new product and the triggered schema. We propose that brand names may become critical in such a situation and determine the impact of this information on the categorization and evaluation of the product. Moreover, we propose that brand names may operate in two different ways, depending on the level of the product hierarchy to which it is associated.

Consider brand names A, B, and C being respectively associated with the superordinate (e.g., brand name A associated with beverage), basic (e.g., brand name B associated with soft drink) and subordinate (brand name C associated with diet) levels within the product hierarchy. We propose that the choice among these levels of brand associations when the introduction category and the representation of the new product present high discrepancy (introduction as a non-alcoholic beverage with diet attributes) will contribute significantly to the evaluation of the new product in one of two ways. First, brand-name associations, specifically for the brand associated most closely with the new product representation (a brand with specific association to diet products), may intensify the discrepancy between this new product and the introduction category, reinforcing the restructuring power of these discrepant attributes, resulting in the formation of a new reference category. Past research has shown that brand names help to define category membership (Cohn, 1982). Thus, when added to the already discrepant product description, a brand name associated with the subordinate level may increase the likelihood that consumers will consider the sub-category (i.e., diet beverages) as the new reference product category. Sujan and Betman (1989) have shown that such shift in the reference product category (which they called sub-typing) occurs in conditions of high discrepancy between a branded product and the reference product category. Moreover, Barsalou (1983) has demonstrated that people construct such ad-hoc categories on an on-going basis in order to make sense of what they experience in daily life. Following work on brand evaluation (Keller, 1993, Aaker and Keller 1990), a highly positive evaluation will result from the perfect fit between the representation of the new product and the reference category that the brand has contributed to redefine.

There is another way in which brand name associations may influence consumer evaluations of new products presenting significant discrepancy with the introduction category. This effect is specific to the brand name associated with the intermediate, basic level in the hierarchy (e.g., brand name B associated with soft drink). In this case, brand associations with the middle level in the hierarchy may provide some of the associative pathway lacking between the representation of the branded product and the introduction category (Anderson and Bower 1980) thus, increasing to some extent the level of fit between the two discrepant concepts and the evaluation of the new product. Therefore, we hypothesize:

H3a: When the representation of the new product presents high discrepancy with the introduction category, the brand name associated with the lower (subordinate) level in the product hierarchy will induce the most positive evaluation compared to the brands associated with intermediary (basic) and higher (superordinate) levels.

H3b: When the representation of the new product presents high discrepancy with the introduction category, the brand name associated with the intermediate (basic) level in the product hierarchy will induce a more positive evaluation than a brand name associated with the higher (superordinate) level.

EXHIBIT 1

INTRODUCING - "QUENCH"

THE STUDY

Methodology

An experiment was conducted to test the above hypotheses. Subjects were exposed to a print advertisement for a new product being introduced under a fictitious brand name in the "non-alcoholic beverage" product hierarchy. This product hierarchy has been used in previous research on the evaluative impact of schema incongruity (Meyers-Levy and Tybout 1989; Stayman,Alden and Smith,1992). In this hierarchy, non-alcoholic beverages is at the super-ordinate level, soft drink at the basic level and sub-category of soft-drinks (e.g., diet cola) are at the sub-ordinate level.

In all conditions, the new product introduced in the market was presented as the same bundle of five features; four features were typical of soft-drinks (i.e., carbonation. tingle, served cold, available in bottles and cans) and one was unique to the new product (chocolate flavor). This description was presented in an ad for a given brand name (associated with one of the three levels in that hierarchy) positioned in a given product category (one of two levels of discrepancy, high or low, with the product representation).

Subjects

Subjects were 120 undergraduate students (54 males,66 females; average age of 20 years) registered in marketing courses in an eastern Canadian university. The study was conducted in a classroom setting.

Experimental design and variables

The experiment followed a between-subject design that combined 2 levels of discrepancy between the new product and the introduction category (moderate and extreme) with 3 levels of brand-name associations (chocolate, soft-drink and beverage) according to a full factorial design. Subjects were randomly assigned to one experimental condition.

Discrepancy between the new product and the introduction category: This factor was manipulated by the level of specificity of the product category in which the new product was introduced. Under moderate discrepancy, the new product was introduced as a new soft-drink. Under extreme incongruity, the new product was introduced as a non-alcoholic beverage. One of the experimental ads is presented in Exhibit 1.

Brand-specific associations: Three fictitious brand names were selected on the basis of a pre-test with the same subject population as for the main study. Pre-test subjects (13 subjects) were asked to indicate how strongly (1 to 7 scale; weakly-strongly) each of the brand names were associated with each of the three levels of the product hierarchy (non-alcoholic beverage, soft drink, chocolate soft drink). They also reported how much they liked the brand name (1 to 7; not at all to very much) and how memorable they perceived it to be (1 to 7; not at all to very much). Means are presented in table 1.

TABLE 1

BRAND NAME ASSOCIATIONS

TABLE 2

RESULTS

As can be seen in table 1, each brand name was uniquely associated with one of the three levels of the product hierarchy. The association of brand name Chockle was significantly higher with chocolate drink rather than soft-drink (t=6.08,df=15, p<.00) or non-alcoholic beverage (t=3.50, df=23, p<.00); Pfizz was associated more with soft drink rather than beverage (t=1.38, df=23, p<.10) or chocolate drink (t=3.95, df=19, p<.00) and Quench was associated more with a non-alcoholic beverage than a soft drink (t=1.40, df=23,p<.09) or a chocolate drink (t=4.36, df=19, p<.00). Liking and memorability did not differ across the brand names (p>0.20).

Evaluation: After having seen the ad, subjects were asked to evaluate the new product. They indicated how much was their liking for the product (1 to 7; not at all to very-much); their perception of its quality (1 to 7; poor-excellent); its appeal (1 to 7; poor to excellent); its taste (1 to 7; poor-excellent); purchase intent (1 to 7; unlikely-likely).

Results

A multivariate analysis (MANOVA ) was performed on the five evaluation variables with new product introduction category discrepancy (moderate and high) and brand-specific association (chocolate; soft-drink; beverage) as between-subject factors. Table 2 presents the statistical distance of the five evaluation variables (product liking, appeal, taste, quality and purchase intent) from the origin. The statistical distance of the five variables is chosen over a simple mean score since it takes into account the covariance between the five evaluation variables (Johnson and Wichern, 1992).

Main effects emerged for brand-associations (Wilks=.7915; F[10,182]=2.26, p<.05) thus confirming H1a that across introduction product categories, the brand conveying associations closest to the representation of product (i.e., Chockle with chocolate soft-drink) induced a more positive evaluation than the two other brands. The brand conveying associations at an intermediate distance from the representation of the product (i.e., Pfizz with softdrink) came second (confirming H1b) and the beverage-association brand being farthest from the representation was evaluated least favarobly confirming H1c). Across brand-association conditions, significantly higher evaluation was observed when the product was positioned as a non-alcoholic beverage than as a soft drink.

Main effects were also observed for new product introduction category discrepancy (Wilks=.8868; F[5,91]=2.91, p<.05) which confirmed H2,as no brand-association effects were observed under moderate discrepancy (p>.50). In contrast, under conditions of extreme discrepancy, the pattern of results is consistent with the predictions. The brand name Chockle (associated with the subordinate level in the hierarchy) was evaluated higher than Pfizz,which is associated with the basic level (p<.02); and Pfizz was evaluated higher than Quench, which is associated with the super-ordinate level (p<.00). These results are illustrated in figure 1. The interaction was also significant (wilks=.6339; F[25,339]=1.77, p<.05) confirming H3A and H3B,that under conditions of extreme discrepancy, the brand name conveying associations with the most specific/lower (subordinate) level of the product hierarchy induced the most positive evaluation, followed by the brand associated with the intermediate level (i.e., Pfizz with soft drink).

FIGURE 1

INTERACTION PLOT

DISCUSSION

The results of this study provide interesting insights on how brand names influence consumers evaluation of new products. Results showed that the positive influence of brand name associations on the evaluation of new product depends on their level of fit with the actual representation of the new product. Brand name association effects were qualified by the degree of discrepancy between the new product representation and the introduction category with significant effects emerging only in conditions of high discrepancy. Results are consistent with our propositions that this moderating effect may be due to the fact that only under these condition of high-discrepancy between a new product representation and the category in which it is positioned, would consumer proceed to the degree of elaboration that is necessary for brand name associations effects to emerge.

Results are also consistent with our propositions that, under high degree of discrepancy between the new product representation and the introduction category, brand name associations may play differentiated roles in new product evaluations. We had theorized that the presence of a brand associated most closely to the actual new product may intensify the discrepancy between this representation (at the subordinate level) and the introduction category (at the superordinate level), forcing the formation of a new reference category with which the representation now present a perfect fit. Alternatively, we had proposed that brand associations with the middle level in the hierarchy provide some of the associative pathways lacking between the representation of the branded product and the introduction category (Anderson and Bower, 1980). This increases to some extent the level of fit between the two discrepant concepts and here the evaluation of the new product. Our results were consistent with these predictions. However, these precise processing propositions will have to be tested further in studies that include process measures and measures of perceived fit.

Our results appear inconsistent with findings from schema research that has shown that moderate level of incongruity is associated with more positive evaluation than extreme-incongruity conditions (Meyers-Levy and Tybout 1989; Stayman, Alden and Smith ,1992). In accounting for this discrepancy between these results and those of the present study, it is important to note that no schema studies have considered congruency-effects specifically related to brand name associations. According to Mandler (1982), people have a natural tendency to like objects that conform to their expectations. However, because schema congruent objects usually don’t attract much attention, they come behind moderately schema-incongruent objects that attract a reasonable amount of attention while allowing for an easy resolution of incongruity. In the present case however, the schema congruity condition induced by the perfect match between the branded product and its newly specified reference product category (o schema) is not plagued by a lack of attention. Thus, the positive impact of schema congruity plus the attention that had been devoted in defining the new reference category should lead to the most positive evaluation amongst all three branding strategies.

Turning to the limitation of the study, results should be interpreted with caution due to the use of a single product category. Even though we used a product hierarchy frequently encountered in schema research in the consumer domain (Meyers-Levy and Tybout,1989; Stayman, Alden and Smith,1992),our findings should be confirmed across more categories. A second limitation stands from the lack of process measures, which would be required for a strict test of the two roles that we proposed for the influence of brand name associations on product evaluations.

In spite of these limitations, the current research provides innovative insights on brand name effects under different positioning strategies. Also, the results provide important information on the interaction between brand and product categories in consumers’ evaluation.

REFERENCES

Aaker, David and Kevin Lane Keller (1990), "Consumer Evaluations of Brand Extensions," Journal of Marketing, 54 (January), 27-41.

Anderson, John R. and Gordon H. Bower (1987), Human Associative Memory: A Brief Edition, Hillsade, NJ: Lawrence Earlbaum Associates.

Broniarczyk, Susan M. and Alba, Joseph W. (1994), "The Importance of the Brand in Brand Extension," Journal of Marketing Research, 31(May), 214-228.

Barsalou, (1989), "Ad hoc Categories," Memory and Cognition, 11 (3), 211-227. 271-288.

Cacioppo, John T. and Richard E. Petty (1982), "The Need for Cognition," Journal of Personality and Social Psychology, 42 (January), 116-131.

Cohen, Joel B. (1982), "The Role of Affect in Categorization," Advances in Consumer Research, 9, 94-100.

Johnson, Richard, A. and Wichern, Dean W. (1992), Applied Multivariate Statistical Analysis, Prentice Hall, Engelwood Cliffs, New Jersey.

Keller, Kevin L. (1993), "Conceptualizing, Measuring, and Managing Customer-Based Brand Equity," Journal of Marketing, 57 (January), 1-22.

Leclerc, France and Schmitt, Bernd H. and Dube, Laurette (1994), "Foreign Branding and Its Effect on Product Perceptions and Attitudes," Journal of Consumer Research, 31(May), 263-270.

Mandler, George (1982), "The Structure of Value: Accounting for Taste," in Affect and Cognition : The 17th Annual Carnegie Symposium, eds. Margret S. Clark and Susan T. Fiske, Hillsade, NJ: Lawrence Erlbaum Associates, 27-48.

Meyers-Levy, Joan and Tybout, Alice M. (1989), "Schema Congruity as a Basis for Product Evaluations," Journal of Consumer Research,16 (June),39-54.

Meyers-Levy, Joan (1989), "The Influence of a Brand Name’s Association Set Size and Frequency on Brand Memory," Journal of Consumer Research, 16(September),

Park, C. Whan, Sandra Milberg and Robert Lawson (1991), "Evaluations of Brand Extensions: The Role of Product Level Similarity and Brand Concept Consistency,! Journal of Consumer Research, 18 (Sept), 112-18

Park, C. Whan; Lawson, Robert and Milberg, Sandra (1989), "Memory Structure of Brand Names," Advances in Consumer Research, 16, 726-731.

Peracchio, L. A. and Alice M. Tybout (1996), "The Moderating Role of Prior Knowledge in Schema-Based Product Evaluation," Journal of Consumer Research, 23 (3), 177-193.

Robertson, Kim (1989), "Strategically Desirable Brand Name Characteristics," Journal of Consumer Marketing, Vol.6, No.4 (Fall), 61-70.

Rosch, Eleanor and Carolyn B. Mervis (1975), "Family Resemblance: Studies in the Internal Structure of Categories," Cognitive Psychology, 7 (4), 573605.

Stayman, Douglas M., Dana L. Alden, and Karen M. Smith (1992), "Some Effects of Schematic Processing on Consumer Expectations and Disconfirmation Judgments," Journal of Consumer Research, 19 (September), 240-255.

Sujan, Mita and Bettman, James R. (1989), "The Effect of Positioning Strategies on Consumers’ Brand and Category Perceptions," Journal of Consumer Research, 26 (November), 454-467.

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