One distinctive environment of undeniable importance to a large proportion of our population is the fast food restaurant. Indeed, fast food restaurants have become one of the most visible and pervasive social phenomena of our time. Nonetheless, the fast food environment has received little attention from communication scholars. This is a puzzling oversight for several reasons. First, design directors for the two major fast food chains in the United States agree that the interior design of their restaurants is the major medium they use to communicate a desired corporate image to the public. Second, the physical features of the fast food environment are so distinctive that they should have a measurable impact on the communicative behaviors exhibited by customers.

In today's competitive restaurant industry, there is a continuous struggle to be number one in profit and sales. Although many types of restaurants still try to compete for business, the fast food industry is now dominant in the restaurant business in the United States. The fast food industry in turn is dominated by McDonald's and Burger King.

McDonald's, the undisputed giant of the business, dominates the fast food industry by sheer size and volume much like General Motors has dominated the automobile industry. In 1986, McDonald's and Burger King had the highest gross food and beverage sales in the industry; McDonald's grossed almost 12.5 billion dollars while Burger King grossed over 5 billion dollars. For 1987 McDonald's and Burger King once again finished in first and second place respectively (Restaurants and Institutions, 1988).

With over 150,000 employees McDonald's is the 17th largest company in the world (Dun's Business Rankings, 1989). Although its gross sales were slightly less than half of McDonald's for the past year, Burger King has had an average net income of close to 200 million dollars for the past five years.

THE COMMUNICATION ENVIRONMENT

The environments within which we interact with others vary widely in scope from entire cities to small groups. Thus, urban planners who design cities are concerned with macrospace or the macro-environment that we associate with large spaces while students of small-group ecology, for example, focus their attention on microspace or the micro-environment (Leathers, 1976). The terms micro-environment and proximate environment are used interchangeably. The proximate environment includes everything that is physically present to the individual at a given moment. The proximate environment of a student in a classroom includes the student's desk, the other students, the teacher, the blackboard, the windows, and the doorway (Sommer, 1966). Frequently included as central, defining features of micro-environments are furniture, lighting, color choices, and decorative items (Sommer, 1969; Sommer, 1974). Thus, the fast food environment is clearly a micro-environment.

Designers of micro-environments such as the business office were concerned almost exclusively with functional questions until recently. Thus, Konar and Sundstrom (1986) wrote that "the functional qualities of offices and workspaces have always figured prominently in office design. The symbolic qualities of those workspaces and offices have typically received far less attention in this context" (p. 203). Today that perspective is changing. Designers still are concerned with such functional questions as how many offices they can get in a building with so many square feet and with the most efficient way to light such a building but they are concerned with more. Designers of corporate micro-environments are also highly concerned with the communicative significance of their decisions in terms of their impact on corporate image, the impressions these decisions make on customers and clients, and the impact on the behaviors of employees.

Physical features of the office micro-environment can and frequently do serve important communicative functions. They can, for example, define or demarcate the status of the occupant of an office (Sundstrom & Sundstrom, 1986). Status markers that serve this communicative function include furnishings, location, privacy, size, and personalization (Konar & Sundstrom, 1986). Similarly, we also know that the presence or absence of aesthetic and professionally related objects can exert a significant effect on the credibility of the occupant of a faculty office (Miles & Leathers, 1984).

The design of the furniture placed in a micro-environment is also an important variable. Considerable research has already been done on the communicative impact of furniture design. Some scholars have concluded that furniture design can communicate specific meanings to users of the furniture (Knachstedt, 1980; Mehrabian & Diamond, 1971; Stacks & Burgoon, 1981), communicate information about an owner's attitude or personality (Argyle & Dean, 1965; Kleeman, 1981; Mehrabian, 1968, 1969; Pile, 1979; Yee & Gustafson, 1983), and contribute to individual differences in affiliative behavior, most notably during user interaction and conversation (Hall, 1966; Kleeman, 1981, 1988; Mehrabian & Diamond, 1971; Sommer, 1966, 1969; Yee & Gustafson, 1983).

The impact of furniture is not limited to its size, shape, and appearance, however. Placement of furniture in the micro-environment has also been found to be important. Results from an early study by White (1953) indicated that patients in a doctor's office were more "at ease" when no desk separated doctor and patient. Joiner (1971) found that higher status occupants of business offices are more likely to place their desks between themselves and the door rather than against a side or back wall. Finally, Preston and Quesda (1974) found that occupants of business offices believe that the primary value of furniture arrangement is control of communicative interaction while occupants of faculty offices believe that type and placement of furniture serves primarily to define the occupant's image.

A limited number of studies have examined the importance of furniture design in fast food restaurants (Best, 1979; Hill, 1976; Panero & Zelnik, 1979; Woudhuysen, 1986) although numerous reviews and studies exist which have focused on the importance of furniture design and placement in the office (Alderman, 1982; Davis, 1984; Kleeman, 1981, 1988; Knackstedt, 1980; Mehrabian, 1976; Panero & Zelnik, 1979; Pile, 1979; Yee & Gustafson, 1983) and in the home (Crane, 1982; Harris & Lipman, 1984; Mehrabian, 1976; Panero & Zelnik, 1979; Sime, 1986).

Context as Communication

In one sense it is easy to embrace the old axiom that meaning is in people, not in things. The latest thinking on the communicative significance of context suggests that this view is both narrow and misleading, however. In fact Rapoport in his innovative book, The Meaning of the Built Environment: A Nonverbal Communication Approach (1982), argues that things both elicit and communicate meanings.

In fact Rapoport (1982), Hall (1974), and Sommer (1983) have provided impressive support for the thesis that environments, and more particularly the distinctive physical features of given environments, do communicate meaning. The argument is based on the assumption that individuals' communicative behaviors are directly affected by the "meanings" that environments have for them. These individuals in turn assume that the placement of furniture or the use of decorative items on a wall in a business office represent conscious and purposive decisions made by a human being. Such features of the micro-environment, therefore, serve a "latent" rather than a "manifest" communicative function. Thus, "by making the selection and placement of elements in a room purposive these environmental elements clearly become communicative" (Rapoport, 1982, p. 13).

In short, context, and the defining features of context, is a distinctive form of nonverbal communication. "The growing concern about perceived crowding, density, crime, or environment quality implies, even if it does not make explicit, the central role of subjective factors, many of which are based on the associations and meanings that particular aspects of environment have for people" (Rapoport, 1982, p. 26).

The Fast Food Micro-environment

Both McDonald's and Burger King now recognize the power of context as a form of communication. This point was repeatedly emphasized during interviews conducted by the first author with Joan Grez, director of design at McDonald's national headquarters (October 27, 1988), and Jim McMillan, director of interior design at Burger King's national headquarters (October 28, 1988). Grez and McMillan stressed that the major medium they use to communicate a desired corporate image to their customers is the design and appearance of their restaurants. They also emphasized that the most important factor affecting their design decisions is their concern that the desired corporate image be communicated to their customers.

McDonald's and Burger King might most succinctly be characterized as "high-load environments." In fact the theory-based notion of high-load environments is useful not only for describing the salient physical features of these fast-food environments but also for providing a basis for predicting how customers' communicative behaviors will be affected by these environments. Thus, Mehrabian (1976) maintains that high-load environments are emotionally arousing and involving. They are arousing because they force individuals who work or interact in them to process much information and to cope with many high intensity environmental stimuli. For example, red wallpaper, no partitions to separate the desk of occupants, high density, and excessive noise levels in a business office would define a high-load environment.

In contrast, a work space with walls, subdued colors and lights, and carpeting to control or eliminate disruptive noises is called a low-load environment. Although the high-load environment is emotionally arousing and promotes communicative interaction, it is often perceived to be unpleasant because the collective impact of the intense micro-environmental stimulation produces discomfort and--over time--fatigue. Low-load environments by contrast are pleasant, relaxing, and should be comfortable although they may be insufficiently stimulating for people to function effectively within their confines.

Baum and Paulus (1987) have elaborated upon the notion of the high-load environment with the "overload model" of space utilization that identifies the effects of "crowding" in a given context. The model is based on the assumption that high levels of density in a social setting may result in a state of stimulus overload. In fact, the feeling of being crowded can be explained as the result of experiencing excessive stimulation. The "overload model" posits that placing individuals together in a limited amount of space (high density) can be involving and arousing but beyond some optimal level the individuals involved will experience discomfort and may ultimately withdraw.¹

The fast-food environments at both McDonald's and Burger King are highly involving because of their high-load characteristics. As we shall see, the major means of promoting involvement in a proximate environment is to create a sociopetal spatial orientation. Sociopetal spatial orientations have individuals facing each other directly while separated by as little distance as is socially appropriate. Fast-food restaurants are committed to the sociopetal use of space. In contrast, a railway station with long rows of seats bolted together back-to-back epitomizes a sociofugal spatial orientation (Gifford, 1987).

McDonald's strong commitment to the use of anthropometric² standards in restaurant design almost, by definition, dictates that theirs will be a high-load environment which simultaneously seems calculated to stimulate customer involvement and discomfort. Although both McDonald's and Burger King use anthropometric information to design their restaurants, McDonald's adheres very closely to the recommendations on size of furniture set out in a reference work by Panero and Zelnik, Human dimensions & interior space (1979). In so doing McDonald's relies almost exclusively on a "designers perspective," as opposed to a "user's perspective," in designing its restaurants (Sommer, 1983).

By conforming so closely to anthropometric standards, McDonald's designs their tables and chairs so that they are only large enough to accommodate human bodies of "average dimension." The decision to meet only the most minimal size and quality standards for their chairs and tables in turn has major communicative implications. The decision helps make McDonald's a particularly "high-load environment." Moreover, results from the observational part of our study will show that there are many other physical features of the physical environment which make McDonald's an unusually high-load environment. McDonald's high-load environment would seem to support the expectation that customers will be both more involved with customers at their table and less comfortable than customers in a lower-load environment.

Since McDonald's and Burger King make many of their advertising and promotional appeals directly to children, it is understandable that they would design their restaurants in a way that would promote involvement. High customer involvement can also serve the objective of fast customer turnover and high profits, however. Thus, the "Conrad Hilton Effect" may help explain some of the design decisions made by McDonald's in particular. Hilton discovered that the more comfortable he made lounges in his hotels the more likely it was that individuals would sit for extended periods of time without spending much money. This information led Hilton to reduce the size of his lounges in the interests of increasing profits (Gifford, 1987).

McDonald's Image Problem

O'Hair, Kreps, and Frey (1990) maintain that applied communication research should address and help provide solutions for specific problems. They write that "Although applied research may well advance questions about theory and methodology posed by communication scholars from within the discipline, the primary purpose of applied communication research is to use theory and methodology in order to understand how communication works within particular settings to solve specific problems" (p. 4).

This study focuses on the most serious image problem that McDonald's has experienced. Dannatt (1988) suggested the nature of McDonald's problem when he wrote that:

The designers of McDonald's are legendarily shy, but they faithfully follow Ray Krock's masterplan of interior decoration, right down to the bland prints on deckchairs. The seating, designed at that notorious 10 degree forward tilt to stop customers lingering more than 10 minutes, feels exactly the same all over the world, an odd blend of comfort and ill ease (p. 62).

McDonald's has probably given a new and particularly unsavory meaning to the term "fast" food with seats tilted forward at a 10 degree angle. The tilted seat alone has done much to nourish the public's perception that discomfort is built into McDonald's restaurants. The seemingly unavoidable inference is that McDonald's conscious intent has been to make the customer uncomfortable to hasten her/his departure and stimulate profits. The 10 degree angle of the seats in McDonald's represents only one of a number of design decisions that might have the effect of reinforcing the perception that McDonald's restaurants are designed to be uncomfortable.

The use of anthropometric standards to design their restaurants also tends to highlight the discomfort issue. Thus, McDonald's restaurants are designed via the use of anthropometric standards which in turn were derived from research done on NASA; research for NASA test pilots is based on the premise that a minimum amount of space must be used to accommodate the human body and make possible the functions it must perform.

McDonald's has never publicly acknowledged that they are concerned about the widespread public perception that discomfort is built into their restaurants. Their recent actions suggest that they are attempting to combat the "discomfort issue," however. Joan Grez, national design director for McDonald's (1988), now emphasizes that McDonald's attaches a high priority to designing their restaurants to be comfortable. Second, McDonald's has been stressing for almost a decade that the customer will encounter a "homey" atmosphere in their restaurants which is certainly suggestive of a high degree of comfort. Finally, McDonald's seems to have freed their designers from the constraint of designing their restaurants so that the maximum number of customers must be put in the minimum amount of space. Woudhuysen (1986) suggests that McDonald's has moved directly to deal with their most serious image problem, the public perception that customer comfort has been sacrificed in order to achieve maximum profits. Thus, Woudhuysen (1986) writes that "Now, however, things are different, and McDonald's allows its designers greater freedom to make people comfortable, apparently eschewing the usual fast-food philosophy of the quickest possible customer turnover" (p. 61).

Statement of Purpose

The present study focuses on the subject of whether, and in what ways, the distinctive physical features of the fast food micro-environment at McDonald's and Burger King affect their customers' behaviors. A direct comparison is made between the behaviors exhibited by customers at McDonald's and Burger King and a direct comparison is made as well between the theoretically-relevant features of these two micro-environments. This comparison seems particularly appropriate in view of the fact that Burger King is McDonald's chief rival and that McDonald's and Burger King sit atop the fast food industry.

Although the leading interior designers at McDonald's and Burger King agreed to a total of four interviews for the present study, they seemed reluctant to talk in detail about the defining features of the corporate image that they want to communicate to their customers. The interviews do certainly support one conclusion, however. McDonald's and Burger King are both centrally concerned with how involving and how comfortable (or uncomfortable) the dining experience is in their restaurants.

The reasons why both McDonald's and Burger King want to provide an involving experience for their customers have already been discussed. McDonald's and Burger King are also committed to making their restaurants comfortable but for somewhat different reasons. McDonald's has been forced to make a public pledge to make their restaurants more comfortable in order to deal with the "discomfort issue." In contrast, Burger King has made a strong commitment to customer comfort without experiencing any external pressure to do so. Jim McMillan, Burger King's director of interior design, stressed that designing their restaurants so that they are comfortable has long been one of Burger King's highest priorities.

The present study is comparative in the sense that the behaviors of customers at McDonald's and Burger King are assessed to determine relative degree of involvement and comfort. Furthermore, the salient physical features of the micro-environments at McDonald's and Burger King that have the greatest potential to affect customer involvement and comfort are identified and their potential impact on customers' involvement and comfort is discussed. Multiple sources of information are used to make those judgments with the primary emphasis on a coding of customers' behaviors that have previously been proven to be accurate indicators of comfort and involvement.

Given the focus of this study, four research questions are addressed:

(1) Is there a significant difference in the level of involvement of customers at McDonald's and Burger King?

(2) Is there a significant difference in the level of comfort of customers at McDonald's and Burger King?

(3) What design choices and physical features of the micro-environments at McDonald's and Burger King might contribute to a different level of customers' involvement?

(4) What design choices and physical features of the micro-environments at McDonald's and Burger King might contribute to a different level of customers' comfort?

METHOD

In this study a triangulation approach was employed so that multiple sources of information could be used in the attempt to answer the four research questions. As is often the case with qualitative research which features direct observation of relevant communicative contexts and behaviors, research procedures were refined as the study developed.

Sources of Information

In this study, information was generated from five different sources: (1) A review of the relevant literature with an emphasis on the anthropometric standards McDonald's and Burger King designers use to decide on the size of their tables and chairs; (2) Preliminary field notes; (3) Coding of customers' behaviors in McDonald's and Burger King restaurants; (4) Detailed, focused field notes; and (5) Telephone interviews with the national design directors for McDonald's and Burger King.

The preliminary field notes were particularly useful in helping to determine the feasibility of subsequent observational and coding procedures. The first author made preliminary field notes on micro-environmental variables that seemed to be particularly relevant to the four research questions during his visits to ten fast food restaurants (five McDonald's and five Burger King restaurants). In particular, preliminary field notes were taken which described the number, size, type, and placement of tables, the presence of decorative items such as plants, and mirrors, the use of partitions, and the type and intensity of lighting used.

At this point, the decision was made to focus the study on five McDonald's and five Burger King restaurants. All ten restaurants observed were suburban restaurants. Seven of these restaurants were in suburban areas adjacent to a university community in the southeast with a metropolitan population of over 100,000. The other three restaurants were located in the outer suburbs of a southeastern city with a metropolitan population of over two million. In short, all ten restaurants that were visited were "suburban" as opposed to "inner-city" fast-food restaurants.

After these ten restaurants were chosen, the first author returned to each of these restaurants on at least two occasions. He then took detailed, focused field notes with emphasis on the features of the fast food micro-environments and customers' behaviors which seemed to be particularly relevant to making judgments about the level of customers' involvement and comfort.

Behavioral Indicators of Involvement and Discomfort

The research questions in this study dictated that behavioral indicators of involvement and discomfort be identified and defined so that these behaviors could be coded when they were exhibited by customers at McDonald's and Burger King. Involvement behaviors were defined as "voluntarily expressed behaviors that signal participation, interest, and occupation with the other person in an interaction" (Burgoon, Buller, & Woodall, 1989, p. 314). Burgoon et al. (1989) argue that involvement and immediacy behaviors are essentially synonymous. Forgas (1978) agrees with Burgoon et al. with regard to the similarity of involvement and immediacy. In fact, Forgas stated that the best way to judge one's involvement with someone else during interaction is to observe his/her immediacy behaviors with the other person. Several examples of immediacy behaviors include close distance, direct body and facial orientation, forward lean, and positive reinforcers such as smiling, head nods, and pleasant facial expressions (Cappella & Greene, 1984; Coker & Burgoon, 1986).

The type of bodily movement that is known to communicate discomfort in both a symbolic and literal sense is the adaptor. Ekman & Friesen (1969) maintain that adaptors, in their original form, were bodily movements used by individuals to cope with some type of physical discomfort--the itch is scratched, for example. In their current form adaptors are defined as a communicatively significant type of bodily movement that typically is not consciously used and is reflective of a negative psychological state such as anxiety.

Adaptors in turn take the form of self-adaptors and object-adaptors. Self-adaptors involve touching of one's own body that often includes touching of one's head or face, for example, while object-adaptors refer to the use of hands to touch or play with objects. In any event, a number of bodily movements that take the form of adaptors are now recognized to be accurate indicators of discomfort (Nespoulous and Lecours, 1986). We know, for example, that when job interviewees exhibit such gestural adaptors as touching their hair, upper torso, and neck that these bodily movements are interpreted by the job interviewer as signs of discomfort (Goldberg & Rosenthal, 1986). Finally, fidgeting is a type of bodily movement that is widely recognized as an accurate indicator of, or measure of, discomfort (Mehrabian & Friedman, 1986).

The following variables were chosen as behavioral indicators of involvement: smiling, amount of talking, head nodding, and leaning. Customer discomfort was operationalized via the following behavioral indicators of discomfort: fidgeting, extraneous foot movement, postural shifts, and side-to-side movement.³

Observer Agreement

There are many ways of checking reliability or observer agreement. As documented in earlier research by Gottman (1979), the observer may use a "reliability checker" to reassure herself/himself that the coding was in fact accurate. In this case a straightforward training procedure was used to familiarize the reliability checker with coding procedures and to undertake the reliability check.⁴ A reliability check was made at twelve tables at both a McDonald and a Burger King. The following figures are percent agreements between the coder and the reliability checker; Cohen's Kappa appears in parentheses: Behavioral Indicators of Involvement: smiling=87% (.72); amount of talking=72% (.58); head nodding=95% (.83); and leaning=83% (.71). Behavioral indicators of Discomfort: fidgeting=97% (.91); extraneous foot movement 100% (1.00); postural shifts=77% (.72); and, side-to-side movements=92% (.88).

Observational and Coding Procedures

The research questions dictated that observational and coding procedures be developed that would ultimately yield reliable information as to the relative level of customers' involvement and comfort at McDonald's and Burger King.

The observational/coding procedures took the following form. The first author would arrive at a given restaurant and seat himself strategically so that he could observe directly the behaviors of a maximum number of customers while using the cover of a plant or some other object in the restaurant so that his observations were inconspicuous. In addition to his coding sheets he brought with him a Graduate Record Exam Study Guide to create the impression among customers who noticed him that his purpose for being there was to eat and prepare for the GRE exam.

The observer/coder, the first author, usually sat at a table in a corner of the restaurant often at the back where he had the best view and could observe the most customers in the restaurant. He would look for a table that had at least two customers seated and would begin to write down information about table number and type, seating orientation, and the number in the party. The behaviors of the customers seated at a given table were observed three times and behaviors observed were coded on coding sheets.

Customer behavior was scanned from left to right at the table and recorded on a coding sheet. Each customer was scanned for behavior on three occasions, except when the customer left before the observer was finished. Each visit to a restaurant lasted between one and two hours, depending on the business and ease of coding of the behaviors of customers seated at tables. By the time observation/coding was completed a total of 208 tables were observed: 102 at McDonald's and 106 at Burger King. A total of 473 customers were observed; 232 at McDonald's and 241 at Burger King. There were twenty-one total visits.

Statistical Testing

Chi square tests were run for each of the four behavioral indicators of involvement and for each of the four behavioral indicators of comfort to determine whether there were significant differences in the amount of these behaviors customers displayed at McDonald's and Burger King. Results for each behavioral indicator are reported as percentages of the total number of behaviors observed. Table 1, for example, reports the amount of smiling at McDonald's versus Burger King. Thus, customers' behaviors were observed a total of 610 times at McDonald's to determine whether smiling was or was not occurring. In 90 out of the 610 instances, or 14.8% of the time, when customers' behaviors were observed they were smiling; by comparison customers at Burger King were smiling only 9.3% of the times they were observed. Telephone Interviews

Four telephone interviews were conducted for this study. Joan Grez, director of design for McDonald's (October 27, 1988), and Jim McMillan, director of interior design at Burger King (October 28, 1988), were interviewed by the first author before the data-gathering and data-analysis phases of this study began. Their perspectives were useful in helping to develop the conceptual focus of this study. After the data-gathering and data-analysis phases of the study were completed, the first author conducted a follow-up telephone interview with Joan Grez on November 2, 1990, and a telephone interview with Buzz Alexander, director of construction for Burger King, on November 9, 1990. These last two interviews proved useful as a way of sharing the results and analysis with high level officials at McDonald's and at Burger King.

RESULTS

This study was designed to produce information that would help provide answers to four research questions. Results reported in Tables 1-8 provide data that make possible the comparison of levels of customer involvement and comfort at McDonald's and Burger King. These results focus directly upon the first two research questions of this study. The third and fourth research questions are addressed latter in the Results section where physical features of McDonald's and Burger King's micro-environment that are relevant to involvement and comfort are described and analyzed.

The results reported in Tables 1-8 are easy to understand if several facts are kept in mind. Four dependent variables were identified as measures of customer involvement in this study--smiling, talking, leaning, and head nods. The behaviors of McDonald's and Burger King customers were compared to determine whether there were statistically significant differences between them on each of these four measures of customer involvement. A separate table is devoted to results for each of these measures of customer involvement--see Tables 1-4. For example, results reported in Table 1 focus exclusively on the amount of smiling done by customers at McDonald's and Burger King. The four dependent variables identified as measures of customer discomfort were fidgeting, extraneous foot movements, side-to-side movements, and postural shifts. The results in Tables 5-8 for the customer four discomfort measures are presented in exactly the same way as for the customer involvement measures. Finally, the independent variables were table type, seating orientation, and cushioning of chair. Separate chi square tests were run to determine if there were significant differences between McDonald's and Burger King customers for different seating types, seating orientations, and cushioning (or no cushioning) of chairs.

The results in Tables 1 and 2 suggest that the level of customer involvement was significantly higher at McDonald's than at Burger King. Smiling and amount of talking clearly communicate involvement in an interpersonal context and there was much more of both at McDonald's.

The significant Main Effect for smiling ( ²=8.3, p< .01), which is reported in Table 1, shows that customers at McDonald's were smiling 14.8% of the times they were observed in comparison with Burger King customers who smiled only 9.3% of the time. The chi square for Main Effects for smiling is reflective of the fact that frequency of smiling of McDonald's customers (when smiling figures for all types of customers' seating are combined, i.e., deucers, 3-seaters, 4-seaters, adjacent, across, and offset seating, and chairs with and without cushioned backs) is significantly greater than for Burger King's customers (when smiling figures for all types of customers' seating at Burger King are combined). When the amount of smiling of McDonald's and Burger King customers with a given seating orientation is compared, we find a number of instances where there was significantly more smiling by customers at McDonald's. Thus, 16.3% of McDonald's customers who were seated at 4-seaters smiled when observed while only 11.2% of their counterparts smiled at Burger King ( ²=4.4, p< .05), and significantly more McDonald's customers seated on chairs with no cushions (14.7% versus 0.00%) smiled than did Burger King customers seated on no-cushion chairs ( ²=7.5, p< .01).

Results in Table 2 indicate a clear and definite difference in the amount of talking at McDonald's and Burger King; customers talked much more often at McDonald's. The significant Main Effect for amount of talking ( ²=15.7, p< .001) highlights the fact that customers observed at McDonald's (when all of seating orientations are combined) talked almost twice as often as the customers observed at Burger King (talking 22.5% of the time observed versus 12.7%).

The difference in the amount of talking and the pattern of more talking at McDonald's is clear and striking. No matter what the table type the amount of talking was greater at McDonald's than at Burger King. McDonald's customers eating at deucers ( ²=7.3, p< .001) and at 4-seaters ( ²=6.9, p< .01) talked significantly more often than their counterparts at Burger King.

Although the results in Table 1 and 2 clearly suggest that customers at McDonald's are more involved than those at Burger King, results in Table 3 seem to suggest that customers at Burger King were more involved than at McDonald's. As we shall subsequently point out, however, amount of customer leaning in these fast food micro-environments may be a much less reliable indicator of customer involvement than amount of smiling and talking.

The results in Table 3 show at least two things. Customers at both McDonald's and Burger King did a lot of leaning no matter what type of table they were seated at, what their seating orientation was, or whether they sat in a chair with a cushioned back or with no cushion. Thus, customers were observed 715 separate times at Burger King and 539 times, or 75.4% of the times observed, they exhibited forward lean; customers observed at McDonald's were leaning forward 62.2% of the times observed. This finding represents a significant difference for Main Effect ( ²=27.9, p< .001). Secondly, the amount of leaning for customers at Burger King was also significantly greater than for customers at McDonald's for customers seated at deucers, 4-seaters, in the across seating orientation, and for those in chairs with cushioned backs.

Finally, the results in Table 4 indicate that there are no significant differences between McDonald's and Burger King's customers either for Main Effects or for each independent variable. In fact these results indicate that customers did very little head nodding in either fast-food restaurant.

The information reported in Table 5 and 6 suggest that customers were less comfortable at McDonald's than at Burger King although the results obtained for coding of behavioral indicators of comfort are less compelling than for involvement. Results for fidgeting seem particularly noteworthy, for example. Fidgeting, as noted earlier, has been perceived traditionally as a clear and unequivocal indication of discomfort. The significant difference for Main Effects reported in Table 5 shows that customers at McDonald's exhibited significantly more fidgeting than at Burger King ( ²=9.1, p< .01). Results in the same table show that customers seated at deucers in McDonald's exhibited significantly more fidgeting than at Burger King ( ²=6.1, p, <.05), and the deucer is the workhorse of the fast food industry where so many customers typically sit.

The results in Table 6 should be viewed with caution. There is a pattern in these results which suggests that McDonald's customers exhibited more extraneous foot movements than Burger King customers but only one difference is significant. McDonald's customers seated at deucers did exhibit a significantly greater amount of extraneous foot movements (11.2% versus 2.5%) than customers at Burger King who were eating at two-seaters ( ²=4.1, p <.05). Overall, customers exhibited a greater amount of extraneous foot movements at McDonald's than at Burger King (7.1% to 5.0%), customers seated at 3-seaters exhibited more extraneous foot movements at McDonald's (8.9% to 0%), and McDonald's customers seated in chairs with no cushions exhibited more extraneous foot movements (9.7% to 5.4%).

Finally, Table 7 and 8 reveal that there were no statistically significant differences in the amount of side-to-side movements and postural shifts exhibited by customers at McDonald's and Burger King. Customers at McDonald's exhibited a greater amount of side-to-side movements than their counterparts at Burger King on eight of the nine comparisons made although some of the differences are very small. Similarly, results in Table 8 show that customers at McDonald's exhibited more postural shifts than customers at Burger King for seven of the nine comparisons made on the postural shifts indicator of discomfort. The non-significant results reported in Tables 7 and 8 do not, however, justify a conclusion that the side-to-side movements and postural shifts of the customers observed in this study by themselves are indicative of a higher level of customers' discomfort at McDonald's than at Burger King.

Observational Information

The many hours spent in McDonald's and Burger King restaurants also produced much observational information that can be used to address the third and fourth research questions. Those questions ask whether the design choices and physical features of the micro-environments at McDonald's and Burger King might contribute to different levels of customers' involvement and comfort. Observations that seemed particularly relevant to these two questions were recorded in both the preliminary field notes and in the final, detailed field notes.

In terms of potential impact on level of customers' involvement and comfort the most important design decisions made seem to be the size of the tables and chairs. This is so because the size of these furniture items in McDonald's and Burger King proved to be the most important determinant of the spatial orientations of interacting customers. Certainly the distance which separates two people can have a major impact on both their level of involvement and comfort.

The size of the "deucer" tables at McDonald's and Burger King is the same: 21" x 24." However, the size and appearance of deucer seats is discernably different. Burger King's deucer seat measures 18½" wide x 17" long while McDonald's deucer seat is 15½" wide x 16" long. In Burger King's newest restaurants the deucer seats are mounted to the floor and they swivel both sideways and back and forth so that forward lean is facilitated. The deucer seats at Burger King sit well back from the table so that customers facing each other and sitting in upright positions will be separated (chest-to-chest) by a distance of 3 feet 5 inches. In contrast, two people seated at a deucer at McDonald's with their feet under the table will be separated by no more than 2 feet 2 inches.

The appearance of deucer tables and chairs at the two fast food restaurants is also notably different. Burger King uses wooden tables and chairs; McDonald's uses plastic in both instances. Because of their size and appearance, Burger King chairs and tables appear to be of higher quality and they appear to be more comfortable. For example, a Burger King deucer chair has eleven supporting wooden spokes in the chair back; McDonald's deucer seats have five plastic spokes or strips for support in the chair back. Both authors of this article experienced noticeable discomfort when seated at McDonald's deucer tables. The discomfort resulted because of the obviously small size of the seats and the inferior back support.

In addition to furniture three other physical features of the micro-environments at McDonald's and Burger King seem to be particularly important: (1) lighting; (2) color choices; and (3) mirrors and partitions. The contrast in lighting at McDonald's and Burger King is certainly an important feature of the micro-environment. Research on lighting in workspaces has shown that lighting is one of the most important factors contributing to the comfort of individuals who work there. In fact studies of lighting in professional micro-environments indicate quite clearly that excessive illuminance may cause discomfort. Luminance is the apparent brightness of a surface and is the joint product of illuminance (light falling on the surface) and the surface reflectance (Ellis, 1986).

McDonald's in contrast to Burger King is intensely lighted on both the outside and the inside. The high intensity lighting used at McDonald's is known as "high intensity discharge" lighting or H.I.D. (Smith & Bertolone, 1986). During the observational phase of this study, a number of customers at McDonald's were observed complaining about the intense light which was attributable to numerous fluorescent lights and windows, much larger than those at Burger King, which have neither shades nor tinting; Burger King controls the intensity of lighting within their restaurants with both window shades and tinting of their windows. McDonald's used fluorescent lighting in all restaurants which were observed while Burger King frequently used light bulbs with shades. On the outside too McDonald's uses vertically placed florescent lights. The level of lighting at McDonald's is so intense and high that a McDonald's restaurant can be seen from far away by an approaching motorist.

Color choices made by designers at McDonald's and Burger King are also quite different. McDonald's consistently opts for colors that are known to be stimulating and emotionally arousing (Smith & Malandro, 1985). On approach a McDonald's restaurant stands out because of the golden (actually yellow) arches with McDonald's name displayed in white letters against a bright red background. In case this color combination is not enough to catch your attention, the American flag is conspicuously displayed in front. On the inside McDonald's uses the same strong colors with eye-catching wall paper hues of pure white and yellow. Burger King in contrast specializes in more subdued color combinations. Burger King frequently uses rustic colors in an apparent attempt to relax customers or put them at ease.

Both McDonald's and Burger King use mirrors and partitions. Mirrors are presumably used to create the illusion that more space is available than actually is. McDonald's mirrors are often prominently placed in the main dining room or near the order-taking area. Mirrors are also used at Burger King but are used less frequently and are not usually placed in high traffic areas.

Finally, both McDonald's and Burger King use some partitions. In McDonald's partitions often seem to serve more of a symbolic than a functional purpose in the sense that they might provide the customer with increased privacy. McDonald's uses wooden railings to separate one eating area from the other, but the railings are held up by see-through spokes that hardly serve as partitions. Thus most of the customers in the main dining area have the opportunity to engage in direct visual inspection of most of the customers. Furthermore, customers in McDonald's in contrast to Burger King can be directly and easily observed by passing motorists since McDonald's has large windows with no tinting or shades and the level of illumination inside is so high that little if anything is concealed from the prying eye.

DISCUSSION

The results from this study serve to emphasize the importance of communication as context. Data reported in Tables 1-8 show that McDonald's customers in the suburban restaurants in this study exhibited behaviors that indicated that they were both more involved and less comfortable than their counterparts at Burger King. While a number of variables may have had an impact on the behaviors of McDonald's and Burger King's customers, it is important to recognize two things at minimum. First, the distinctive physical features of McDonald's suburban restaurants means that McDonald's by definition is clearly a "higher-load micro-environment" than the micro-environment at the Burger King restaurants that were observed. Secondly, the theory-based conceptualization of "high-load environments" supports the expectation that McDonald's customers will experience a dining micro-environment that is both more involving and less comfortable than the dining micro-environment at Burger King.

The results of this study draw attention to the important link between context and behavior. They also draw attention to the fact that experts disagree as to the potential of the built environment to affect the behaviors of individuals who interact within a given type of built environment. In fact there currently are three major conceptualizations that attempt to specify the potential impact of the built environment on the behaviors of people in a given built environment: architectural determinism, environmental possibilism, and environmental probablism (Bell, Fisher, Baum, & Greene, 1990).

Architectural determinism is based on the rather extreme assumption that the physical features of the built environment are basically the only cause of the behaviors of individuals who use the built environment. In contrast environmental possiblism assumes that the built environment has little, if any potential, to affect behavior. Environmental probablism currently has the most support among scholars of the environment and seems to be most useful in considering the impact of the context in a study such as the current one. Environmental probablism is based on the assumption that a given environment can strongly affect behaviors, that the probable impact of the environment on the behaviors of individuals who use an environment can be predicted, and that other factors interact with context to affect behaviors.

When the results from this study are interpreted, it is important to recognize that generalizations from an initial study such as this one must be qualified. Clearly we cannot generalize results from this study to McDonald's and Burger King restaurants on the international level or to all types of McDonald's and Burger King restaurants across the United States. Nonetheless, it is important to note that this study focuses upon a particular type of restaurant that is built by McDonald's and Burger King, the suburban restaurant.

All of the restaurants in the present study were suburban restaurants. The design and physical features for McDonald's and Burger King's suburban restaurants is highly standardized across the United States; the design directors for McDonald's (Grez, October 27, 1988) and for Burger King (McMillan, October 28, 1988) agreed in interviews with the first author that the physical features of their suburban restaurants in various parts of the country differ very little. In her most recent interview with the first author Grez (November 2, 1990) emphasized that there is a great deal of consistency in the design of McDonald's suburban restaurants across the country; for example, the type and dimensions of the furniture used in McDonald's restaurants is the same from state to state. The design directors stressed that the big difference(s) in the dining micro-environments for these two fast-food chains is between the inner-city and suburban restaurant. For example, both fast-food chains use stand-up tables with no chairs in their inner-city restaurants in Miami, Chicago, New York, and Los Angeles.

The central focus of this study is on context as communication. In particular, the study sought to determine whether the distinctively different fast food micro-environments at McDonald's and Burger King were associated with differences in the behaviors exhibited by the customers, respectively, at McDonald's and Burger King. In fact, the customers' behaviors observed in this study suggest that contexts can communicate messages. In the case of McDonald's, for example, the challenging, interpretive questions are (1) what message(s) was communicated by context that in turn resulted in a serious "image problem" for McDonald's?; and (2) how might McDonald's modify the contextual features of their suburban restaurants in order to make a more favorable impression on their customers?

The symbolism of McDonald's design decisions appears to be far from desirable. In trying to address the discomfort issue McDonald's may be inadvertently communicating an inconsistent message. The explicit message is a public commitment to make their restaurants more comfortable. However, the implicit message is quite different. Indeed, the implicit message is strongly communicated through the salient physical features of their suburban restaurants; that message is that McDonald's builds discomfort into their restaurants. In this regard McDonald's may be the victim of a paradoxical situation. In their effort to provide a more involving environment for their customers--with intense lighting, brightly colored wallpaper, and a highly sociopetal spatial orientation, for example--McDonald's seems to have made their restaurants more uncomfortable.

McDonald's must of course make the decision as to whether the "discomfort issue" is sufficiently serious to justify the re-design of their suburban restaurants in order to make them more comfortable. At the level of value judgments, the discomfort issue may be traced back to the fact that McDonald's relies almost exclusively on a "designer's perspective" in making decisions on the physical features of their restaurants; in the past the opinions of McDonald's customers have rarely been solicited on the design and appearance of their suburban restaurants. In order to assure desirable interaction between designers and users, McDonald's might embrace a new perspective known as Social Design. Social Design, as developed by Robert Sommer (1983), requires that designers and users (i.e., customers in this instance) both have input in the design of the built environment in order to avoid "misfits between people and the built environment" (p. 10). Social Design combines the insights gained from designer/user interaction with the latest information gained from social science research.

McDonald's clearly can make the dining micro-environment of its suburban restaurants more comfortable and inviting. In fact on the international level McDonald's has revealed a truly impressive capacity to design restaurants that reflect a sensitivity to local needs and sensitivities. The negative communicative implications of their present design decisions are spelled out in considerable detail in the present study. At the same time, the detailed physical description(s) of McDonald's micro-environment that is provided in this study could be used as one source of information to make needed modifications in this fast-food micro-environment. There is good reason to believe that context can be made to communicate a more positive message to McDonald's customers.

NOTES

1. In considering the implications of the distances which characteristically separate interacting customers it is important to distinguish between the concept of density and of crowding. Density defines space in physical terms as the number of people per unit of space or, alternatively, as the amount of space available for use by each person. Crowding by contrast is a psychological concept. Crowding is that condition which exists when an individual's attempts to achieve a desired level of privacy have been unsuccessful in the sense that more social contact continues to occur than is desired (Altman, 1975).

2. Anthropometry is the science that deals specifically with the measurement of the human body, and its various parts, to determine differences in individuals and groups (Panero & Zelnik, 1979). Although both McDonald's and Burger King use anthropometric standards in determining the size of their tables and chairs, McDonald's relies much more heavily on this type of information. Burger King's decisions are more interactive in that they also put considerable emphasis on information on customer preference for the size and design of chairs and tables which is obtained via questionnaires, surveys, and open-response forms. Since anthropometric standards are based to a considerable degree on NASA research designed to determine what is the minimum amount of space required for test pilots to perform needed functions, it is not surprising that Panero and Zelnik would recommend that McDonald's and Burger King use the smallest possible chairs and tables in the interest of efficient utilization of space. These authorities do recommend, however, that four feet is the ideal separation for seated customers in a fast food restaurant. Surprisingly, McDonald's has chosen to violate this standard by designing their chairs and tables in such a way that they expect customers to typically be separated by no more than 2½ feet (Grez, 1988).

3. Smiling was operationalized as any apparent smile on the customer's face for at least two seconds, regardless of the context of the smile. Talking was defined as the percentage of customers observed at McDonald's and Burger King who were talking, when their behavior was observed, as opposed to the percentage who were not talking. Head nodding consisted of a customer's head moving in the vertical position for at least two seconds. Finally, leaning was defined as a discernible forward arch in the customer's back. Fidgeting was defined as discernible back and forth movement of one's posterior and upper thighs that seemed to serve no functional purpose other than to communicate discomfort. Extraneous foot movement occurred when it was judged that customers were moving one foot or both of their feet in a nonpurposive, excessive manner (e.g., shaking feet, tapping feet, etc.). Postural shifts were defined as any detectable movement of part of the body such as a move from an open to a closed position via shifting leg positions, crossing ankles, or an overall change in bodily position. Side-to-side movement was simply defined as shifting body weight from side to side.

4. The first author discussed the procedure of the reliability check with the reliability checker, how the author and checker would sit while in the restaurant, and how they would maintain secrecy while in the restaurant. In addition, the author spent time going over the operationalizations of the nonverbal behaviors with the reliability checker. After trial coding using diagrams of nonverbal behavior illustrated on paper, the author felt confident that the reliability checker was adequately prepared. The authors would like to thank the reliability checker, Jeff Hart. Hart is an undergraduate student at the same university where the last author is a faculty member.

REFERENCES

Alderman, R. L. (1982). How to make more money at interior design. New York, N. Y.: Whitney Communications Corporation.

Altman, I. (1975). The environment and social behavior. Monterey, CA: Brooks/Cole.

Argyle, M., & Dean, J. (1965). Eye contact, distance, and affiliation. Sociometry, 28, 289-304.

Baum, A., & Paulus, P. B. (1987). Crowding. In D. Stokols & I. Altman, Handbook of environmental psychology, (pp. 533-570). New York: John Wiley.

Bell, P. A., Fisher, J. D., Baum, A, & Greene, T. E. (1990). Environmental Psychology. Third Edition. Fort Worth, TX: Holt, Rinehart, & Winston.

Best, A. (1979). Fast food. Architectural Review, 166, 110-113.

Burgoon, J. K., Buller, D. B., & Woodall, W. G. (1989). Nonverbal communication: The unspoken dialogue. New York: Harper & Row.

Cappella, J. N., & Greene, J. O. (1984). The effects of distance and individual differences in arousability on nonverbal involvement: A test of discrepancy-arousal theory. Journal of Nonverbal Behavior, 8, 259-286.

Coker, D. A., & Burgoon, J. K. (1986). The nature of conversational involvement and nonverbal encoding patterns. Human Communication Research, 13, 463-494.

Crane, C. C. (1982). Personal places: How to make your home your own. New York, N.Y.: Whitney Library of Design.

Dannatt, A. (1988, March). Global stomach: London. Designer's Journal, 62.

Davis, T. (1984). The influence of the physical environment in offices. Academy of Management Review, 9(2), 271-283.

Dun's Business Rankings, 1989.

Ellis, P. (1986). Functional, aesthetic, and symbolic aspects of office lighting. In J. D. Wineman, (Ed.), Behavioral issues in office design. New York: Von Nostrand Reinhold.

Ekman, P., & Friesen, W. V. (1969). The repertoire of nonverbal behavior: Categories, origins, usage, and coding. Semotica, 69, 49-97.

Forgas, J. P. (1978). The effects of behavioral and cultural expectation cues on the perception of social episodes. European Journal of Social Psychology, 8, 203-213.

Gifford, F. (1987). Environmental psychology: Principles and practice. Boston: Allyn & Bacon.

Goldberg, S., & Rosenthal, R. (1986). Self-touching behaviors in the job interview: Antecedents and consequences. Journal of Nonverbal Behavior, 10, 65-80.

Gottman, J. M. (1979). Marital interactions: Experimental investigations. New York: Academic Press.

Hall, E. T. (1974). Handbook for Proxemic Research. Washington, D. C.: Society for the Anthropology of Visual Communication.

Hall, E. T. (1966). The hidden dimension. New York: Doubleday.

Harris, H., & Lipman, A. (1984). Social process, space usage: Reflections on socialization in homes for children. British Journal of Social Work, 14, 49-65.

Hill, L. (1976). Restaurant reviewers talk design...not food. Interiors, 136, 84-85.

Joiner, D. (1971). Office territory. New Society, 7, 660-663.

Kleeman, W.B. (1981). The challenge of interior design. Boston, Mass.: CBI Publishing.

Kleeman, W. B. (1988). The politics of office design. Environment and Behavior, 20(5), 537-549.

Knachstedt, M. V. (1980). Interior design for profit. New York, N. Y.: Kobro.

Konar, E, & Sundstrom, E. (1986). Status demarcation and office design. In J. D. Wineman, (Ed.), Behavioral issues in office design. New York: Von Nostrand Reinhold.

Leathers, D. G. (1976). Nonverbal communication systems. Boston: Allyn & Bacon.

Mehrabian, A. (1968). Relationship of attitude to seated posture, orientation, and distance. Journal of Personality and Social Psychology, 10, 26-30.

Mehrabian, A. (1969). Some referents and measures of nonverbal behavior. Behavior Research Methods and Instrumentation, 1, 203-207.

Mehrabian, A. (1976). Public places and private spaces: The psychology of work, play, and living environments. New York: Basic Books.

Mehrabian, A., & Diamond, S. G. (1971). Effects of furniture arrangement, props, and personality on social interaction. Journal of Personality and Social Psychology, 20, 18-30.

Mehrabian, A., & Friedman, S. L. (1986). An analysis of fidgeting and associated individual differences. Journal of Personality, 54, 406-429.

Miles, E. W., & Leathers, D. G. (1984). The impact of aesthetic and professionally related objects on credibility in the office setting. The Southern Speech Communication Journal, 49, 361-379.

Moleski, W. H., & Lang, J. T. (1986). Organizational goals and human needs in office planning. In J. D. Wineman, (Ed.), Behavioral issues in office design. New York: Von Nostrand Reinhold.

Nespoulous, J. L., & Lecours, A. R. (1986). Gestures: nature and function. In J. L. Nespoulous, P. Perron, & A. R. Lecours (Eds.), The biological foundations of gestures (pp. 49-61). Hillsdale, NJ: Erlbaum.

O'Hair, D., Kreps, G. L., & Frey, L. (1990). Conceptual Issues. In D. O'Hair & G. Kreps (Eds.), Applied Communication Theory and Research (pp. 1-22). Hillsdale, NJ: Lawrence Erlbaum.

Panero, J., & Zelnik, M. (1979). Human dimensions & interior space. New York: Whitney Library of Design.

Pettigrew, L. A. (1988). The importance of context in applied communication research. The Southern Speech Communication Journal, 53, 331-338.

Pile, J. (1979). Decorating your office for success. New York: Harper & Row Publishers.

Preston, P., & Quesda, A. (1974). What does your office say about you? Supervisory Management, 19, 28-34.

Restaurants and Institutions (1988, December 23), p. 48.

Rapoport, A. (1982). The meaning of the built environment: A nonverbal communication approach. Beverly Hills, CA: Sage.

Sime, J. D. (1986). Creating places or designing spaces? Journal of Environmental Psychology, 6, 49-63.

Smith, F. K., & Bertolone, F. J. (1986). In D. C. Hines (Ed.), Bringing interiors to light: The principles and practices of lighting design. New York: Whitney Library of Design.

Smith, L. J., & Malandro, L. A. (1985). Courtroom communication strategies. New York: Kluwer.

Sommer, R. (1966). Man's proximate environment. Journal of Social Issues, 22, 59-70.

Sommer, R. (1969). Personal space: The behavioral basis of design. Englewood Cliffs, N. J.: Prentice-Hall.

Sommer, R. (1974). Tight spaces: Hard architecture and how to humanize it. Englewoods Cliffs, NJ: Prentice-Hall.

Sommer, R. (1983). Social design: Creating buildings with people in mind. Englewood Cliffs, NJ: Prentice-Hall.

Stacks, D. W., & Burgoon, J. K. (1981). The role of nonverbal behaviors as distractors in resistance to persuasion in interpersonal contexts. Central States Speech Journal, 32, 61-73.

Sundstrom, E. & Sundstrom, M. S. (1986). Workplaces: The psychology of the physical environment in offices and factories. Cambridge: Cambridge UP.

White, A. G. (1953). The patient sits down. Psychosomatic Medicine, 15, 256-257.

Woudhuysen, J. (1986, April). McDonald's--Burger Kings. Designer's Journal, 58-62.

Yee, R., & Gustafson, K. (1983). Corporate Design. New York, N.Y.: Whitney Communications Corporation.