Research Support School of Business Administration December 1993 NAVIGATING BY ATTIRE: INDIVIDUALS' USE OF DRESS IN ORGANIZATIONS Working Paper #721 Anat Rafaeli Hebrew University of Jerusalem Jane Dutton C.V. Harquail Stephanie Lewis University of Michigan

I i /~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ L L i i I j I i c t I

NAVIGATING BY ATTIRE: INDIVIDUALS' USE OF DRESS IN ORGANIZATIONS Anat Rafaelil School of Business Administration Hebrew University of Jerusalem Mt. Scopus Jerusalem ISRAEL Jane Dutton Graduate School of Business University of Michigan C.V. Harquail Graduate School of Business University of Michigan Stephanie Lewis Department of Psychology University of Michigan December, 1993 Comments Most Welcome Please use electronic mail to communicate comments and suggestions to Anat Rafaeli MSARAFA@PLUTO.CC.HUJI.AC.IL tPortions of this study were completed when Anat Rafaeli was a visiting faculty member with the University of Michigan. We wish to thank all the participants in the study who gave so generously of their time. We thank Susan Ashford, Art Brief, Bill Foraker, Mike Pratt and Julie Younglove for comments in earlier drafts of this paper.

Dress in Organizations 2 ABSTRACT Dress in organizations is visible and ubiquitous, yet understudied. In this study we examine the dynamics surrounding the dress of administrative employees in an academic institution. Building on extensive qualitative data, we suggest that dress is an important behavioral and cognitive tool that enables individuals to enter organizational roles, read and understand the organization's structure, and adapt to their context. We draw on and link these themes to the organizational literature on role theory, cognitive sensemaking, and adaptation theory. The findings enhance our understanding of dress in organizations, and highlight the contributions that attention to dress can make to members and students of organizations.

Dress in Organizations 3 An age-old question for organizational scholars has been, "how do individuals fit into the organizational context?" Some researchers have focused on how selection processes align individuals and organizations (Schwab, Rynes and Aldag, 1987). Others have looked at how organizations and individuals achieve alignment through the process of socialization (Feldman, 1976; Chatman, 1991). This paper explores how the everyday behaviors of organizational members —in particular, members' dress behaviors —align them with the social group called the organization. On one hand, this is a qualitative study of the use of dress by organizational members. It builds directly on the work of Rafaeli and Pratt (1993) and Pratt and Rafaeli (1993) to explore how individuals' dress behaviors relate to organizational functioning. On the other hand, this is a qualitative study of how small micro-acts of individuals (e.g., what members wear at work) create, sustain and alter more macro-organizational structures and processes (e.g., the organization's division of labor, the creation of its image to outside constituents). Previous work on organizational dress has primarily focused on the organizational-level dynamics of dress. In particular, Rafaeli and Pratt (1993) address the question, "Is dress relevant to organizational behavior?" by arguing that dress both reflects and creates a variety of organizational dynamics. They offer a framework that identifies three dimensions of organizational dress: attributes, homogeneity, and conspicuousness. Attributes of dress comprise the physical characteristics of clothing —color, material, and style. Dress homogeneity refers to the extent to which there is variance in the dress of organizational members. Organizational dress may be considered random heterogeneity (there is a great deal of variance in dress among organization members), stratified homogeneity (the dress within subgroups is similar, yet the dress between subgroups varies a great deal), or complete homogeneity (all members of the organization dress similarly). Finally, dress conspicuousness refers to the extent to which organizational dress varies from dress outside the organization. Within this framework, Rafaeli and Pratt propose that organizational dress reflects the internal values and structure of

Dress in Organizations 4 the organization, and the organization's relationships with the external environment (particular cultures or institutions). In this study, we complement the earlier framework by exploring the manner in which dress is engaged by individuals in organizations. Dress is as visible and ubiquitous in organizations as it is in all other aspects of human society -- organization members cannot escape some interaction with dress as part of their membership. In particular, individuals psychologically engage with dress in two ways. First, since individuals make decisions about what to wear and then act to implement those decisions, they engage with dress as actors. Second, individuals see and interpret (often unconsciously) what others are wearing, and thus also engage with dress as observers. Organizational dress, therefore, involves both dress behaviors -- the choice and wearing of dress -- and dress cognitions -- observations of others' dress and the use of these observations as cues in making sense of one's context. Given our interest in individuals' engagement with dress, the research question at the outset of our investigation was: How do individuals use. and relate to, the patterns of dress in their organizational context? Our goal was to examine closely and inductively individuals' perceptions and behaviors with respect to dress at work. We conducted an in-depth qualitative study of the dress of administrative employees. The results suggest that dress is an important behavioral and cognitive instrument that organizational members use (consciously and unconsciously) to navigate and relate to their organizational context. More specifically, dress assists individuals in engaging with, and executing, various roles. Dress also serves as an important contextual cue that helps each individual to understand his/her place in the organization and the relative place of others. Finally, dress is a means to aid individuals adaptation to their context. Thus, from a functional standpoint, dress serves as an aid in role making, an organizational compass, and an adaptive tool. These three themes illustrate and link three important streams of research, respectively -- role theory, cognitive sensemaking, and adaptation theory. But before we elaborate on these theoretical linkages to our inquiry, we first turn to the methods we employed.

Dress in Organizations 5 METHOD Overview During the winter and spring of 1991, we collected data from the administrative support staff in the business school of a large midwestern university. The university is comprised of 18 schools and colleges, and employs more than 22,000 faculty and staff. There are over 36,000 students, of whom approximately 70%-are undergraduates. The business school employs 138 faculty members and 137 support staff. Annual enrollment in the business school is approximately 2,500, of whom 21% are undergraduates. The business school also has an active executive development program which relies mostly on the same faculty, yet operates in a separate wing of the school, and offers more than 40 executive programs and management seminars each year. We initiated this study because of our interest in dress and received the support and formal endorsement of the dean and the senior staff administrator in the school. The stated goal of the study was to explore the dynamics that surround and influence the dress of administrative employees in a professional organization. Three factors encouraged us to focus on the administrative staff. First, this group is central for maintaining a smoothly operating organization. Second, this group is structurally central, as the administrative staff interacts with all levels and types of constituencies in the organization. Third, there is a relative lack of attention to studying administrative support staff despite their organizational centrality (O'Leary & Ickovics, 1990). Thus, we hope to lend insight into the everyday work of these employees by examining everyday acts of dress. Data Collection We used three means of data collection:

Dress in Organizations 6 Semi-structured interviews. We conducted in-depth, semi-structured interviews with a stratified random sample of the school's administrative staff. We selected the sample of twentytwo informants (20 women, 2 men) from the population of full-time, permanent employees holding administrative responsibilities. Informants' positions ranged from clerk, receptionist, and secretary to unit director, manager, and associate dean. Informants were employed in various departments in the school including the library, the dean's office, faculty support, student admissions and services, computing services, document processing, and executive education. We stratified the sample in a manner that guaranteed representation of the various functional and hierarchical distinctions in the school. Average tenure of informants in the organization was 9 years, ranging from 1 to 24 years; average tenure in the present position was 4 years, ranging from 3 months to 11 years. The sample also matched the distribution of the administrative support staff in terms of gender (94% of the staff is female). Because of the gender homogeneity of the administrative support staff, we do not believe it is feasible to generalize our findings to men's use of dress in organizations. The head administrator notified all of the administrative employees in the school that the study was being conducted. We personally contacted all selected informants and asked for their consent to be interviewed. Every person we contacted agreed to be interviewed, and interviews took place in their offices or in a school lounge, lasting between 45 minutes and 3 hours. We recorded and transcribed all but two interviews; one informant refused to be taped, and the tape recorder malfunctioned during another interview. For interviews not taped, interviewers recorded detailed notes. Interviewers showed informants an interview schedule at the beginning of each interview and encouraged them to answer the questions listed on this schedule (see Appendix). All of the informants answered the full set of listed questions, and we encouraged elaborations. Such elaborations occurred most frequently when we probed for particular responses. Probing informants to elaborate on their answers was useful both in establishing rapport and in gaining an in-depth understanding of the dress issue.

Dress in Organizations 7 Observations. We collected two types of observational data. Collecting these data was relatively easy since we were all faculty members or students in this organization at the time of data collection. First, following each interview, we took detailed notes about the dress and appearance of the informants, their work stations, and work environment. In addition, we conducted systematic observations of the business school. We divided the school into four geographical areas. A researcher visited each area for the explicit purpose of detailing the attributes of those areas (e.g., layout, materials used, decorations), and noting the dress of employees present in these areas. We recorded detailed notes following these observations. Second, during the course of the study, we observed major events in the school that we thought might pertain to the study. Members of the research team attended two types of events: 1) events intended for the administrative staff, such as the annual lunch hosted by the dean; and 2) events planned and managed by the administrative staff, such as talks given by visitors or various workshops provided for students. Research team members took detailed notes on these events, including but not limited to the appearance of the participants at the event. During the course of such events, we talked with various participants. Such conversations were not treated as formal interviews, although they did occasionally offer helpful insights. Documents. We acquired and coded a range of documents that we thought might prove relevant to dress, including formal school policies, training manuals, and internal newspaper articles pertaining to dress in the business school. Data Analysis We followed the suggestions of Glaser and Strauss (1967), Miles (1983), and Strauss and Corbin (1990), by analyzing the data in three distinct, though tightly interrelated, phases. Phase I: Searching for Themes. First, we scanned all the data and searched for dominant themes. Scanning was first done by individual members of the research team alone,

Dress in Organizations 8 and then discussed during weekly meetings, where potential themes were raised and discussed. The themes we generated ranged from nearly intuitive, fundamental assertions (e.g., "dress is used to form a first impression," and "dress is influenced by weather") to a number of more intriguing, unanticipated themes (e.g., "dress helps individuals identify different parts of the organization," "dress is used to convey trust and knowledge to customers/vendors"). Phase II: Developing a Coherent Conceptual Framework. Next, we searched for a conceptual structure that would integrate the themes in a coherent manner. Our goal in this phase was to articulate a parsimonious yet interesting framework that would help us understand and describe the dynamics associated with dress for the employees of the business school. This phase required extensive brainstorming and recurrent formal and informal conversations among the research team. These conversations often involved drawing schematic representations of potential conceptual frameworks that depicted relationships among various themes. Initially we conducted these conversations in face-to-face meetings; later conversations involved exchanges of written and electronic messages. As this phase unfolded, three questions guided our assessment of alternative conceptual frameworks: 1) Does this framework explain a large number of themes? 2) Is this framework the most parsimonious one for explaining these themes? and 3) Is this framework helpful and insightful for organizational scholars? In the crux of this phase, we identified and articulated the thesis proposed in this paper: For the individuals we studied, dress offers a mechanism for more smoothly executing the various roles they are required to play within the organization. for organizing their understandings of their place of employment and for adapting themselves to this social setting. This framework conceptually integrates a large number of themes identified in the data, and provides valuable insights for organizational scholars. Once we identified the conceptual framework, we revisited the list of themes generated in Phase I of the data analysis. We systematically considered whether, and how, each of these

Dress in Organizations 9 themes fit into the framework. This process produced a theme list that was relevant to the conceptual thesis that we wished to advance. Some themes, although evident in the data, could not be linked to the conceptual framework proposed, and therefore are not included in this paper. Phase III: Coding Data into Themes. In the third and final phase of data analysis, we systematically coded and categorized all the data that had been collected against the themes retained in Phase II. We conducted the coding electronically, following Holsti's (1969) and Rosengren's (1981) recommendations for content analyses. At the end of this process we had produced a new data set that consisted of quotes from the original data categorized by themes that were integral to the conceptual framework. The remainder of this paper draws from this database and describes this framework, and also relates our efforts to prior theoretical developments. INDIVIDUALS' USES OF DRESS IN THE ORGANIZATION The formal policy regarding dress in the organization we studied was extremely openended. A university handbook simply stated: Each employee is responsible for presenting a personal appearance which recognizes the need for good grooming and neatness in order to avoid distracting others and to comply with safety standards. This very broad code left many employees unaware of the existence of any formal policy regarding dress. No one systematically informed employees of the code, and supervisors provided new applicants who asked about dress standards with very general guidelines such as "you have to look presentable" or "it is important that you look good." The formal policy regarding dress was sufficiently ill-specified that a senior manager told us: "When we interview people for a job we always make it clear that there is no dress code." Thus, the dress of employees in this organization was generally not a mandated response to formal organizational requirements.

Dress in Organizations 10 Despite the lack of a strict, formal dress code, our data revealed systematic patterns of dress in the organization. Three themes in the data indicated that dress in the organization was not a random phenomenon: 1) there were strong norms among informants about appropriate and inappropriate dress; 2) dress in the business school was noticeably different from dress in other parts of the university; and 3) there was evidence of stratification by dress in the organization. First, strong norms existed in the organization regarding appropriate and inappropriate dress. The dominant dress norm could be captured by the label "professional" or "business-like" (Molloy, 1975; 1977; Rafaeli & Pratt, 1993). Dress that is business-like is typically conventional, conservative, lacks ornamentation, and uses subdued or placid colors (Lurie, 1983; Molloy, 1977). For example, one informant explained the dress of the school's administration: "Most of the people in administration typically have always worn suits or jackets." Another informant, in characterizing her own attire, noted: "What I'm wearing today is what I call business attire, which is fairly conservative, plain jacket, plain skirt, casual, but not too casual." Dressing "professionally" also meant not wearing certain types of clothing. As one informant told us: And they expect your appearance to be on the professional business-type level. No flashy colors, no sexy dresses, no shorts or drop shoulder shirts. In addition, the existence of dress norms was evident in the shared prescription against wearing jeans (except occasionally on Fridays). Without being prompted, all informants mentioned that employees of the business school should not wear jeans. Several informants noted that many professors did not maintain this standard, despite their common membership in the organization. The shared sentiment was that the violation of this norm is inappropriate and damaged the organizational image, even when-done by faculty. As noted by one informant: Some faculty wear jeans or shorts. Especially the young ones. But they should know better. What does it make this place look like? Second, the norms to dress professionally distinguished business school members from those in the rest of the university, who held norms for much more casual dress. One could easily

Dress in Organizations 11 classify individuals in the halls of the business school as members or non-members of the business school community based on what they wore. Thus, dress in the business school could be classified in Rafaeli & Pratt's (1993) typology as conspicuous, in that it made the organization stand out from its immediate environment. The data suggested that informants were readily aware of the observable distinctions in dress between the business school and the rest of the university. To illustrate, one informant remarked: [People in other parts of the campus] comment that even our messengers —our student messengers —they can always tell the business school messengers, they don't even need any introduction at the beginning of the year. It's, 'Oh, you must be from Business Ed.,' because of the way they are dressed, and their hair. I think the word is, their 'scrubbed appearance'. Third, while the business school members shared an overall norm for professional dress, our data also revealed systematic dress stratification in the organization. Following Rafaeli & Pratt's (1993) classifications, the dress pattern can be described as stratified-homogeneity. That is, employees in some units with higher social or formal status consistently dressed in more formal styles than employees in other departments. Informants were aware of this stratification. This awareness was evident in simple comments like "people in the dean's office are more dressed up than in other parts of the school." More astute observations were also apparent: When I walk around the school and I look at someone, I can usually tell, by the way they're dressed, who they are. An MBA? A secretary? A faculty member?...the students dress the most casual, but... I think they dress a little bit more professionally here in the business school [than elsewhere in the University]. Our analyses of the data suggested that these systematic patterns in organizational members' dress were not incidental. Rather, we propose that these tacitly agreed upon dress norms are reinforced by the fact that they are helpful and propitious to the individuals involved. We now turn to the ways in which dress norms aided school administrative staff.

Dress in Organizations 12 Three Mechanisms of Dress The data suggest that dress enables organization members to relate to the organization in three ways: 1) dress enables individuals to execute particular roles associated with their job; 2) dress enables individuals to cognitively comprehend the organization as a complete and complex system; and 3) dress enables individuals to adapt themselves and their environment to increase their "fit" with the people and processes surrounding them. These three uses of dress pertain respectively to three streams of theory and research on social psychological dynamics in organizations: role theory, cognitive sensemaking, and adaptation theory. These three literatures together capture the types of relationships that exist between the organization and its members. Role theory argues that individuals perform roles assigned to them by their employers (e.g., Katz & Kahn, 1978; Turner, 1978), and thus refers to the tasks and responsibilities \ imposed upon individuals by their organization. Conceptualizations about individual sensemaking propose that individuals rely on various cues in their environment to develop a coherent understanding of the environment. This stream of literature describes the cognitive efforts invested by individuals to comprehend the organization (Louis, 1980; Isabella, 1990; Weick, 1979). Finally, adaptation theory suggests that individuals read their environment and adapt their behaviors to fit into this environment (e.g., White 1959/1974; Ashford & Taylor, 1990). An adaptation perspective on organizational dress combines the role taking and sensemaking views by arguing that individuals use cognitions first to understand their context and then alter their behavior accordingly. Our integration of these three perspectives, and the role that dress plays in each, is summarized in Table 1. As suggested in Table 1, we see two dimensions that help describe the manner in which dress is used by individuals: Dress can be described according to the extent to which it involves concrete behaviors or acts (behavioral focus) that are observable by others, versus the extent to which it involves cognitive processes (cognitive focus) that occur primarily in individuals' minds, making them less observable by others.

Dress in Organizations 13 Insert Table 1 About Here As implied in Table 1, the use of dress in organizations can vary in the extent to which dress influences individual behaviors and cognitive processes. When individuals wear particular clothes to better perform their assigned roles, then dress behaviors are focused on behavioral acts which bring about desired outcomes for-members. In this context, dress is an instrument used to assume a particular role. In contrast, when individuals use dress patterns around them as a cue to notions such as division of labor or organizational structure, then dress has a cognitive focus, aiding in the development or clarification of members' maps of the organization. Relying on this point of view, we propose that dress is an important cue for individual sensemaking of the organization. Combining these two foci, cognitive observations of dress coupled with dress behaviors enable individuals to develop a better fit with their context by changing themselves and/or their context. Thus, as suggested in Table 1, individuals' use of dress to enhance adaptation to the organizational context engages both a cognitive reading of the organization and behavioral actions that facilitate an individual's relationship with the organization. We propose that dress is a tangible artifact that plays a role in creating a relationship between individuals and their organizational environments —dress enables individuals to act according to the requirements of the context in which they operate, to read and understand this context, and to enhance the fit between themselves and the context. We describe below the three social psychological perspectives which inform each view of how dress connects individuals to their organizational contexts. Dress and Role Taking Individuals have multiple roles available to them in the course of their participation in social systems (Katz & Kahn, 1978; Turner, 1978; Swann, 1987). All roles have scripts or prescribed ways to act associated with them (DiMaggio, 1991), and individuals are expected to

Dress in Organizations 14 follow the appropriate prescriptions to fulfill their roles (Katz & Kahn, 1978). Katz and Kahn further argue that organizational members respond to the expectations of their role with "role behaviors" (1978: 195). As a beginning, we argue that dress behaviors - choosing and wearing clothes -- are important role behaviors. As role behaviors, dress behaviors reflect and influence role choice in two distinct ways. First, dress is an expressive role behavior, in that dress is an expression of the meaning of one's perceived or objectively-assigned organizational role (e.g., if one wears a business suit it means one's taking on the role of "professional"). Second, dress is an instrumental role behavior. Individuals choose certain dress to aid in effective role execution. For instance, a woman working as a sales clerk may choose to wear flats instead of heeled shoes because her role requires her to stand up for several hours at a time. Similarly, the 'blue collar' and 'white collar' shirts that we now use to stereotype social classes have their origins partially in instrumental role behavior: The working-class shirt color arose from the limited textile and dyeing technologies of the early industrial era, inasmuch as early machine-made shirt fabrics able to withstand the frequent launderings and chemicals required to remove the grime of hands-on industrial labor would not bleach completely white, and blue dye concealed stains more effectively. Individuals may also use dress behaviors as a mechanism to actually "take on" a particular role at a given time. This proposition is consistent with literature regarding the role of visible, external cues as "role signs" to identify a particular role to oneself and others (Turner, 1978; Kanter, 1972; Goffman, 1959/1963; Banton, 1965). The concept of dress as a "role sign" contains implicit within it the idea of choice of role (using dress to signal role implies that one would signal change of role by change of dress, and thus perhaps change role by changing dress). Our data specifically suggest that dress is an important mechanism for individuals with regard to the expression and execution of four different roles: the role of an employee, the role of an authority figure, the role of a boundary spanner and the role of organizational image bearer. Engagement in these roles, in turn, triggers certain scripts or ways to behave and certain

Dress in Organizations 15 thoughts or ways of thinking (DiMaggio, 1991). While these roles may be occasionally interrelated, they are sufficiently distinct at a conceptual level to stand alone in our analysis. Table 2 offers examples of data supporting and illustrating this assertion. Insert Table 2 About Here Dress and Role of Employee. A salient theme in the data involved informants' categorization of their attire as "organizational" and correspondingly as "appropriate" versus "inappropriate" for work. The first quote in Table 2 illustrates this theme. In this quote, an informant explicitly notes that although she has two sets of clothes, one can be worn for work while the other one cannot. Similarly, in explaining how she decided what to wear to work on a particular day, another informant noted: "I have this set of clothes I can wear to work, and I just pulled this outfit from there." Two other informants expressed this role taking in the following way: "I wear work attire; not something that I'd wear to a bar," and "We wear work attire. Not something that you wear when you go out on a date, or go out with your friends." By classifying their clothes in this way, individuals distinguish and categorize themselves. They are first and foremost employees when they wear clothes from their "work appropriate" wardrobe at work. The wearing of non-work clothes distances one from this social identity. Furthermore, the process of dressing for work appears to be entwined with the process of planning particular work roles that individuals expect to perform. In response to our question "how did you decide what to wear this morning?," informants repeatedly referred to particular roles that they expected to perform. Meetings generated a particular dress style, attending a conference another, making a presentation a third, while "just sitting at your desk" was likely to generate a fourth, and least descriptive, style of dress. In addition to its expressive and instrumental purposes, dressing with regard to the roles one expects to perform on a given day can be understood as a method of "autocommunication" (Broms & Gahmberg, 1983).

Dress in Organizations 16 Autocommunication is a mode of cueing oneself to behave in a certain way. Not only will a person dressed for a role express that role to others, she will remind herself of that role throughout the course of the day. In other words, dress not only facilitates entry into the role of an employee, it also appears to offer individuals a vehicle for switching on a particular behavioral script and cognitive orientation that fits a role variation appropriate for the tasks of the day. This dynamic was eloquently summarized by one informant who explicitly told us: "I dress in the morning for whatever is going to go on that particular day. In my mind I go through my calendar and dress accordingly." Taking the role of "employee" also involves simultaneously letting go of other identities that are inherent to one's self-concept (Gecas, 1982). In wearing attire that has been designated "organizational," individuals temporarily shed or hide other, non-organizational roles and adopt the role of employee (cf. Rafaeli, 1989b). In a sense, a process of identity compartmentalization occurs (Turner, 1978), and part of the individual is closed off when the role of employee is entered. One informant explicitly summarized a sentiment expressed by many: "There's a whole part of my personality that's not expressed when I'm dressed for work." Thus, we seldom observed stereotypically feminine colors (lavender, pink) and styles (frilly dresses, colorful jewelry), although our sample was predominantly female. Informants characterized their dress as "business-like," which fits the particular work organization rather than feminine, which would fit them as a group (cf. Wolf, 1990; Rafaeli and Pratt, 1993; Lurie, 1981). These dress patterns helped to articulate that femininity was outside the individual's role in this organization. Several informants explicitly voiced this theme. For example: "I can't wear my Laura Ashley dresses to work. I look like a school girl. People don't recognize that look as an eclectic taste. I'd feel silly." Such compartmentalization of one's work role from the other components created a sense of incompleteness for some of the study participants: "[If I were to dress like the real me] I would probably wear some wild earrings. It would say that I am fun loving." Consistently,

Dress in Organizations 17 informants welcomed opportunities to break the implicit dress code, and to be "themselves" in their dress. The ritual of coming home from work involved "changing into my jeans and sneakers as soon as I can." Moreover, various organizational rites and rituals were viewed as an opportunity to violate implicit dress norms. Fridays were informally designated as "jeans days," implying an opportunity to "dress down," and a Christmas party or dean's luncheon were repeatedly mentioned as welcomed opportunities for "dressing up." Dress and the Role of an Authority. Creation of the image that one has authority relies on the effective management of impressions in interactions with others. Our data consistently suggested that informants who felt that conveying authority is an important component of their organizational role also felt that dress is a useful mechanism for communicating this role component. In other words, dress helps some organizational members enter an authority figure role. Obviously, dress is not the only mechanism for obtaining and asserting authority. Yet, as illustrated in Table 2, informants felt that dress assisted them in generating authority-based trust and respect from others in the organization. Informants identified particular dress styles as more effective at accomplishing certain aspects of the authoritative role. Specifically, they perceived more formal, "business-like" dress as helpful in conveying the responsibility of an authoritative role. In contrast, they considered classic feminine dress as undermining authority. As Harragan (1977: 339) noted: A woman who wants to manage affairs, control subordinates, and exert authority must avoid any kinds of dresses which portray her as weak or indecisive. Any taint of the "little girl" look is anathema - pinafores, ruffles, bows, cut prints, flounce skirts, clinging fabrics, or distinctively "feminine" frills will contradict any effort to be viewed as forceful. In this vein, business-like dress was seen as helpful in "drawing the line" in convincing others that one is "knowledgeable," or "efficient," and in asserting that one is "someone with respect," so as to claim that "my answer is the final answer." All these examples suggest that dress can help formulate the impressions necessary for inducing compliance from other persons (for scholarly concurrence, see Fussel, 1983; Goffman, 1967; Joseph, 1986; Lurie, 1981;

Dress in Organizations 18 Molloy, 1975; Rafaeli and Pratt, 1993). Dress appeared in our data to be one of the easiest means to signal the possession of an authoritative role in organizations (a finding predicted by Rafaeli and Pratt, 1993). Dress may be an effective tool for evoking authority because it is a powerful signal to both self and others that one has status and power in an organization. This may explain why in an empirical study of nurses in a rehab unit the struggle over a dress code was an important way for unit members to debate, contest and resolve issues regarding who had power in the unit (Pratt & Rafaeli, 1993). One informant relayed to us her own surprise about the power of dress as a vehicle for taking on more authority in the organization: When I first took this job, I was real leery about going into a management position supervising people. Within the next 6 months I tried different things. I changed some policies and procedures, that kind of stuff.... But still wasn't comfortable. So I started changing my dress, wearing more suits and so on... I eventually noticed the difference... little by little. For this informant, wearing certain types of clothing signaled to others and to herself that her authority position had changed. Her behavior and the behavior of others changed accordingly. The experience of this informant mirrors the findings of other researchers who suggest that dress plays a major role in delineating and communicating the role of "boss," making it easier for others, especially other women, to take the role of subordinate (a study by O'Leary, 1987 cited in O'Leary and Ickovics, 1990: 41). Dress and the Role of a Boundary Spanner. A third role that individual dress behavior appeared to facilitate is that of boundary spanner. Perceptions of one's responsibility as a representative of the organization to outside clients or customers guided dress behaviors. Several themes evident in the data support this proposition. First, informants generally saw their dress as conveying attributes of themselves which could be used to infer their probable relationships to others: In my dress, the first impression I would like to give to somebody who didn't know me is that I care about myself which means that I'm going to care about someone else or that someone can trust me to do something for them, no matter what it is.

Dress in Organizations 19 For this informant, care in dress signaled competence and potential for care of others. Second, informants repeatedly noted their responsibilities toward the organization's customers when explaining the way they dressed. Thus, they saw their role as being defined by the customers with whom they interacted. To illustrate, an informant from the executive education center noted: "We are a service center. Like a hotel or a restaurant. People pay a lot of money to come here. So we should treat them nicely... we should be dressed up." A faculty secretary noted: "It's the market of your job [that is important]. My clients are faculty and students so I can dress more casually." Signals involve at least two participants -- sender(s) and recipient(s) -- and require some commonality in perceptions/understandings to be effective. Consistent with previous research about employees in boundary spanning positions, informants' assumptions regarding dress were tied to their assumptions about clients (cf. Adams, 1976; Rafaeli, 1989a). For example, informants who worked in the executive education center repeatedly noted that because they interact with corporate executives they should dress in corporate business styles: We have clients, executives. [So] most of the people here dress well. They should. Anytime one is around participants, they should be more dressed up than if only seeing students. We deal with TOP executives of the TOP corporations in the country. Take BIG prices. The way employees appear is a part of what these corporate clients get. In contrast, an informant who worked in a student service center noted how the fact that this unit deals primarily with students determined her dress: We have students coming in all day. I don't want to dress too formally, because it will turn them off. I try to dress in a friendly, casual manner to welcome them. This quote reinforces, the point made earlier that individuals use dress to enact a type of relationship with others, but it also emphasizes the importance of shared, often implicit, assumptions of the "messages" of dress in both sender and recipient.

Dress in Organizations 20 This second aspect of dress and boundary spanning may be considered as a special case of a third, more general, theme: informants considered dress a means of establishing good rapport with target persons, a quality which has been identified as an essential component of effective boundary spanning (Adams, 1976). To illustrate, an informant who frequently interacts with students noted: I don't want to dress too nice, because there are other people that I have to help. You know, students will come in their slacks, and I don't ever want to give them the impression that I'm superior over them. I want to look and know that I am kind of at their own level, that they are not intimidated, so that they can use our office as a resource. If I sat behind that desk, I think, in a suit and a shirt with a bow tie or scarf around my neck, I think I would intimidate some of the students so that they wouldn't feel comfortable to come in. The student placement center represents an interesting unit in terms of its boundary spanning functions. The office primarily serves students and is located in a hall used almost exclusively by students. Despite the high utilization of the office by students, the dress of employees was significantly more formal and "business like" than in other student-service offices in the business school. However, this pattern is consistent with our thesis: In order to perform their function, employees in this office had to interact with corporate representatives as well as students. Indeed, the student placement center is a key mediating unit, linking the students who are looking for a job with corporate representatives seeking new employees. Thus, the targets of boundary spanning included students as well as corporate representatives. As noted by an informant, both students and employees of this office adapt their dress to meet the standards set by the corporations: Corporations do a lot of college relations business [here]. You never know who's dropping into my office, so I'd never want to be totally relaxed [in my dress]. I'd never come to work with slacks. Even summer outfits seem too informal.... It's a riot to see the students. One day in jeans, the next day in a suit but with the same back pack. [But] I'm very sympathetic to their situation. It's a mixed environment, where one day they are.talking to their friends, and the next to a corporate representative.

Dress in Organizations 21 One additional element in the data further supports the boundary spanning dynamics of individuals' dress: In the summer and sometimes on Fridays, when few corporate visitors could be found in the halls of the school, dress in the school appeared significantly more casual. Apparently, during these times employees felt that they were working only with other (internal) members of the organization. As a result, the boundary spanning roles and functions were relaxed, allowing employees to be more casual. One informant noted "then it's like we're one big family on vacation." Boundary spanning functions were most salient when an employee had to represent the organization to members of outside groups. Such presentations to students, corporate representatives, or executive education participants required careful dress planning. One informant made the following comment: I prepare what I wear before a presentation. In terms of who I will be with and what will they be wearing. It's the goal of integrating. It's the goal of mixing. It's not the goal of wearing a navy blue suit to impose or oppose. It's wearing a blue suit to blend. Indeed, recurring observations of employees in various public functions revealed relative homogeneity of dress among participants and between participants and the speaker. Employees' dress in presentations to corporate representatives, for example, is significantly more formal than the dress of the same individuals on other days. This was the case even when corporate representatives spoke to a forum of students with the staff as passive participants. The qualifier "formal attire required" often publicly accompanied invitations and announcements of such events. Although this qualifier was formally intended for students, staff members also appeared dressed more formally at these events than at other times. Apparently attendance at these events is considered a form of organizational representation, calling for dress that meets the standards of those attending, including the presenters and the students. Adopting a boundary spanning role did not always require one to directly interact with outsiders. Rather, the boundary spanning function of a whole unit could affect how unit members dressed. Even informants from the back offices in the executive development center

Dress in Organizations 22 noted the obligation they felt to dress in a business like fashion: "The business we're in calls that we dress up even if meeting clients or executives isn't true for our own job." Employees in all units in this business school take extra care when they know that they might be in direct contact with representatives of organizations outside the school, especially those representing the corporate world. Thus, as noted by one informant, receptionists located in offices that only occasionally interacted with the corporate environment still considered the visits of corporate representative to the office as an occasion when it was essential to dress up: I write down in my calendar if there's going to be a big meeting in the office, or if [my boss] is meeting a high executive and I try to dress in a nice skirt or a blouse or something. That's the only time that I am really cautious of what I wear. In short, dress was a mechanism that facilitated the performance of boundary spanning functions by helping an individual break down barriers between organizational members and outside groups. Dress and Role of Image-Bearer. In addition to the individual roles that members hold, each business school employee is a member of the business school, and thus represents it to outsiders. Therefore, a part of each member's role is to convey the organization's values and image to outsiders. Our data revealed that there are two components to this assertion: Individual dress behaviors reflect the organization's values to outside stakeholders, and are guided by the desire to comply with the organizational goal of "professionalism". First, our analyses suggested that employees assume an association between the dress of organization members and perceptions of outside constituents. This assumption leads employees to wear styles of dress that will create a good impression for outsiders. The external image that the organization would like to project determines, in part, what colors, materials, or styles of dress employees wear. Informants were aware of the assumed link between employees' attire and organizational image. General comments such as, "If you go by and see someone sitting there that's got on some ripped old thing or whatever...I think it's a reflection on the whole school", illustrate this

Dress in Organizations 23 link. Some informants expressed even more specific assumptions that linked particular dress styles to a particular organizational image: We as a school have to look good to the outside. I'm here in the front office; anyone walking into the school sees me. I want them to know that in the business school we're friendly. So I dress nice. But not too nice which would be cold. Such observations were not limited to informants who see themselves as boundary spanning "windows" into the organization. Informants from all parts of the organization, including departments that were in inconspicuous locations, mentioned a presumed association between their personal appearance and the organization's image to outsiders. The most common image that informants used to describe what they thought their dress should convey for the organization was "professionalism." One informant summarized the prevalent perception: "This is a professional organization. So we should all look professional. And I dress in a way that will look professional." Almost all informants mentioned that employees should be dressed "professionally" or in a professional manner. The data indicated some variations on the interpretation of this term, but in general, the term seemed to be used as a shortcut code to capture a sense of the overarching organizational goal of being the best. Staff at all levels of the organization echoed this sentiment. At a staff luncheon, the dean explicitly noted this theme: Our goal is to succeed, and be the best. The very best. In order to do that we need to... do an excellent job for students and for companies... we need to develop an excellent professional environment. That includes serving their learning skills but also includes common standards of quality, helpfulness, courtesy, and service, because it's a part of the total image that people have. According to this organizational leader, an excellent organization is also a professional organization. Our data revealed that employees used this assumption in their perceptions of and decisions about dress: They selected attire that appeared professional. In this way members' dress behavior played a role in sustaining the desired organizational image. In sum, dress is an important behavior that individuals use to enter into and fulfill their organizational roles -- employee, authority, boundary spanning, and image-bearing roles. Not

Dress in Organizations 24 only do individuals use dress to distinguish their organizational roles from their nonorganizational roles, but dress is also useful in distinguishing between various organizational roles. Dress is also helpful in enabling individuals to perform their roles more effectively. Dress as a Cognitive Tool: Identifying Parts of the Organization Through Dress In addition to the behavioral focus implied by the relationship between dress and organizational roles, we found that members also used dress as an informative cue in comprehending the organization. Modem organizations are complex systems that often seem ambiguous and amorphous. Because of this complexity, various mechanisms are necessary to ensure individual members' understanding of the organization. Observable indicators of organizational culture, such as rites, rituals, and symbols, have been proposed as mechanisms that create and institutionalize distinctions and classifications in organizations (Trice & Beyer, 1984, 1993). As eloquently summarized by Goffman (1961: 27) who cites Radcliffe-Brown: As the number of persons becomes large, clan segmentation becomes necessary as a means of providing a less complicated system of identification and treatments... the full range of diversity is cut at a few crucial points, and all those within a given bracket are allowed or obliged to maintain the same social front. Extending Goffman's assertions, we propose that dress is an important component of the "front" that identifies different parts of the organization and serves as contextual cues that members can use to "decode", understand, and represent the organization. As noted above, Rafaeli and Pratt (1993) suggest the existence of dress stratification in many organizations. These authors define "dress stratification" as homogeneity of dress within distinct subgroups in an organization, accompanied by differences between sub-groups. Stratification of dress was evident in the business school: Dress in certain parts of the organization, such as the executive education center and the dean's office, was much more formal and business-like than in other parts, such as document processing and faculty areas. Moreover, informants appeared well aware of this stratification; for example, dress was

Dress in Organizations 25 repeatedly included in their descriptions of the organization and in locating themselves in the organization. Given the visibility of dress, it easily serves as cues that individuals use to make sense of their environment. Building on Weick's (1979) framework for organizational sensemaking, Isabella (1990) described the interpretive activities of organization members. According to Isabella, individuals observe and interpret events in their environment to gather information that guides their understanding and provides scripts for future behavior. We propose that dress serves as one type of contextual cue that individuals perceive and interpret to make sense of their organization -- to comprehend how the parts of the organization together comprise the whole, and how they fit into and can work within this whole. Thus, dress is a salient observable cue that allows individuals to construct cognitive maps for reading and navigating the organization. Two themes in the data support this thesis: 1) awareness of stratification of dress according to function; and 2) awareness of stratification of dress according to hierarchical location. We propose that recognition of distinctions in dress helps individuals perceive the more abstract notions of hierarchical and functional specialization. Table 3 reveals some informants' observations on dress distinctions in the organization, shows they were aware of these hierarchical and functional distinctions, and that they discussed dress distinctions when referring to different organizational units. This recurrent theme supports our proposition that dress acts as a visible cue for reading and working in an organizational structure. Insert Table 3 About Here Dress and the Functional Division of Labor. Conventional wisdom about organizations suggests that individuals in different parts of organizations dress differently (Molloy, 1975, 1977; Cassedy & Nussbaum, 1983). Consistently, our observations revealed that different standards of dress existed in different physical and functional parts of the organization.

Dress in Organizations 26 As the following quote suggests, dress distinctions do not appear to be driven by particular job titles, but rather by employment in particular organizational functions: The secretaries in our office dress better than secretaries in other offices... you do notice differences between executive ed and other offices, but also between our office and other offices in executive ed like the materials department. Accordingly, faculty secretaries noted that because they were performing secretarial functions in the academic wing of the school, their dress requirements were "academic" rather than "corporate." The faculty secretary quoted first in Table 2 notes that employment in the dean's office requires the more professional dress symbolically captured in the term "suits." The relative freedom or restrictions on dress may even direct individuals' career choices. Some informants were clearly attracted to the dress of departments that conferred greater status, while others saw dress restrictions associated with higher status as very constraining. To illustrate, another faculty secretary noted: "I wouldn't accept a promotion in the dean's office or executive education, where I'd have to dress up more." Stratification of dress according to function may be so vivid that it causes outsiders to consider it more important than other unit attributes. We noted earlier the unique status of the career and placement center as a unit that serves both students and corporate representatives. Students who frequent this unit appeared to notice the dress standards that existed in this functional unit and saw them as more formal dress standards than in other units with which the students had contact. The student newspaper published on April Fool's Day included an ad that reveals students' observations of functional dress stratification. The ad notes:

Dress in Organizations 27 Applications are now being accepted for the following full-time positions: * Director of Placement * Associate Director of Placement * Assistant Director of Placement Job Requirements: 1) Nice Clothes Interested candidates please submit a full length photo to the Dean's Office by April 15, 1991 Dress and Hierarchical Stratification. Colloquial wisdom also holds that individuals in more senior positions dress and look better. Glueck (1989: 27) notes, for example, how corporate executives essentially wear a uniform, when he describes the C.E.O. of Guardian Life Insurance: Silver haired and dignified, wearing gold rimmed glasses and his C.E.O. uniform - dark suit, vest, and rep tie - Angle looks every inch the model of a major corporate mover. Similarly, Kanter's observations in one large corporation specifically suggest that appearance of the senior members of the organizational hierarchy tends to be revealing. Kanter (1977: 47) noted: Managers at Indsco had to look the part. They were not exactly cut out of the same mold, like paper dolls, but the similarities in appearance were striking... not that there were formal dress rules in this enlightened organization, like at the legendary IBM uniform, but there was an informal understanding all the same. In a similar vein, Van Maanen and Kunda (1989: 61) note that uniforms in the magical world of Disneyland: provide instant communication about the social merits and demerits of the employee within the little world of Disney workers. Uniforms also correspond to a wider status ranking that casts a significant shadow on employees of all types.

Dress in Organizations 28 Finally, Rosen (1991) in his observations of dress at the Spiro and Associates advertising agency notes that the applicability of dress norms varied between business and creative individuals, which served to perpetuate the power differential between these two groups: The primary bifurcation in a typical advertising agency, both socially and culturally, is between the creative and business roles. The distinction extends to dress at Spiro and Associates, where there is an explicit dress code for those performing business functions, but none for those performing creative functions. The dress code, in fact addressed only to business males, states that suits (of appropriate color, tailoring and fabric —not polyester, for example) must be worn at work... The difference in the dress norms between roles in the agency corresponds to its power hierarchy. (pp. 79-80). Our unobtrusive observations of members of the business school were not particularly helpful in identifying individuals' locations in the hierarchy. However, observations by employees during public meetings confirmed that individuals who were identifiable as higher status were also more formally, and more conservatively, dressed. Furthermore, interviews consistently revealed that informants associated "better" dress with the higher hierarchical levels. As Table 3 summarizes, typical comments were that secretaries and other support staff dressed less formally in their dress than managers, and deans or associate deans dressed more formally than regular faculty. Moreover, informants employed their knowledge of hierarchical dress stratification in making their own decisions about how to dress. They followed guidelines that focused on maintaining the organizational hierarchical structure by complying with implicit dress norms that existed for each hierarchical level in the organization. Several data pieces illustrate this assertion. At the most general level, an informant told us: "You have to always be sure that you're not outdoing the people higher up than you in terms of your dress." More specifically, informants made dress decisions based on their personal location in the hierarchy: I have a pink jump suit that I wear sometimes. I wear it at home all the time, going shopping and so on. I wore it here once this year. But if I were management, I'd never wear it. I'd be extremely professional. I think at my position, I can wear stuff like that.

Dress in Organizations 29 and a different informant noted: If I were working in a more high-level role, for example, supervisor, or something like that, I'd feel more comfortable to dress up more. It's not necessary in my role. These quotes suggest that informants were both aware that dress varies with hierarchical stratification and that their dress behavior explicitly helped maintain the relationship embedded in the stratification. These behaviors, in turn, reinforced the usefulness of dress as a cognitive device: Dress helped sort people in the organization and aided in members' understandings of where they belonged. In sum, dress helped to articulate the business school's distinctions between departments, and within hierarchical levels across departments. A shared and distinct pattern of dress provided contextual cues that organization members used to identify and relate to the various parts of the organization. Members' use of dress to understand and relate to the structural divisions in the business school had a cumulative effect on the school's division of labor. The dress stratification and its effect on what employees chose to wear maintained the division of labor and status in the organization. Thus, micro-behavior reproduced a macro-division of labor (Collins, 1981; DiMaggio, 1991). Dress and Individual Adaptation Seeing dress as a tool used for individual adaptation with the environment combines the behavioral and cognitive functions of dress discussed in the above two sections. Adaptation theory (White, 1959/1974; Ashford & Taylor, 1990) suggests that individuals read their environment to gain understanding about it and then adapt their behaviors to "fit" with the environment. This may involve individuals' changing their environment or modifying their own behavior to make it appropriate to the organizational environment. One way in which individuals may adapt to their environment is to alter their behaviors to fit with their understanding of their environment, without necessarily trying to change it. This is often the case with new employees —they gather information about their new environment, and

Dress in Organizations 30 then adjust their behavior to fit in with it. Dress is an obvious example of this process. For instance, a new employee is unsure about appropriate attire, so s/he asks explicit questions and/or notes carefully what others are wearing in order to discern what is acceptable, and then dresses in order to fit the dress norms that s/he identified. Many individuals take the adaptation process further and use dress to change their context. They may not be satisfied with the "fit" that exists between themselves and their context, and change their own behaviors to achieve a better fit. Here, individuals take a proactive stance and act in order to make a statement about themselves that fosters some sort of change in their relationship with the environment. This is essentially a process of negotiating one's place in the organizational context. Two patterns in the data support the assertion that dress is a tool for individual adaptation to the environment: First, informants described how they actively adjusted their dress to fit what they perceive to be appropriate dress standards. Second, informants viewed dress as a means of altering their place or situation in the organization changing expected standards of dress to which they tried to adjust. Adapting to fit in. Changing oneself to fit into one's work context is an important way in which individuals increase the fit with their environment. Indeed, informants in our study appeared to maintain a dress standard that was consistent with what they observed in their environment. As noted earlier, their dress is not dictated by a formal dress code. Rather, they appeared to accept and adapt to the tacit agreement that existed in the organization regarding appropriate dress. Thus, informants explained that they actively sought information on what others in the organization wore before and upon entering the organization as well as throughout their time as organizational members. As noted by one informant, adjustments in dress to fit with the context were considered an important element of signaling effective socialization, as well as a means of enhancing group cohesiveness: When I first started working here I felt uncomfortable. I didn't have the right clothes and I didn't have enough of them. And I didn't have new ones... within a few months I went out and bought clothes that I felt worked better here. It was a

Dress in Organizations 31 certain amount of culturalization that took place, when I first started working here, that taught me how to dress. By making the effort to dress similarly to the members of one's group, informants felt they more clearly belonged in the group as insiders, with the accompanying psychological comfort of "fitting in". Dress embodies adaptation when individuals actively attend to and invest in dress to make sure that their dress fits with the local standards. We propose that such adaptation of dress signals to the member and to others in the organization that the member is thinking and acting "appropriately". Where this sense of appropriateness is absent, we believe members experience discomfort. All but one of our informants talked about how they used dress as an adaptation device. We saw adaptation take two forms in this study: preventive and corrective. Preventive adjustments of dress refer to dress actions taken by employees before coming to work on a given day. Such actions prevent the discomfort that occurs when one is inappropriately dressed. Informants noted that they were more likely to engage in preventive adjustments early in their organizational tenure, when they were still learning the organizational ropes: [When] I first started working here I observed what people wore around here. And I'd go into stores and I'd look and I'd sort of match up my purchases with what I'd seen here. I started to shop with those guidelines that I sort of picked up around here for myself. Despite the emphasis on early experiences, preventive adjustment of dress is necessary every time one decides what to wear to work. With this type of planning, members avoid personally uncomfortable situations and blend in more effectively with the organizational context. Our data included recurring references to considerations of alternative modes of dress. Each alternative was judged as either appropriate or inappropriate for wearing in the organization in general, or for a particular task on a given day. Corrective adjustments refer to adjustments in dress made after one has selected a particular dress outfit. These are the adjustments made by individuals in response to the discomfort of being dressed inappropriately. For example, informants cited many instances

Dress in Organizations 32 when they miscalculated the appropriateness of particular attire, causing discomfort and prompting corrective action: I made a mistake one time. I came in wearing a particular jump suit. An outfit. And I actually decided that I should go home and change. It was too casual. And frankly, it would show off a woman's figure too much and be too distracting. It might have come off as being too provocative. My supervisor commented that she wasn't sure they allowed jump suits, but she didn't tell me to go home and change. I decided to do it. A recurring concern voiced by informants was that they would spill something or tear something in their clothes, which would cause them to look "unprofessional". Such discomfort could be resolved if the individual could go home and change. However, such adjustments are not always possible. When an employee lives far from the place of work, it can be difficult if not impossible to make necessary corrective dress adjustments. In such cases, employees must make extra preventive efforts. These efforts were clearly evident in our data: Informants explicitly noted that because they live far away, they planned their dress the evening before to reduce the possibility of mistakes made during the morning rush. Changing to find a new situation to fit. Informants reported that they used dress as a means of expressing group affiliations. By changing the group or unit with which one affiliated a member can alter the norm or expected dress standards which he or she is expected to fit. As Harragan (1977: 337) tells women who wish to succeed in business: Keep an eye on the costume of superior to ascertain the 'tone' or 'look' that is voluntarily adopted by upward moving men. Be very careful not to dress in conformity with lower echelon jobs. As if she was citing Harragan, an informant from the student placement center told us: "I always tell students and new people: Look at the managers, the people higher than you and how they dress. That is what you -should aspire to." Our data- revealed -that informants used dress to convey affiliation with management or leadership in the organization. Dress was a viable means for individuals to acquire status.

Dress in Organizations 33 Specifically, our informants appeared to choose their attire according to the people in the organization who they wished to be associated with or to emulate. They repeatedly noted how they selected particular role models in determining their dress; such models were consistently of higher status and usually in a higher formal position than the informant. Recurring references to looking at one's boss, the dean, and the associate dean for dress cues were evident in the data. In some cases, gauging dress with such affiliations in mind may enhance one's sense of selfefficacy, as the following quote suggests: I love suits. I'm always buying suits... they make me feel good even though I'm not what you call a P & A [Professional & Administrative]; I'm a clerical. It gives me and you the idea that I'm important. Managers in our sample appeared aware of this dynamic which, in turn, affected the level of self-consciousness they had about their own dress behaviors. As one manager noted: I think I affect the way the people I supervise dress, though they don't affect me. They don't wear slacks probably because I don't wear slacks. I've never said I don't approve of wearing slacks in the office. I don't know how they got the message, but I suspect they did. This section has described ways in which individuals actively use dress as a cognitive and behavioral tool to adapt to their environment -- to fit into the environment and to connect oneself with aspects (e.g., status levels) or people of that environment. Overall, we have discussed three ways in which organizational members use dress: performing roles, sensemaking, and adaptation. While we have distinguished between these three dress mechanisms, they are not completely independent perspectives. We have already stated that adaptation involves understanding the organizational context and changing behaviors, but there are other ways in which the three processes are interrelated. For instance, engaging in dress behaviors to fulfill, one's role is not.a task empty of cognitive content. Rather, individuals must have some understanding of what their role entails to know what dress is appropriate. Also, roles and sensemaking are related, in that an individual cannot use dress as a cue for making sense of the organization unless all organizational members use dress for the purpose of

Dress in Organizations 34 "taking on" their work roles. That is, if members do not dress according to their roles, then dress does not serve as an accurate cue for understanding the structure of the organization. Thus, while the distinctions between the three processes that we have delineated are helpful in understanding dress as used by organizational members, it is important to remember that they are related to each other, and are somewhat organization-specific. DISCUSSION AND IMPLICATIONS Our exploration of the dress behavior of the administrative staff of a business school has several important theoretical implications for the understanding the use of dress in organizations. It also offers more general insight about how individuals are connected to the organizations in which they work. While our study is a qualitative investigation of organizational dress, it is also an investigation of how individuals' decisions about what to wear at work create and maintain social order that imbues an organization with a collective coherence. In this sense, our study helps articulate how dress acts as a medium through which the micro (individual behavior regarding dress) links to the macro (a pattern of stratified organizational roles, power sources and power usage, a collective organizational image) levels of organization in dress. Thus, we see this study as part of a broader set of studies in organization science that try to link micro and macro patterns of behavior (Collins, 1981; DiMaggio, 1991). Our study contributes to our understanding of various perspectives on the relationship between individuals and organizations by exploring the significance of dress in terms of role theory, cognitive sensemaking, and adaptation theory. Dress helps individuals perform the roles assigned to them by the organization, read and understand their organization, and adjust to the expectations of this organization. Thus, while dress has been previously ignored in organization studies, it may actually be an important aid in the processes that are essential for effective individual and organizational functioning. Our study suggests that dress is a vital tool of navigation for organizational members. It is a versatile switching device: It allows an individual to emphasize different roles by deciding

Dress in Organizations 35 what to put on in the morning. Rather than seeing dress as merely an objectification of organizational values and beliefs (as it is often depicted by organizational culture researchers), this study proposes that dress is a rich and versatile navigational instrument. Our respondents used dress to signal and execute organizational roles. They also used dress to locate themselves in relationship to others both inside and outside of the organization, and as a means for establishing and breaking down barriers to others. Dress served as a mode of expressing organizational loyalty, defining organizational boundaries and playing the organization's politics. It was a means by which individuals could proactively achieve alignment with the organization's environment. Thus, rather than seeing dress as merely an instrument of conformity and of deindividuation, we see organizational dress as a more complex social behavior which has individual and organizational significance. We have proposed that dress is one means through which individual role performance is enhanced (by wearing the right clothes). We saw that dress facilitated role performance by orienting individuals towards what they thought they would be doing at work during the day, before they even stepped into the business school building. The taking on of certain dress put persons into a work mode and accompanying affective state and helped to "turn off" alternative identities. We have also shown that dress not only helps individuals perform their roles, but also helps them understand how these roles fit into the larger organizational picture. Paying attention to what one wears helps to orient and situate a member, giving coherence and purpose to their place in the organization. Stone (1981) suggests this occurs because appearance, of which dress is a part, gives an individual an identity, which establishes "what and where the person is in social terms" (p. 399). In this organizational setting, this process was particularly useful as it helped individuals not only situate themselves but locate and understand themselves in relationship to a larger division of labor and distribution of authority. The role theory and sensemaking perspectives represent rather passive views of how individuals behave in organizations: Individuals are assigned various roles which they proceed to "enter" and perform. Individuals observe, read and interpret the organizational context. Our

4 Dress in Organizations 36 analysis reveals that the same means used to enter and perform various roles (dress) can also be used in a more active mode —individuals use dress to shape and adjust their relationship to their context. Considering dress as a means for adaptation and adjustment means that individuals can use dress to influence the way others view them and thus influence the way in which they incorporate themselves into their organizational context. Our view suggests that individuals can potentially play an active role in shaping the environment in which they work. If individuals adjust their attire in order to fit into the organizational context, then actions by an individual may eventually help shape that context. A series of small coordinated actions by an informal group of individual actors can create a new reality into which other individuals ultimately adjust themselves -- thus illustrating the value of "small moves." While our data are too short-term to reveal extensive engagement of this dynamic, we do have some evidence of it occurring. For example, one informant felt that suits with shorts could be appropriate for work because they are both fashionable and comfortable, and eventually she started wearing them. She felt that this simple act eventually led to the emergence of a new norm that made shorts for women a legitimate piece of organizational attire. Our observations indeed revealed that a significant number of administrative employees wore suits with a shorts component. This study suggests that the concept of dress links processes that have not been linked before in the organization literature: It enables individuals to read and create a cognitive map of their organization, and adjust their own behavior to this cognitive reading. Having such a map is a first step in facilitating individuals' adaptation or "fit" into their organizational context. Thus, the emerging work on the importance of person -- job -- organizational fit that attempts to unravel how such fits are accomplished may benefit from focusing on aspects of organizational life such as dress. One important implication of our finding that dress offers an avenue for individual adaptation and adjustment is that organizations seeking change and innovation should allow variation in organizational dress. In such conditions (of relative flexibility in dress), new trends

Dress in Organizations 37 in dress may be introduced either randomly or intentionally. Such trends, in turn, may bring about more profound changes in the two other social psychological dynamics that dress is argued here to engage: individuals' assumptions about the organization (drawn from their cognitive sensemaking of the organization), and individuals' approaches to the execution of various roles. Future research may investigate the dynamics of dress in environments of different levels of flexibility in dress. Since nearly all of our informants were women, however, we cannot conclude that the understandings gained from this study necessarily apply to men. In short, men as a group may think, feel, and react differently to organizational dress than women. Similarly, since all of our informants were drawn from one subgroup —administrative staff —we cannot assume that other subgroups (e.g., faculty) or other organizations with different internal structures experience dress in the same way. Further investigations are necessary to understand the dynamics, attitudes, and interpretations that accompany dress for different individuals in different organizational contexts. Similarly, our study has limited generalizability to organizational settings which are unstable or in decline. The organization we studied was in a fairly steady state at the time we collected our data. Its performance was solid and its external reputation was on the rise. Resources were relatively munificent. There had been no recent layoffs nor fundamental strategic changes. We believe these characteristics of the organization are important to note as the power distribution in the school and the institution's identity (Albert & Whetten, 1985) were fairly stable. As a result, in contrast to the results of the study by Pratt and Rafaeli (1993), dress was not a symbol used' by members to define and negotiate the organization's identity or its authority pattern. Instead, we see dress as meaningfully tied to the everyday acts, understandings and adjustments of individuals who are trying to "get along" and perform well in the organization. We conclude that dress is a potent and useful instrument for accomplishing and understanding these everyday organizational behaviors.

Dress in Organizations 38 In conclusion, we believe that the study of the use of dress by individuals in this single organization has yielded a variety of insights for organizational scholars. For those interested in the links between individual action and collective structures and processes of dress, our study allows the generation of testable hypotheses. These hypotheses could test how the dress choices of organizational members create, sustain or destroy the way labor is divided, the way authority is distributed or the way the organizational image to insiders and outsiders is built, sustained or destroyed. We can see also that the study of dress can be used to apply and elaborate role theory, a sensemaking and an adaptation perspective. We have already elaborated many of these possibilities. Future research could address how members' dispositional attributes (e.g., selfesteem, self-monitoring styles) or situational characteristics (e.g., role power, role ambiguity) affect dress behavior and dress-related feelings, and their relationships to outcomes such as commitment, citizenship behavior, or levels of organizational identification. More broadly, we encourage future research to explore the particular dynamics of individuals' uses of dress in organizations. As this study has demonstrated, dress is a powerful and versatile instrument which helps individuals navigate the turbulent waters of organizational life. We encourage celebration and understanding of its use.

Dress in Organizations 39 TABLE 1 THE RELATIONSHIP BETWEEN INDIVIDUALS AND ORGANIZATIONS REGARDING DRESS Social Psychological Perspective Role of Dress Employees' Primary Activity Extent of Behavioral Focus Extent of Cognitive Focus Focus Role theory Employee Dress is one means Acting in high low performance of individuals use to context organizational roles enter, take on, and perform roles Cognitive Employees' Dress offers a vivid Understanding low high sensemaking cognitive and ubiquitous cue context theory representation of about various parts and the organizational hierarchies that make context up the organization Adaptation Employee Individuals use dress to Adaptive moderate moderate theory adaptation into "fit" themselves with moves based organizational the context in which on context they operate. understanding and acting in context

Dress in Organizations 40 TABLE 2 DRESS AND ROLE TAKING Role of Employee My dress at work is extremely conservative: suits, or skirts and blazers. I have to have extremely different wardrobes. One is rumpled khakis, which is what boaters wear to crawl around and get dirty or be in the garden or whatever. I don't have an inbetween. I am myself in jeans, a T-shirt and sneakers. But I can't wear that to work. Role of an Authority I dress to put me as someone with respect. Someone that they will trust and respect and listen to. I have to tell a lot of people "no" sometimes. Business and professional dress can help in that. I would hope that my dress will tell you that I am someone, not necessarily management, but in a management position and that I am someone you should trust if you ask a question or if I give you an answer or advice, that I am giving you the correct and not misleading information. Role of Boundary Spanner We are a service center. Like a hotel or a restaurant. People pay a lot of money to come here. So we should treat them nicely. These are the most senior executives in the country. If we're around these people, we should be more dressed up. If I dress casual, people will get the impression that they should feel comfortable to come into the office, and that we are going to help them. I look comfortable and I am comfortable and that says that if they had a problem or a question or even if they had some off the wall thought, like they wanted to know if the vending machines took Canadian quarters, they would feel comfortable enough to ask me. Role of Image-Bearer We as a school have to look good to the outside. I'm here in the front office; anyone walking into the school sees me. I want them to know that in the business school we're friendly. So I dress nice. But not too nice which would be cold. If you go by and see someone sitting there that's got on some ripped old thing or whatever...I think it's a reflection on the whole school.

Dress in Organizations 41 TABLE 3 DRESS AND COGNITIVE PERCEPTIONS OF THE ORGANIZATION Functional Distinctions in Dress People in the dean's office... [who do corporate type work].. have to wear suits all day. We here... [in the student service center]... we just wear skirts and dresses. There is a difference between the library and the computer lab, and there is a difference between the front desk and the back office in how they dress. They do different things. So they wear different clothes. There is a difference between dean and faculty. Dean is dark suits. Faculty might throw on a blazer. I'm not saying a suit is necessary, but there is a difference. Hierarchical Distinctions in Dress People in management always have suits on. For the secretaries, its lot of skirts and blouses and sweaters and slacks. The dress in our department... varies with position. The coordinators, we all dress pretty much the same... the secretaries have their own style.

Dress in Organizations 42 APPENDIX INTERVIEW PROTOCOL Hello, I am and I am writing an article for a professional magazine on what people wear when they work. Can I talk with you for a few minutes? 1. Describe your role in the B-School. How long have you worked here? 2. Can you tell me about your work path? What jobs did you have to get to this point? 3. Describe the way you are dressed today. 4. Why are you wearing this? 5. How do you feel being dressed this way? Probe: Do you feel comfortable? Do you feel like yourself? 6. Do the people you work with affect your dress? How? 7. Tell me about situations or times in the Business School when you have felt uncomfortable about the way you dressed? a. What was the situation? b. Why did you feel uncomfortable? 8. Tell me about situations or times in the Business School when you felt really good about the way you were dressed. 9. Do you recall how you felt when you first had to dress for work? a. Did your feelings change over time? How? b. How long did that take? 10. What does the way you are dressed tell me about you? 11. What would you add or change about the way you are dressed to tell me more about you? 12. What do you think the Business School expects of you in terms of what you wear? a. How did you learn about that? b. Have you changed the way you dress since working at the Business School? How? Why?

Dress in Organizations 43 13. How is (how will) your job relate(d) to what you wear? 14. What else would we need to know in order to understand what life is like as a staff member in the Business School? 15. What do you like about the Business School? 16. What do you wish could be changed to make the School a better place?

Dress in Organizations 44 REFERENCES Adams, J.S. 1976. The structure and dynamics of behavior in organizational boundary roles. In M.D. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology, 1st Ed., 1175-1201. Chicago: Rand McNally. Albert, S. & Whetten, D. 1985. Organizational identity. In L.L. Cummings and B.M. Staw (Eds.), Research in Organization Behavior, 7, 263-295, Greenwich, Conn.: JAI Press. Ashford, S.J. & Taylor, M.S. 1990. Adaptation to work transitions: An integrative approach. Research in Personnel and Human Resources Management, 8, 1-39, Greenwich, Conn.: JAI Press. Banton, M. 1965. Roles: An Introduction to the Study of Social Relations. New York: Basic. Broms, S.H. & Gahmberg, S.H. 1983. Communication to self in organizations. Administrative Science Quarterly, 28, 482-495. Cassedy, E. & Nussbaum, K. 1983. 9 to 5: The Working Woman's Guide to Office Survival. New York: Penguin Books. Chatman, J.A. 1991. Matching people and organizations: Selection and socialization in public accounting firms. Administrative Science Quarterly, 36, 459-484. Collins, R. 1981. Micro foundations of macro sociology. American Journal of Sociology, 86, 984-1014. DiMaggio, P. 1991. The micro-macro dilemma in organizational research: Implications of role-system theory. In Joan Huber (Ed.), Macro-micro Linkages in Sociology, 76 -98. Newberry Park: Sage Publications. Feldman, D.C. 1976. A contingency theory of socialization. Administrative Science Quarterly, 21; 433-452. Fussell, P. 1983. Class. New York: Ballantine Books. Gecas, V. 1982. The self-concept. Annual Review of Sociology, 8, 1-33. Glaser, B.G. & Strauss, A.L. 1967. The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine. Gluek, G. 1989. Capturing C.E.O.'s for posterity. New York Times Magazine, December, 23, 27. Goffman, E. 1961. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday.

Dress in Organizations 45 Goffman, E. 1963. Behavior in Public Places. New York: The Free Press. Goffman, E. 1967. Interaction Ritual. New York: Pantheon Books. Goffman, E. 1979. GenderAdvertisements. New York: Harper Torchbooks. Harragan, B.L. 1977. Games Mother Never Taught You: Corporate Gamesmanship for Women. New York: Warner Books. Holsti, O.R. 1969. Content Analysis for the Social Sciences and Humanities. London: Addison-Wesley. Isabella, L. 1990. Evolving interpretations as a change unfolds: How managers construe key organizational events. Academy of Management Journal, 33, 1, 7-41. Jenkins, M.C. & T.V. Atkins. 1990. Perceptions of acceptable dress by corporate and non- corporate recruiters. Journal of Human Behavior & Learning, 7(11): 38-46. Joseph, N. 1986. Uniforms and Nonuniforms: Communication Through Clothing. New York: Greenwood Press. Kanter, R.M. 1972. Commitment and Community: Communes and Utopias in Sociological Perspective. Cambridge: Harvard University Press. Kanter, R.M. 1977. Men and Women of the Corporation. New York: Basic Books. Katz, D. & Kahn, R. 1978. The Social Psychology of Organizing. New York: Wiley. Louis, M.R. 1980. Surprise and sensemaking: What newcomers experience in entering unfamiliar organizational settings. Administrative Science Quarterly, 25, 226-251. Lurie, A. 1981. The Language of Clothes. New York: Random House. Miles, M.B. 1983. Qualitative data as an attractive nuisance: The problem of analysis. In J. Van Maanen (Ed.), Qualitative Methodology, 117-135. London: Sage. Molloy, J. 1975. Dressfor Success. New York: Warner Books. Molloy, J. 1977. The Woman's Dressfor Success. Chicago: Follett. O'Leary V.E. and Ickovics, J.R. 1990. Women supporting women: Secretaries and their bosses. In H.Y. Grossman and.N.L. Chester (Eds.), The Experience and Meaning of Work in Womens' Lives. Hillsdale, N.J. Lawrence Elbaum Associates. Pratt, M.J. & Rafaeli, A. 1993. The role of symbols in fragmented organizations: An illustration from organizational dress. Working paper, University of Michigan. Rafaeli, A. & Pratt, M. 1993. Tailored meanings: On the meaning and impact of organizational dress. Academy of Management Review, 18, 1, 32-56.

1I Dress in Organizations 46 Rafaeli, A. 1989a. When cashiers meet customers: An analysis of the role of the supermarket cashier. Academy of Management Journal, 32, 2, 245-273. Rafaeli, A. 1989b. When clerks meet customers: A test of variables related to emotional expressions on the job. Journal of Applied Psychology, 74, 385-393. Rosen, M. 1991. Breakfast at Spiro's: Dramaturgy and dominance. In P.J. Frost, L. Moore, M.R. Louis, E.C. Lundberg and J. Martin (Eds.), Reframing Organizational Culture, 77-89. NewburyPark: Sage Publications. Rosengren, K.E. 1981. Advances in Content Analysis. Beverly Hills, CA: Sage. Schwab, D.P., Rynes, S.L. and R.J. Aldag. 1987. Themes and research on job search and choice. Research in Personnel and Human Resource Management, 5, 126-166. Stone, G.P. 1981. Appearance and the self: A slightly revised version. In: G.P. Stone and H.A Faberman (Eds.), Social Psychology Through Symbolic Interaction, 187-202. New York: MacMillan Publishing. Strauss, A. & Corbin, J. 1990. Basics of Qualitative Research. London: Sage Publications. Swann, W. B. 1987. Identity negotiation: Where two roads meet. Journal of Personality and Social Psychology, 53, 6, 1038-1051. Trice, H.M. & Beyer, J.M. 1984. Studying organizational cultures through rites and ceremonials. Academy of Management Review, 9, 4, 653-669. Trice, H.M. and Beyer, J.M. 1993. The Cultures of Work Organizations. Englewood Cliffs, NJ: Prentice Hall. Turner, R.H. 1978. The role and the person. American Journal of Sociology, 84, 1-23. Van Maanen, J. & Kunda, G. 1989. 'Real feelings': Emotional expression and organizational culture. In B.M. Staw & L.L. Cummings (Eds.), Research in Organizational Behavior, 11, 43-103. Greenwich, Conn.: JAI Press. Weick, K.E. 1979. Cognitive processes in organizations. In B.M. Staw (Ed.), Research in Organizational Behavior, 1, 41-74. Greenwich, Conn.: JAI Press. White, R. F. Strategies for adaptation: An attempt at systematic description. In G.V. Cuelho, D.A. Hamburg and J.E. Adams (Eds.), Coping and Adaptation, 47-68, New York: Basic Books, Inc. Wolfe, N. 1991. The Beauty Myth. London: Cambridge University Press.