Special Session Summary Issues in Categorization

Citation:

Michaela Wanke and Geeta Menon (1998) ,"Special Session Summary Issues in Categorization", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 154.

European Advances in Consumer Research Volume 3, 1998      Page 154

SPECIAL SESSION SUMMARY

ISSUES IN CATEGORIZATION

Michaela Wanke, Universitat Heidelberg, Germany

Geeta Menon, New York University, U.S.A.

Assigning objects to categories is a fundamental process in judgment and decision making. Not surprisingly, consumer research is built extensively on categorization research. Vice versa, consumer research has also contributed to much of our knowledge on issues in categorization research. Of course, for consumer research the most important question is, how such processes affect consumer judgments and decisions. Although three very different contents were investigated the present papers all centered around this question. The goal was to overcome domain-specific boundaries and confer different applications of research on categorization processes.

Dhar & Simonson (Consumer Preference of Items in Related Mental Accounts: Matching vs. Balancing) look at how consumer choices within one consumption experience (category) affect each other. Much of consumer behavior involves making spending decisions that are related by temporal proximity or belong to the same event or transaction. For example, when spending an evening out, a consumer will decide how much to spend on transport, entertainment, and food etc. According to classical choice theory, the choice for each component should depend only on individuals’ preferences and the set of available alternatives in each component. Dhar and Simonson propose that people’s choices in each component of the bundle may be influenced by choices and outcomes of other components that belong to the same mental category. More specifically, they posit the matching principle: the tradeoff invoked in the first component is also favored in the other components. The predictions were supported in a series of laboratory studies. For example, subjects preferred to purchase higher quality snacks for the visit to a baseball game when they purchased seats with better view. After several potential explanations were ruled out the results suggest that when outcomes belong to the same category, subjects feel it is important to maintain the hedonic experience. The results are inconsistent with mental budgeting, the notion that the amount of money spent in the second component is smaller when there is a larger savings in the first component.

The main result of this work is that consumers strive to match choices, which are part of the same consumption experience (category). While this work focused on choices between different options within one category, WSnke (Advertising Effects on the Categorization of Brand-Members) looked at how consumers evaluate stimuli when they are presented as part of the same category. In order to improve the brand-image marketers often highlight the brand’s flagship in advertising. Their hope is that the top-of-the-line product will help to create a favorable brand image, which in turn will beneit each brand member. However, activating a brand’s flagship provides a highly positive standard of comparison against which every other brand member looks less attractive. This may prove a dilemma for marketers. The presented results show that whether the first or the second process comes to bear may be influenced by the advertising strategy. When a star and a more moderate model of the same brand were presented by ads that emphasized their common brand the more moderate model was evaluated more favorably as compared to an ad presentation where both (star and moderate model) were featured individually. In particular, the favorable influence of the star in the brand-identity condition helped the more moderate model against a similar competitor of a different brand. When the ads were individualized, the introduction of the star decreased the evaluation of the more moderate model and of a similar competitor compared to a condition where no star was present. Thus, high-lighting the flagship may hurt as well as benefit other brand-members. This paper once more drew attention to how marketing instruments such as advertising may influence categorization processes. Both papers (Dhar & Simonson; WSnke) found evidence for the inter-exemplar assimilation, which is either reflected by choices matching the other category members, or by influences on the product evaluations.

Coming from a different perspective, Menon (Are the Parts Better than the Whole? The Effects of Decompositional Questions on Judgments of Frequent Behaviors) investigated the effectiveness of breaking a behavior category into sub- categories in increasing the accuracy of the elicited frequency judgments as they are often asked in surveys. For example, a national consumer omnibus survey asks its panel members the frequency with which they purchase specific brands of detergents and cooking oil. Such frequency judgments are used by marketers in making market-size and brand-share forecasts. This omnibus survey also elicits information such as the frequency of the usage of a brand of cooking oil in different types of culinary preparations. Such information is used for market segmentation based on usage and the benefits that accrue. Thus, the accuracy of these reports is of utmost importance. Results of a study indicate that sub-categorizing a behavior through the use of a decompositional question serves to enhance the accuracy of the reports, contingent on the following conditions. First, the effectiveness of this question is moderated by the regularity of the target behavior such that the accuracy of frequency judgments pertaining to regular behaviors is not enhanced, but that pertaining to irregular behaviors is. Second, the effectiveness of this question is mediated through the cognitive process used for the task, such that the use of the decompositional question enhances episodic recall. This works well for irregular behaviors (since episodes are not well represented in memory) and decreases the associated perceived cognitive effort, but not for regular behaviors (presumably because memory-based information such as rates of occurrence are already highly accessible in memory and can be used in generating a frequency judgment), and increases the associated perceived cognitive effort. This research, therefore, provides specific guidelines for question wording in the interest of obtaining accurate data.

With subcategorization the last paper addressed an issue potentially relevant for the other two papers and categorization in general. Future research needs to address, for example, how subcategorizing consumption decisions of one category would affect the matching strategy or the transfer of brand appeal.

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