Product Typicality and Arousal: Some Initial Evidence From a Field Study of a Shopping Mall

ABSTRACT - Traditionally, the relationship of typicality and preference as regards consumer products has been measured in the following way: researchers identify a product category and then measure subjects' judgments of typicality of the instances as members of the category. Then, they measure subjects' attitudes toward the instances. Thus, for example, subjects might be asked to respond to the idea of a daily publication in folio form, with a colored, glossy cover but plain paper and black ink on the inside pages. The target category is newspaper. Traditional research suggests that the degree to which subjects judge this object a typical newspaper will be positively related to the degree to which they like the object (Loken and Ward 1990; Nedungadi and Hutchinson 1986). Other research (Meyers-Levy and Tybout 1989; Ward and Loken 1988) suggest limitations to this approach. We argue that an important factor being ignored in these studies is that products/instances may share characteristics from several categories at the same time. In this paper we apply insights obtained from prior research on the relationship between typicality and preference to shoppers' emotional responses regarding a new kind of shopping center. In particular, we consider shoppers' reports of pleasure and arousal experienced while shopping in the center.

Citation:

Kalman D. Applbaum, Kenneth Gray, and Mary-Ann McGrath (1994) ,"Product Typicality and Arousal: Some Initial Evidence From a Field Study of a Shopping Mall", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 316-319.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 316-319

PRODUCT TYPICALITY AND AROUSAL: SOME INITIAL EVIDENCE FROM A FIELD STUDY OF A SHOPPING MALL

Kalman D. Applbaum, Northwestern University

Kenneth Gray, Northwestern University

Mary-Ann McGrath, Loyola University of Chicago

ABSTRACT -

Traditionally, the relationship of typicality and preference as regards consumer products has been measured in the following way: researchers identify a product category and then measure subjects' judgments of typicality of the instances as members of the category. Then, they measure subjects' attitudes toward the instances. Thus, for example, subjects might be asked to respond to the idea of a daily publication in folio form, with a colored, glossy cover but plain paper and black ink on the inside pages. The target category is newspaper. Traditional research suggests that the degree to which subjects judge this object a typical newspaper will be positively related to the degree to which they like the object (Loken and Ward 1990; Nedungadi and Hutchinson 1986). Other research (Meyers-Levy and Tybout 1989; Ward and Loken 1988) suggest limitations to this approach. We argue that an important factor being ignored in these studies is that products/instances may share characteristics from several categories at the same time. In this paper we apply insights obtained from prior research on the relationship between typicality and preference to shoppers' emotional responses regarding a new kind of shopping center. In particular, we consider shoppers' reports of pleasure and arousal experienced while shopping in the center.

It has been demonstrated in consumer research studies (Loken and Ward 1990; Nedungadi and Hutchinson 1985) but also suggested in psychology (Barsalou 1985) and cognitive and symbolic anthropology (Douglas 1966, Leach 1964, Shepard 1978, Zerubavel 1991) that the more typical a product is of a category, the more respondents will like it. The reason atypical things are disliked, Zerubavel suggests, is that:

"Being situated within several mental fields at the same time, intermediate entities necessarily defy the...perceived mutual exclusivity of categories, thereby questioning the very viability of the boundaries separating them from one another. Essentially ambiguous, they present a serious threat to rigid classificatory structures, since, by their very existence, they call attention to the inadequacies of such structures. As such, they are often perceived...as quite dangerous" (1991:35).

The relationship between typicality and arousal, however, also manifests a counterindicating pattern. In particular, Meyers-Levy and Tybout (1989) demonstrated that mild ambiguity that results when an object is an intermediate entity ('schema incongruous') may increase arousal, as consumers are motivated to reduce the ambiguity. By this reasoning, objects that are unambiguously members of single categories, because they are easily categorized, may lead to comparatively low arousal. Objects that are ambiguous among several categories may lead to low arousal as well because they can be easily rejected as poor category members.

The inspiration for Meyers-Levy and Tybout's empirical work on schema congruity (or incongruity) and product evaluation came from a remarkable paper by G. Mandler (1982). Mandler elaborated in psychological terminology what had been observed by Aristotle and was later developed into theories of poetry and cognition by Roman Jacobson and Leonard Meyer (for Music, 1956), which is that we derive pleasure from mentally filling in the blanks between a stimulus in a series and how we expect that series to continue given the 'structure' of the series up to that point. Thus, for example, in poetry, a meter or rhyme scheme can be deviated from in order to impart a pleasing effect. Similarly, an accepted structure (such as a traditional sonnet form) can be altered for the same purpose, or the juxtaposition of two words one wouldn't ordinarily place next to each other can lead to arousal as the reader is asked to bridge the gap between the two word-concepts.

All these point to 'process effects.' Meyers-Levy and Tybout explain: "Mandler argues that the effect generated by responding to moderate incongruity will be more favorable than that typically generated by responding to either congruity or extreme incongruity... Moderate incongruities are regarded as 'interesting and positively valued,' thereby leading to more positive responses than ones elicited by schema congruity...the very process of resolving incongruity is thought to be rewarding... By contrast, extreme incongruity is defined as incongruity that cannot be resolved or can be resolved only if fundamental changes are made in the existing cognitive structure...lead[ing] more to frustration than resolution. Thus extreme incongruities typically elicit more negative evaluations than do moderate incongruities" (Meyers-Levy and Tybout 1989, 39).

Meyers-Levy and Tybout operationalized Mandler's concept to explain this nonmonotonic or inverted U-shaped relationship. Later, Meyers-Levy, Louie and Curren (1994) examined this proposition further with the aim of measuring the extent to which the phenomenon may account for people's responses to brand name extensions. Meyers-Levy et. al. found that, contrary to intuition that says that "people would judge products most favorably when they carry brands that are congruent with associations tied to the product..., brand name extensions are evaluated more favorably when a product's brand name is moderately incongruent with the product than when it is either congruent or extremely incongruent with the product" (Meyers-Levy et. al. 1994, 49). Thus, Levi Strauss could sell casual shoes as an extension of their brand name but when they ventured into structured clothing (suits), it was a flop.

The traditional straight line approach to measuring preferences and typicality has also been controverted by Ward and Loken (1988) who argued that for some categories, typicality and attitude are negatively correlated. For instance, given the category sports car, respondents will often react negatively rather than positively to highly typical instances of the category. The reason for this, presumably, is that the virtue of a sports car lies in its being unusual rather than typical. Therefore, for product categories in which uniqueness is valued, typicality will be negatively related to attitude.

In the presentation following, we comply with these lines of reasoning, and suggest further that the same positive 'process effect' of reforming schema incongruity can and should be applied also to the process whereby tertiary product attributes are brought into cognitive focus. Objects are evaluated for ultimate category coherence (a surrogate here for 'typicality' or 'schema congruity') as much by the agreement among their attributes (in recognition of the fact that products/instances may share characteristics from several categories at the same time) as by their overall fit into categories. Moreover, tertiary or less prominent product attributes also play an important role in the evaluation of objects for schema congruity.

It is best to explain 'tertiary product attributes' with an example. Consider Coca-Cola. Coca-Cola is the quintessential soft drink. In fact, its mention is all but synonymous with the category soft drink. And yet, it also has other characteristics which may be relevant to its drinkers. After soft drink, Coke is also a beverage. While not the quintessential member of the category beverage, it is nevertheless a relevant entrant. Asked which category Coke is more typical of - soft drink or beverage - few people would choose the latter. Coke is perfectly a soft drink but only secondarily a beverage. That orange juice on some occasions is considered by consumers a comparable alternative to Coke, however, means that Coke's being a member of the category beverage is not trivial. Let us take this example a bit further. While Coke is both a soft drink and a beverage, it is also a brown thing. On scale with the other two categories, Coke really is an imperfect representation of the category brown things. And yet, it is undeniably a member of that set. Moreover, Coke's membership in that category is not a nonsensical or trivial fact. Note how Pepsi (the other most typical, brown soft drink) recently launched a clear version of their soda. Obviously, the brownness of Pepsi was an attribute worth reconsidering. Thus, we may speak of the degrees to which Coke is (a) a soft drink, (b) a beverage, and (c) a brown thing. (a) and (b) and (c); not (a) or (b) or (c).

Let us take another example. At one point there was introduced into the market a yogurt drink. The product failed to define for itself a new category in consumers' minds and it was withdrawn from the market. The relevant categories appear to have been beverage, milk shake, snack and yogurt. Had consumers been asked which of these categories yogurt drinks best fit into, perhaps it would have been difficult for them to decide, except to say that it had some characteristics of each. The ambiguity of the product among these categories (at least one of which isn't even a liquid), surely contributed to its failure as a product.

What do these two examples have in common? Stated most simply, our hypothesis is that if a new product is moderately to strongly typical of two categories (and possibly weakly typical of a third category), then arousal will be high and the product will be a possibility for marketplace success.

The first information we wish to determine regarding a product/entity in our research setting is 'Which of the potential target categories display the typicality-arousal relationship and to what degree?' In the case of the yogurt drink, for instance, subjects would be asked (1) How typical is this of a beverage? (2) How typical is this of a milk shake? (3) How typical is this of yogurt? (4) How typical is this of a snack? [In our survey, we replaced typicality with similarity (e.g., how similar is this to a milk shake) because we believed that this question would be easier for respondents to answer without special instructions by the interviewer. Many researchers in psychology have defined typicality as similarity to a category, so similarity is a fair surrogate for typicality (see, Smith 1990).] We now state our propositions:

Proposition 1: There may be positive relationships between arousal and typicality with respect to more than one target category for an object or entity.

Proposition 2: Arousal will be greater when a product is typical of two categories than when a product is typical of zero, one or three categories.

THE STUDY AND PRELIMINARY RESULTS

Our preliminary results are based upon a field rather than a laboratory experiment. Other than the benefits of ideal access to a fieldsite, the authors felt that if they could question subjects "as they experienced" the new product concept that there would be greater reliability and interest in the data.

The new product is a kind of combination factory outlet mall, traditional mall and neighborhood shopping district. It also has overtones of a carnival, communicated in its decor and in the promotions created by the parent company. The shopping center (which is how we will refer to it from this point forward) offers a somewhat unique and different shopping experience. The building is about a mile long (it is non-circular) and contains more than 150 stores (about 2 million square feet), including 11 large "anchor" stores. Though many of the stores are closeout sales centers, the atmosphere much more closely approximates a retail setting than a factory outlet setting. There are two food courts and a cinema is under construction.

METHOD

20 shoppers at the shopping center were approached and asked to complete a 10 minute survey. Part one asked a number of demographic questions. Part two contained 10 of the 12 seven point semantic differential scales that had been developed by Mehrabian and Russell (1974) and applied by Pavelchak, Antil and Munch (1988) to measure pleasure and arousal due to the shopping experience. Part three of the survey contained the typicality scales. Three target categories were used: traditional mall, factory outlet mall, and neighborhood shopping district. Respondents were asked to judge how similar the shopping mall is to each of these categories on a six point scales. (Additional tasks that are not relevant to the present hypotheses were also included in the survey. These were at the end of the survey).

PRELIMINARY RESULTS AND DISCUSSION

Regression Analysis

Typicality was positively related to arousal for more than one target category. For each subject, a pleasure and arousal score was obtained by averaging the 5 scales (reliabilities have not yet been computed with the small sample size). Each dependent variable was regressed onto typicality as a traditional mall (Mallsim), typicality as a factory outlet mall (Facsim), typicality as a neighborhood shopping district (Neighsim), and agreement with the statement "GM is a new kind of place" (Newkind. The regression is significant for arousal (see Table 1). Mallsim, Facsim and Newkind are significant predictors of arousal, while Neighsim is not (see Table 2). Specifically, Mallsim and Facsim are positively related to arousal, while Newkind is negatively related. Thus, P1 is supported. The non significance of Neighsim as a predictor appears not to be due to a restricted range for this variable, as the standard deviations are nearly identical for the three typicality variables (although the mean Neighsim score is low, see Table 3).

Proposition 2

We divided the sample into two groups. Group 1 consisted of the nine respondents who found Gurnee Mills typical of exactly two categories (i.e., they rated it 4 or higher for two categories). These respondents were expected to experience high arousal. Group 2 consisted of the 10 respondents who found Gurnee Mills typical of zero, one, or three categories (there was missing data for one respondent). These respondents were expected to experience low arousal. As can be seen in Table 4, these hypotheses were supported. Subjects who found GM typical of two categories were significantly more aroused than those who found it typical of zero, one, or three categories. There is no significant difference between the groups for pleasure.

TABLE 1

ANALYSIS OF VARIANCE FOR REGRESSION

TABLE 2

PREDICTORS FROM REGRESSION FOR AROUSAL

TABLE 3

DESCRIPTIVE STATISTICS FOR TYPICALITY (SIMILARITY) VARIABLES

TABLE 4

MEANS AND T-TEST FOR AROUSAL, PROPOSITION 2

Discussion

Our preliminary results suggest that tertiary product attributes are relevant to measurements of arousal vis-a-vis typicality. That abstract product attributes are relevant to consumer decision-making is already amply demonstrated (cf. Johnson 1988). However, the contribution of abstract and/or tertiary attributes to an arousal response toward products (a contribution of particular interest with respect to new products) is less perfectly understood. The present line of research begins to explore that correlation.

REFERENCES

Barsalou, Lawrence (1985) Ideals, central tendency, and frequency of instantiation as determinants of graded structure in categories. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 629-654.

Douglas, Mary (1966) Purity and Danger. New York: Praeger.

Johnson, Michael D. (1988), "Comparability and Hierarchical Processing in Multialternative Choice" Journal of Consumer Research Vol 15. No 3.

Leach, Edmund R. (1964) "Anthropological Aspects of Language: Aniaml Categories and Verbal Abuse." In Eric H. Lennenberg, ed. New Directions in the Study of Language, pp.23-63. Cambridge: MIT Press

Loken, Barbara. & James Ward (1990). Alternative approaches to understanding the determinants of typicality. Journal of Consumer Research, 17, 111-126.

Mandler, G. (1982), "The structure of value: Accounting for taste." In M.S. Clark & S.T.Fiske (Eds.), Annual Review of Psychology: The 17th annual Carnegie symposium (3-36). Hillsdale, NJ:Erlbaum.

Mehrabian, Albert and James Russell (1974). An Applied Approach to Environmental Psychology, Cambridge, MA: MIT Press.

Meyer, L.B. (1956), Emotion and Meaning in Music. Chicago: University of Chicago Press.

Meyers-Levy, Joan and Alice Tybout (1989) Meyers-Levy, J. & Tybout, A. M. (1989). Schema congruity as a basis for product evaluation. Journal of Consumer Research, 16, 39-54.

Meyers-Levy, Joan and Therese A. Louie and Mary T. Curren (1994), "How Does the Congruity of Brand Names Affect Evaluations of Brand Name Extensions?" Journal of Applied Psychology Vol 79, No. 1, 46-53.

Nedungadi, Prakash. & J. Wesley Hutchinson (1985). The prototypicality of brands: Relationships with brand awareness, preference and usage. In E. Hirschman & M. Holbrook (Eds.), Advances in Consumer Research (Vol. 12, pp. 498-503). Provo, UT: Association for Consumer Research.

Pavelchak, Mark, John Antil and James Munch (1988), "The Super Bowl: An Investigation into the Relationship Among Program Context, Emotional Experience and Ad Recall" Journal of Consumer Research Vol. 15 No. 3.

Shepard, Paul (1978) Thinking Animals. New York: Viking.

Smith Edward E. (1990), "Concepts and induction," in Foundations of Cognitive Science, ed. M. Posner, Cambridge, MA: MIT Press.

Ward, James & Barbara Loken (1988). The generality of typicality effects on preference and comparison: An exploratory test. In M. Houston (Ed.), Advances in Consumer Research (Vol. 15, pp. 55-61). Provo, UT: Association for Consumer Research.

Zerubavel, Eviatar (1991) The Fine Line, Chicago and London: University of Chicago Press.

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