Typicality As a Determinant of Affect in Retail Environments

Citation:

John W. Barnes and James C. Ward (1995) ,"Typicality As a Determinant of Affect in Retail Environments", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 204-209.

Advances in Consumer Research Volume 22, 1995      Pages 204-209

TYPICALITY AS A DETERMINANT OF AFFECT IN RETAIL ENVIRONMENTS

John W. Barnes, University of TexasBEl Paso

James C. Ward, Arizona State University

This paper explores the influence of retail environments on consumer moods and emotions. In particular, we focus on the influence of environmental prototypicality on measures of transient mood, emotion (P.A.D.), and attitude. The linkages among environment typicality, emotion, and mood have not been explored in the marketing literature. A path analysis of our measures shows that environmental typicality has strong direct effects on feelings of dominance, pleasure, and arousal; indirect effects on mood; and direct and indirect effects on attitude. Implications for theory and management are drawn.

The environments in which consumers shop, eat, and play arouse emotions and moods. These feelings can, in turn, influence consumer attitudes and behaviors. Although retailers have at least implicitly understood these relationships for decades, consumer researchers only recently began their systematic empirical study. Early studies showed that the emotion eliciting qualities of retail and service environments could be characterized in terms of Mehrabian and Russell's dimensions of pleasure, arousal, and dominance (Donovan and Rossiter 1982). Furthermore, these and later studies showed that the experience of these dimensions is linked to attitudes, intentions, and behaviors such as approach and avoidance (Donovan and Rossiter 1982), willingness to buy (Baker, Levy, and Grewal 1992), and unplanned spending (Donovan, Rossiter, and Nesdale 1993). Later research has begun to investigate the environmental qualities antecedent to the P.A.D. dimensions, that is, the characteristics of the environment that elicit emotion.

Hull and Harvey (1989) distinguish between micro characteristics of the environment and molar characteristics. Micro characteristics are the physical characteristics of the environment, the details such as color, light, and sound used to create a particular atmosphere. These are the characteristics that most research on the environment-emotion link in marketing has focused upon. For example, Bellizzi and Hite (1992) found a relationship between retail environment color and feelings. Baker, Levy, and Grewal (1992) found that lighting, music, and the number and friendliness of employees influenced shopper emotions.

Molar characteristics are "emergent properties" that result from the sum of such details. Although researchers in environmental psychology have shown an interest in affective reactions to more molar characteristics of environments such as their complexity, mystery, and coherence (Kaplan 1987), their naturalness (Kaplan and Kaplan 1982), and their schema consistency (Purcell 1986), the study of how such qualities arouse emotion and mood in consumption settings has been neglected. An exception to this neglect is a recently presented study by Smith and Sherman (1993) that demonstrated a link between broad dimensions of store image and transient mood. In this study, we will explore the relationship of an important molar characteristic of retail environments C their perceived prototypicality C to emotion and mood, a linkage of managerial significance and theoretical interest that has so far been neglected in our literature.

The question of how environmental typicality relates to emotion, mood, and attitude is theoretically interesting and managerially important. From a theoretical perspective, the expected relationship between environmental typicality and emotion is not at first obvious. A consumer in a highly typical environment might feel either positive or negative emotions, depending upon the kind of place he or she is seeking. A consumer seeking the familiar might feel good in a typical environment; a consumer seeking a novel experience might feel bored. However, these are speculations. The categorization literature has not examined the effect of typicality on emotion and mood, and neither has the environment literature in marketing.

The categorization literature suggests that the relationship between typicality and attitude is most often positive (Loken and Ward 1990), although this is not always the case. Ward and Loken (1987) found that consumers seeking variety, prestige, or scarcity negatively valued typicality. In the sole analysis of the effects of retail environment typicality on attitude, Ward, Bitner, and Barnes (1992) found a positive relationship. But their analysis did not attempt to assess emotional reactions to environmental typicality or their role in creating a positive typicality-attitude realtionship. The role of emotions in this linkage raises interesting questions.

What emotions do more typical environments evoke? For example, do consumers in a highly typical environment feel more or less arousal, more or less dominance, more or less pleasure than those in an atypical environment? How is the pattern of emotions evoked by more or less typical environments linked to attitude? For example, if a typical environment evokes little arousal, is this linked to a positive attitude toward the environment?

From a managerial perspective, the environmental typicalityCemotion relationship should be of significant import to retailers and service providers. Retailers and service providers spend millions of dollars annually on the creation of environments designed to have positive effects on consumer emotions, attitudes, and behaviors. If environmental typicality provokes emotion, then it is an important, but largely unrecognized, aspect of store atmosphere. Both retailers and professional service providers sell the "intangible," some more than others. Both rely upon the environment as not only a tangible cue to their services, but also as a stimulus to favorable emotions likely to influence perception of both the seen and unseen aspects of their offerings. The typicality of their stores and offices is likely an important cue to consumers about the nature of these offerings, and also perhaps an important stimulus to the moods and emotions that can influence their perception.

In sum, our review suggests a need to investigate the links from environmental typicality to emotional qualities of the environment (pleasure, arousal, dominance) and from these to mood and attitude. Our study assesses these linkages via a path analysis of data from a field experiment in which paricipants visited a set of fast food restaurants chosen to vary in typicality.

METHODS

Past studies of environmental effects on consumer behavior have often substituted laboratory pictures and/or scenarios for real environments. Although studies support the validity of such simulations (Bateson and Hui 1992), we believe that the power of environments to influence emotion as well as the nuances of these effects are ideally studied through actual experience. In this spirit, all of our subjects not only visited the restaurants in our study, they ordered food and ate there.

Participants

The participants were 86 students, mostly in their twenties, attending undergraduate marketing classes at a southwestern university. Since fast food restaurants consider young adults to be a major target market, this group represented a significant segment of fast food consumers. The participants were given course credit for participation.

TABLE 1

VARIABLE MEANS AT RESTAURANT LEVEL

Cover Story

The cover story alleged the trips to the restaurants were for on-site taste tests. This story was employed to not only justify trips to the restaurants but also so the participants' attention would be on the usual focus of trips to fast food restaurants C food C instead of being unnaturally drawn to the restaurant's physical environments. The comments and questions of the participants throughout the trips indicated that they believed their purpose was taste-testing, not environmental research.

Trips

The students visited sets of restaurants in small groups supervised by a researcher. Fifteen fast-food restaurants (see Table 1 for names) were chosen as stimuli for the study by three criteria: Affiliation with a national or regional chain, within a 15 minute drive from campus, and primarily serving fast-food. Inspection of Table 1 suggests that these criteria resulted in a representative selection of restaurants including national chains specializing in a variety of food types, and some chains popular locally.

To simplify the logistics of the study, the students were divided into groups of five to eleven, and the fifteen restaurants were divided into sets of three. Visits by the entire group of 86 to each restaurant on a single trip were not possible due to transportation, time, and restaurant capacity constraints. Also, since each visit involved a taste test, the researchers did not wish to force participants to eat fast food at fifteen restaurants. Each group of students visited one set of three restaurants. Each set of three restaurants was visited by at least two groups of students totaling 14 to 19. Restaurants were assigned groups of students by a random selection without replacement procedure. Within each set of three restaurants, the order in which they were visited was counter-rotated for the two groups visiting each set.

The same researcher accompanied each group to the restaurants and administered all questionnaires. At each of the restaurant sites, participants were led through the front door of the restaurant so they could observe both its interior and exterior characteristics. The groups were seated at previously selected locations where they could observe as much as possible of the restaurant interior. To reinforce the cover story, participants taste tested an item of food at each restaurant. After the tasting, they responded to a questionnaire that asked them to rate the food's quality as well as its nutrient content. They also filled-out a short mood scale (described below) while inside the restaurant. The researcher explained the introduction of this scale by mentioning a need to control for any effect of participants' feelings on their ratings of the food.

Upon completion of the three site visits, each group returned to the university to fill out follow-up questionnaires. To continue the cover story, the first questionnaire asked for their nutritional beliefs and habits, how often they ate at fast food restaurants, and demographic information. Subsequent questionnaires contained the measure of typicality, the P.A.D. scale, the attitude scale, and other measures

Perceived Typicality

The participants were asked to rate their perceptions of the overall typicality of each restaurant they had visited on three zero to 10 scales with the following endpoints: very typical C very atypical; very good example C very poor example; very representative C very unrepresentative. These scales have been found in past research to correlate in theoretically expected ways with a variety of other measures of goodness of example as well as other measures (Loken and Ward 1990). The three scale items were summed and averaged for each respondent. The correlations of the three typicality measures with one another were all .93 or above. Overall their alpha was .94.

Attitude

Participants were asked to rate their overall attitude toward each restaurant they visited on three zero to 10 point scales with the following endpoints: low quality C high quality, bad C good; unsatisfactory C satisfactory. The three scale items were summed and averaged for each respondent. The correlations of the three attitude measures with one another were all .95 or above. Overall their alpha was .95.

TABLE 2

CORRELATION MATRIX OF VARIABLES

Mood

Mood was measured by the Peterson and Sauber (1983) MSF (Mood Short Form), a four-item scale that measures transitory mood. As noted, participants filled-out this scale while in the restaurants. The MSF asks respondents to describe "how you feel at the moment" by checking their agreement to four Likert statements: "Currently I am in a good mood;" "At this moment I feel 'edgy' or irritable;" "As I answer these questions I feel cheerful;" "For some reason I am not very comfortable right now." The second and fourth items are reverse scored. Data reported by Peterson and Sauber (1983) support the scales' validity. The scale's alpha was .71.

P.A.D.

Participants' emotional reactions to the fast food restaurant environments were measured by the Pleasure, Arousal, Dominance scale. This scale was specifically developed by Mehrabian and Russell (1974) to measure emotional reactions toward environments. The P.A.D. consists of three subscales. The pleasure scale measures peoples' general positive or negative affective reaction to the environment (i.e., their happiness, satisfaction, content, or lack thereof). The arousal scale measures the degree to which they feel awake, stimulated, and excited; or sleepy, relaxed, and calm. The dominance scale measures the extent to which they feel free to act in the situation (in control, influential) or constrained (controlled, influenced, submissive). Previous studies have validated the scale in retail environments (Donovan and Rossiter 1982). The alphas were .87 (P), .92 (A), and .89 (D).

RESULTS

A correlation matrix for the scales is shown in Table 2. The individual data for each measure was entered into a series of OLS regressions designed to estimate a path model. The path model treated typicality as an exogenous variable influencing emotions and through them such outcomes as mood and attitude. Conceptually, this path model was inspired by Bitner's (1992) conceptual model of the psychological effects of service environments. Like her model, our model begins with a physical characteristic of the service environment (typicality), assumes that this characteristic evokes emotional responses (pleasure, arousal, dominance), and then assumes that these emotional responses influence more cognitive appraisals such as attitude and mood, which has generally been defined as a more "cognitive" affective construct (Gardner 1985).

The figure shows the significant paths in the final model. The legend at the bottom indicates the significance level for each path. A summary of the results for the regressions used to estimate the direct effects and calculations for selected indirect effects are shown in Table 3. As we present the results, we will also interpret them and discuss their significance.

The figure shows that perceived typicality of the fast food environments was strongly related to emotions. The path from typicality to pleasure was .29, significant at alpha<.01. More typical environments elicited greater feelings of happiness, contentment, and satisfaction than less typical environments. We might speculate that typical fast food environments elicit more positive feelings because people desire these environments to be neither "too much nor too little." By "too much" we suggest that a fast food restaurant decorated in an opulent or unusual manner, like many trendy sit-down places, might be perceived as too expensive or too strange by many customers. By "too little" we suggest that a fast food restaurant decorated more cheaply or plainly than usual for national chains will be judged negatively, perhaps as to the quality of its food.

The path from typicality to arousal was .27, also significant at alpha<.01. Typical environments elicited greater feelings of wakefulness, stimulation, and excitement than less typical environments. The positive relationship between typicality and feelings of arousal might seem at first paradoxical, if we assume that the typical is boring. However, several explanations exist for this finding. First, part of the excitement of going to a fast food restaurant, or any restaurant, is anticipation and consumption of the food. Fast food patrons are likely to want standard fare in a standard place. To the extent the environment meets this expectation, they may feel more excited about picking up food or consuming it at the restaurant. Second, fast food restaurant environments that are less typical in a negative sense, that is, more plainly decorated than the typical place, are also likely to be perceived as less exciting. Third, environments less typical in other senses, that is, just different, or of higher than usual quality, might create feelings of disappointment, indifference, and boredom when viewed as fast food restaurants.

The path from typicality to dominance was .48, a strong relationship significant at alpha<.01. Dominance is intended to measure peoples' feeling that they are free to act in a situation versus being influenced, constrained, and controlled. In a typical environment, people may feel they know what to expect, where to go, and what to do. In other words, they may feel more free to act than in a less typical, less familiar environment. Past studies showing strong correlations between ratings of familiarity and typicality provide some support for this suggestion.

FIGURE

TYPICALITY AS A DETERMINANT OF AFFECT IN THE RETAIL ENVIRONMENT

Dominance is also related to feelings of arousal (.56, alpha<.01). Participants who felt more free to act reported greater feelings of arousal. Consumers may feel that they know what to do and what to expect in a typical environment. From another perspective, a typical environment does not confuse, disturb, or intimidate them. Therefore, they feel in control, free to act, and more excited about their trip for fast food. Consistent with the suggestion that typical environments evoke a pleasurable kind and degree of arousal, the path from arousal to pleasure is strongly significant (.65, alpha<.01).

The direct path from typicality to mood was only .11, not significant at alpha=.10. However, typicality is related to mood indirectly through dominance, pleasure, and arousal. We will examine the direct paths of these variables to mood first, and then the indirect effect of typicality on mood through them. The direct path from dominance to mood is .58, significant at alpha<.01. The direct path from pleasure to mood is .43, also significant at alpha<.01. Finally, the direct path from arousal to mood is .18, only marginally significant at alpha<.10. The indirect effects of typicality on mood through these variables are shown in Table 2. The strongest indirect effect of typicality on mood is directly through dominance, .29. The next strongest is through pleasure, .12. The other indirect effects are less than .10, but in sum suggest a substantial indirect effect of typicality on mood. Overall the antecedent variables yield a r-squared on mood of .48. The direct path from dominance to mood suggests feeling free to act is an antecedent to a good mood, a relationship that agrees with common sense, but has been rarely explored in the marketing literature. The borderline path from arousal to mood is not surprising, since most prior literature has described mood has a more cognitive than arousal based construct (Gardner 1985). Overall, the idea that retail environments influence mood is intuitively appealing. Here we see that feeling dominant, in control of the service environment, is one of the strongest antecedents to a good mood. Many studies have shown that a good mood will in turn have a prevasive influence on perception of products and services (Gardner 1985).

Turning to attitude, we find that environmental typicality has a positive direct path to overall evaluation of the restaurants (.20, alpha<.05). This relationship was expected, since past studies of the typicality-attitude relationship have shown positive correlations between these constructs. Feelings of dominance also directly effect attitude (.28, alpha<.05) as does mood (.23, alpha<.05). Overall, the regressions on attitude yield a r-squared of .30, significant at alpha<.01. The indirect effects of typicality on attitude are interesting to examine. These occur through dominance (.13) and mood (total of indirect effects through mood, .14). These results suggest that at least part of the typicality-attitude relationship for fast food environments is due to the effect of typicality on feelings of dominance and mood.

DISCUSSION

We set out to explore the effect of a molar environmental characteristic, typicality, on emotion, mood, and attitude. We found that perceived typicality had strong direct effects on feelings of dominance, pleasure, and arousal; indirect effects on mood; and direct and indirect effects on attitude. Few studies have attempted to link typicality to emotion and mood, and none of the few efforts have been in the context of retail environments. Our findings are unique, and have theoretical and managerial implications we will discuss, but also have limitations. Our study was conducted in only one type of environment, fast food restaurants. The pattern of relations between typicality and emotion could be context dependent, and thus might change in other settings, such as upscale clothing boutiques, in which atypical environments might be valued. Research designed to assess the reliablity of our findings across a variety of environmental contexts is needed. Alternative explanations for the results exist, and also need to be ruled out by later studies. For example, Zajonc's mere exposure effect could have caused more typical, more familiar environments to evoke more favorable emotions.

TABLE 3

DIRECT EFFECTS SUMMARY OF PATH ANALYSIS RESULTS

Theoretically, the results raise some interesting issues for further research. We found positive direct relationships between typicality and dominance, and typicality and arousal. Although we speculated on the origins of these relations, further work should explicate the psychology of these links. The strong relationship of typicality to dominance, and dominance to mood and attitude, is particularly interesting.

In their pioneering work on the application of the P.A.D. framework to retail environments, Donovan and Rossiter (1982) dismissed the importance of dominance as an emotional reaction to retail environments. Although their dismissal was consistent with efforts to simplify the P.A.D. framework to only two dimensions, pleasure and arousal (Russell and Pratt 1980), dominance has emerged in several studies has strongly related to environmental preception and behavior (Hui and Bateson 1991). This study confirms its importance. More work is needed on its antecedents and consequences.

From a broad perspective, we feel that environments are among the more encompassing, multidimensional phenomena marketers study, and thus deserve to be studied from interdisciplinary perspectives. In this study, we made a modest effort in this direction by bringing together constructs previously used to study categorization, environmental psychology, and mood. Future studies should continue and further such efforts.

For managers the results reinforce the importance of perceived typicality as an aspect of the retail environment. Environmental typicality not only affects categorization and attitude, but also emotion. Thus, perceived typicality might be viewed as an important emergent or molar outcome of efforts to manipulate decor, lighting, music and so on to create a desired emotional atmosphere.

REFERENCES

Baker, Julie, Michael Levy, and Dhruv Grewal (1992), "An Experimental Approach to Making Retail Store Environmental Decisions," Journal of Retailing, 68 (4), 445-460.

Bateson, John E.G. and Michael K. Hui (1992), "The Ecological Validity of Photographic Slides and Videotapes in Simulating the Service Setting," Journal of Consumer Research, 19 (September), 271-284.

Bellizzi, Joseph and Robert Hite (1992), "Environmental Color, Consumer Feelings, and Purchase Likelihood," Psychology and Marketing, 9 (5), 347-363.

Bitner, Mary Jo (1992), "Servicescapes: The Impact of Physical Surroundings on Customers and Employees," Journal of Marketing, 56 (April), 57-71.

Donovan, Robert, John Rossiter, and Nesdale (1993), presented at special session on "The Influence of Environmental Factors on Consumer Behavior: A Decade Later," in Advances in Consumer Research, Vol. XX, Leigh McAlister and Michael L. Rothschild, eds., Association for Consumer Research, 630.

Gardner, Meryl (1985), "Mood States and Consumer Behavior: A Critical Review," Journal of Consumer Research, 12, 281-300.

Hull R. Bruce and Antony Harvey (1989), "Explaining the Emotion People Experience in Suburban Parks," Environment and Behavior, 21 (3), 322-345.

Kaplan, Stephen (1987), "Aesthetics, Affect, and Cognition," Environment and Behavior, 19(January), 3-32.

Kaplan, Stephen and Rachel Kaplan (1982), Cognition and Environment, New York:Praeger Publishers.

Loken, Barbara and James Ward (1990), "Alternative Approaches to Understanding the Determinants of Typicality," Journal of Consumer Research, 17 (September), 111-26.

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

Peterson, Robert and Matthew Sauber (1983), "A Mood Scale for Survey Research," in 1983 AMA Educators' Proceedings, eds. Patrick Murphy et al., Chicago, IL: American Marketing Association, 409-414.

Purcell, Allan T. (1986), "Environmental Perception and Affect: A Schema Discrepancy Model," Environment and Behavior, 18, 3-30.

Russell, James A. and G. Pratt (1980), "A Description of the Affective Quality Attributed to Environments," Journal of Personality and Social Psychology, 38 (August), 311-322.

Smith, Ruth Belk and Elaine Sherman (1993), "The Effects of Store Image and Mood on Consumer Behavior: A Theoretical and Empirical Analysis," in Advances in Consumer Research, Vol. XX, Leigh McAlister and Michael L. Rothschild, eds., Association for Consumer Research, 631.

Ward, James C., Mary Jo Bitner, and John Barnes (1992), "Measuring the Prototypicalityand Meaning of Retail Environments," Journal of Retailing, 68 (2), 194-219.

Ward, James C. and Barbara Loken (1987), "The Generality of Typicality Effects on Preference and Comparison: An Exploratory Test," in Advances in Consumer Research, Vol. XV, Michael Houston, eds., 55-61.

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