Research Support School of Business Administration July 1993 WHAT IS THE DIFFERENCE BETWEEN ORGANIZATIONAL CULTURE AND ORGANIZATIONAL CLIMATE? A NATIVE'S POINT OF VIEW ON A DECADE OF PARADIGM WARS Working Paper #729 Daniel R. Denison University of Michigan

I I I I I I I I I i Ii II i I i I I I i i I

What IS the Difference Between Organizational Culture and Organizational Climate? A Native's Point of View on a Decade of Paradigm Wars Daniel R. Denison School of Business Administration University of Michigan Ann Arbor, MI 48109 313-763-4717 July 1993 My thanks to Geert Hofstede and his colleagues and students at Rijksuniversiteit Limburg in Maastricht, The Netherlands, for their insights and forbearance during my presentation of an earlier version of this paper in November, 1992. I would also like to thank Mary Yoko Brannen, Paul Carlile, Jane Dutton, George Gordon, and Karl Weick for their comments on earlier drafts of this paper.

What IS the Difference Between Organizational Culture and Organizational Climate? A Native's Point of View on a Decade of Paradigm Wars Abstract Several recent culture studies have applied quantitative survey methods and identified comparative "dimensions" of culture in a way that appears to contradict some of the epistemological foundations of culture research within organizational studies. In addition, these studies also reveal a noteworthy similarity between recent culture research and earlier research on organizational climate. This paper examines this similarity by comparing the culture and climate literatures in three areas: their definition of the phenomenon, their epistemology and methodology, and their theoretical foundations. Implications and consequences of the divergence of these two literatures are discussed along with alternative strategies for future research.

In the decade since the culture perspective burst on to the organizational studies scene, the perspective has waxed and waned in influence and vitality (Frost, 1985; Smircich and Calas, 1987). Emerging from the odd confluence of the formidable forces of postmodernism and Japanese competition, the culture perspective helped to introduce a decade of paradigm wars that have served to redefine the epistemological fault lines within the discipline (Burrell and Morgan, 1979). These paradigm wars have sharpened our skills at epistemological repartee, but, ironically, culture research has still fallen short of both theoretical and practical expectations, even as it has become an established topic area within the field. Indeed, undeniable signs of "maturity" are increasingly apparent in the recent culture literature. Culture textbooks have been published (Ott, 1989; Trice and Beyer, 1992), and an unusually large number of books have recently appeared (Denison, 1990; Sackman, 1991; Kunda, 1992; Martin, 1992; Frost, et al., 1992; Kotter and Heskett, 1992). In addition, several comparative studies have recently appeared in leading organizational journals, applying conventional quantitative research techniques to the study of culture (Chatman, 1991; O'Reilly, Chatman, and Caldwell, 1991; Hofstede, Neuijen, Ohayv, and Sanders, 1990; Jermier, Slocum, Fry, and Gaines, 1991; Gordon and DiTomaso, 1992). These studies have appeared, even though other authors have argued that the application of conventional quantitative techniques to culture research runs the risk of reducing culture to "just another variable in existing models of... organizational performance" (Siehl and Martin, 1990, p.274). More perplexing, however, is the fact that many of these most recent quantitative culture studies have become virtually indistinguishable from an older, and now neglected research area of organizational climate. Why is it, for example, that when O'Reilly, Chatman and Caldwell (1991) ask questions about risk-taking as an organizational trait, the field of organizational studies labels it as "organizational culture," yet when Litwin and Stringer (1968) asked similar questions about risk-taking, that the field labeled it as "organizational climate?" Does this similarity have implications for the recent history and future trajectory of research on organizational culture? 3

Although there have been several attempts to compare these two literatures and explore areas of integration (Schneider, 1985; 1990; Reichers and Schneider, 1990), the similarities and differences between culture and climate research have generally been neglected in recent discussions of the evolution of the culture perspective (Smircich and Calas, 1987; Ott, 1987; Schein, 1990; Trice and Beyer, 1992). This paper attempts to address these issues by comparing the culture and climate literatures on three main points in an effort to understand the differences and similarities between these two perspectives and the implications of this comparison for future research. The paper examines, in turn, (1) the nature of the phenomenon that is under investigation in each literature, (2) the epistemological assumptions and methodological strategies that have influenced each literature, (3) the theoretical foundations underlying the culture and climate literatures, and concludes by discussing the consequences of these similarities and differences for future research. The purpose of this discussion and reflection is to develop a broader perspective on past research in these two related areas and then to use this commentary and critique to help shape the future research agenda so that it focuses more closely on the phenomenon itself, rather than the paradigm wars that sustain both the real and the apparent boundaries between research perspectives. The Phenomenon This discussion begins with the observation that organizational culture and organizational climate have both similarities and differences. One of the areas of high similarity appears to be the phenomenon itself. Both perspectives examine a broad phenomenon that might be described as the internal social psychological environment of organizations and the relationship. of that environment to individual meaning and organizational adaptation. Both perspectives entertain the possibility of a shared, wholistic, collectively defined social context that emerges over time as organizations struggle with the dual problems of adaptation and individual meaning. To be sure, there are important differences between these two perspectives. Culture researchers have been more concerned with the evolution of such systems over time (Rohlen, 4

1974; Pettigrew, 1979; Van Maanen, 1979; Mohr, 1982; Schein, 1985, 1990; Mirvis and Sales, 1990), while climate researchers have placed more attention on the impact that organizational environments have on groups and individuals (Joyce and Slocum, 1984; Ekvall, 1987; Koys and DeCotiis, 1991). Culture researchers have also argued for the importance of underlying assumptions (Schein, 1985; 1990; Kunda, 1992) and individual meaning (Geertz, 1973; Pondy, Frost, Morgan, and Dandridge, 1983), while climate researchers have often placed greater emphasis on "observable" practices and procedures that are closer to the "surface" of social behavior in organizations, and the perception of these conditions by organizational members. Despite these many differences, however, the broad and rich intersection between these two literatures deserves further exploration. For example, theorists in both areas have struggled with a highly similar set of generic problems. As a first illustration, both perspectives attempt to address the problem of social organizations being, at the same time, both the product of individual interaction and a powerful influence on individual interaction. That is, organizations are comprised of individual interactions, but are also a determining context for those interactions (Giddens, 1979; Riley, 1983; Schneider and Reichers, 1983; Ashforth, 1985). Both literatures have attempted to understand this process of reciprocal evolution, but have often been more successful at solving either one aspect or the other, but not both at the same time. A second example, closely related to the first, is the "multi-layered" nature of culture or climate (Schein, 1985; 1990; James and Jones, 1974; Glick, 1985). The alternatives presented in each of the perspectives once again have notable similarities. In each literature, for example, there are attempts to conceptualize "objective" realities such as artifacts or practices and to distinguish them from the individual perceptions or cognitions that comprise values, beliefs, and assumptions. In culture research there is a frequent distinction made between the overt, surface manifestations of a culture such as artifacts, structures, symbols, rituals, or practices, and the underlying assumptions or values that those manifestation exemplify. In climate research, a similar debate exists surrounding the quasi-objective "set of conditions" that exist in an organizational system 5

and the subjective perception of those conditions by organizational members. Some theorists in fact argue that the "set of conditions" is the "climate," while others argue that the "climate" is in fact the selective perceptions (Glick, 1988; James, Joyce, and Slocum, 1988). A careful comparison of Schein's (1985; 1990) hierarchy of artifacts, values, beliefs, and assumptions in culture research, with James and Jones (1974) or Glick's (1985) discussion of the levels of organizational and psychological climate shows several parallels. For example, Schein (1985), Lundberg (1982) and others have distinguished levels of analysis ranging from core assumptions which represent the "deepest" level of culture, to beliefs and values as an intermediate level, to norms and artifacts that are visible at the "surface level." Climate researchers have also relied on a three-part typology (James and Jones, 1974; Glick, 1985) that distinguishes psychological climate (James, 1990) or the experienced organizational environment perceived by organizational members, from a socially psychological "set of conditions" called organizational climate, to an "objective" and "structural" set constructed of conditions also called organizational climate. While these levels of analysis used in the two literatures do not, of course, match directly, their common attempt to distinguish the manifest from the latent, the cognitive from the social, and the object from the subject share many similarities. Both literatures might also be criticized for giving more attention to the distinction among "levels" of culture, rather than to the integration across levels (Weick, 1993). A third issue that appears in both literatures stems from the "wholistic" or "global" nature of the phenomenon (Schneider, 1975; Schneider and Reichers, 1983; Ekvall, 1987). Both literatures struggle with the inherent expansiveness of an explicitly broad and inclusive phenomenon. Accordingly, it is often difficult, if not impossible, to define the content of the "domain" of culture or climate independent of the interests of individual theorists and researchers. Thus, the content of culture as defined by Schein (1985; 1990), Hofstede (1991), Martin (1992), Kunda (1992), Kotter and Heskett (1992), Hofstede (1991), or Peters and Waterman (1982), varies greatly. Climate research, as Denison (1990) has shown, shows a very similar pattern; the content varies by theorist, and there seems to be no natural limit to the climate domain other than 6

the ability of theorists and researchers to evoke new adjectives to describe perceived social psychological environments. A related issue also reflected in both literatures concerns the relation between this unitary whole and its constituent parts. Within the culture literature (Gregory, 1983; Riley, 1983, Van Maanen and Barley, 1984; Martin, 1992) and the climate literature (Drexler, 1977; Joyce and Slocum, 1984; Ekvall, 1987) researchers have suggested a number of dynamics that distinguish sub-cultures or distinctive sub-unit climates and conceptualizations of their relationship to the organizational whole. In addition, this issue has also taken the form in each literature of defining specific content areas such as a climate for creativity (Cummings, 1965), safety (Zohar, 1980), or service (Bowen and Schneider, 1988) or a culture of absense (Nicholson and Johns, 1985). A final area of similarity between these two literatures (and perhaps the most telling) comes when the "content" of traditional climate research is compared to the "content" of recent culture studies. Obviously, not all, or even most culture researchers ch coose to explicitly describe culture in terms of comparative traits or dimension. But when they do they frequently exhibit a clear similarity to the topics that climate researchers have been concerned with for decades. For example, Hofstede's concept of power distance -- the appropriate social and emotional distance that should be maintained between individuals of different status and power -- is highly similar to the concept of "aloofness" introduced in one of the earliest studies of organizational climate (Halpin and Croft, 1962). Interestingly enough, Halpin and Crofts, working in the context of American public schools systems, cast this dimension in a perjorative light, while Hofstede's observations across national cultures appear to lead him to cast this in far more neutral terms. Nonetheless, the underlying content is highly similar. A careful comparison of the content of culture and climate studies yields many such similarities. Schwartz and Davis's (1981) culture dimensions overlap with the climate dimensions of Taylor and Bowers (1973), O'Reilly, Chatman, and Caldwell's (1991) dimension of risk-taking is the same as Litwin and Stringer's (1968) dimension, and Wilkins (1978) concept of social control perhaps even bears some similarity to Porter and Lawler's (1973) concept of autonomy. 7

Table 1 presents a more detailed comparison of this similarity by examining a set of five summary dimensions that have been described by four separate authors, two from the climate literature and two from the culture literature. A number of similarities are highly apparent and appear to call into question the clear differences between the two phenomenon that are often assumed in the literature. Given this degree of similarity, should culture and climate be regarded as two separate domains? [insert Table 1 about here] Methodology and Epistemology During the early evolution of the culture perspective, the distinction between culture and climate was quite clear. Schwartz and Davis (1981) perhaps put it most clearly when they said that whatever culture is, it is not climate ("one way to understand culture is to understand what it is not..." p.32). Studying culture required qualitative research methods and an appreciation for the unique aspects of individual social settings. Studying organizational climate, in contrast, required quantitative methods and the assumption that generalization across social settings was not only warranted, was often the primary objective of the research. If a researcher carried field notes, quotes, or stories, and presented qualitative data to support their ideas, then they were studying culture. If a research carried computer printouts and questionnaires, and presented numbers to support their ideas, then they were studying climate. In fact, one of the most constructive contributions of the culture perspective to organizational studies has been epistemological reform (Pettigrew, 1979; Frost, Moore, Louis, Lundberg, and Martin, 1985; 1992; Van Maanen, 1979; Schein, 1985; 1990). Researchers who saw the need for epistemological revolution and reform in organizational studies mounted an attack on the orthodox logical positivism of organizational studies cloistered in a Trojan Horse called "culture." Once inside, they spread their perspective to other researchers, topics, and areas 8

of study within the field. This revolution has had far-reaching effects, and helped to transfer many of the ideas of postmodernism from social science and literary theory to organizational studies (Parker, 1992). While it is not the intent of this essay to fully explore the influence of postmodernism on organizational studies, it is difficult to ignor the pervasive impact of this perspective. Postmodernists have often indicted postitivist social science for "elevat[ing] a faith in reason to a level at which it becomes equated with progress" (Parker, 1992, p.3). As such, postmodernists are often harshly critical of attempts to systematize, define, and impose rational comparative logics on the social and organizational world. Instead, it is suggested that "all of our attempts to discover truth should be seen for what they are -- forms of discourse." (Parker, 1992, p.3). Thus, knowledge must be situated in time and place, and hence relativized. To wit; The very act of comparing, an effort to uncover similarities and differences, is a meaningless activity because post-modem epistemology holds it impossible ever to define adequately the elements to be contrasted and likened (LaTour, 1988, p. 179). This perspective, of course, wreaked havoc with the classic positivist approach of climate researchers, who often took as their primary objective a mandate to develop a universal set of dimensions that would allow for the generalization and comparison of social and psychological environments. But what was left in its place? The dynamics of this evolution in organizational studies are reminiscent of Eagleton's observation that one of the chief advantages of the postmodernist perspective is that, It allows you to drive a coach and horses through anyone else's beliefs while not saddling you with the inconvenience of having to adopt any yourself." (Eagleton, 1983, p.144). Thus, despite the influence of this epistemological agenda, it has done much less to define a substantive research agenda in the area of organizational culture (Frost, 1985). Instead, the agenda of culture researchers has recently taken a curious turn, moving from its reformist epistemological foundations to a more conventional agenda of comparative research and generalization across settings. This trend is perhaps best exemplified by a series of recent culture 9

articles that have either used quantitative methods exclusively, or have used some combination of quantitative and qualitative methods ( Hofstede, Neuijen, Ohayv, and Sanders, 1990; Chatman, 1991; O'Reilly, Chatman, and Caldwell, 1991; Jermier, Slocum, Fry, and Gaines, 1991; Gordon and DiTomaso, 1992; Denison and Mishra, 1993). The work of Geert Hofstede (Hofstede 1980a; 1980b; 1991; Hofstede, Neuijen, Ohayv, and Sanders, 1990) represents an interesting anomaly in this discussion of the evolution of the substance and epistemology of culture research. Hofstede's quantitative and comparative work on national culture received wide acclaim during the 1980s, during a period when researchers of organizational culture studiously avoided quantitative methods and comparison across settings (Schein, 1985). How is it that quantitative comparisons of national culture across many nations received widespread acceptance, while quantitative comparisons of organizational culture across organizations within a single nation were seen as unwarranted? Hofstede and his colleagues further confounded the emergent epistemological sensibilities of traditional culture researchers when they published a study of organizational culture comparing twenty Dutch and Danish firms and demonstrated substantial organizational differences between firms on several dimensions of organizational culture, which were linked to the dimensions of national culture developed in earlier research (Hofstede, Neuijen, Ohayv, and Sanders, 1990). Thus, when viewed over time, the culture literature reveals a broad range of epistemological approaches that overlap to a significant degree with the earlier research on organizational climate. While epistemological reform has made it difficult for researchers to present comparative data without acknowledging the limits to comparison, to present quantitative data without acknowledging the complementary value of qualitative data, or to make generalizations without acknowledging the inevitability of situated action and meaning, there is still a notable convergence of epistemologies. Like the convergence of substance described in the previous section, this trend makes it more difficult to distinguish culture and climate research solely on the basis of epistemology or method. 10

Theoretical Foundations The most significant differences between the culture and climate literatures thus appear to lie not in the nature of the phenomenon or the methods used to study it, but in the theories that have been developed or borrowed from other branches of the social sciences to interpret the phenomenon and the data. This section contrasts the roots of the climate literature in the theoretical perspective of Lewinian field theory (Lewin, 1951) with the roots of the culture literature in the symbolic interaction and social construction perspectives (Mead, 1934; Berger and Luckmann, 1966; Weick, 1969; 1979). Many of the differences between climate and culture can be understood by examining Lewin's basic concept of the relationship between individuals and their social environment and then considering the implications of this framework for the study of organizations. Lewin expressed his basic concepts in terms of an equation: B = f(p,e) in which B = behavior, e = the environment, and p = the person. Quite apart from the unending complexities of actually "computing" the predictions of such an equation, Lewin's framework makes the far more basic assumption that the social world can be neatly divided into Bs, ps, and es. Thus, in order to study a phenomenon such as organizational climate (or culture) from Lewin's perspective, the person must, by definition, be separate from the environment. This perspective characterizes the approach taken in the climate literature quite well. The "agents" of an organizational system, such as management, are often assumed and not studied. The "subjects" of that system, most often employees, workers, or subordinates, are the primary objects of study. The impacts of the system on its subjects is examined with a non-recursive logic that conveniently neglects the process by which the social environment is constructed by the individual members it comprises. 11

In contrast to the Lewinian logic that analytically separates the person from the environment and tends to assume that individuals are either subjects or agents of a social system, the symbolic interaction perspective (Mead, 1934) or social construction perspective (Berger and Luckman, 1966) underlying the organizational culture literature assumes that the person cannot be separated from the environment and that the members of social systems are simultaneously subjects and agents. Furthermore, this literature often defines the primary topic of interest as the recursive dynamics between the individual and the system (Riley, 1983; Giddens, 1979; Lave and Wenger, 1990), rather than the impact of the system on its members. These seemingly modest theoretical differences have wide-ranging ramifications. Three of these are examined here: First, the capacity of each of these perspectives for understanding the evolution of social process over time; Second, the potential of each of these perspectives for comparison across organizational settings; and Third, the connection of each of these perspectives to the ideology of managerialism. The Evolution of Social Process One of the advantages of the symbolic interactionist and social constructionist framework is the perspective that they provide on the evolution of social process over time. The simultaneous creation of meaning and social structure, the evolution of interaction patterns into systems of normative control, and the close connection between the symbolic and the material world can all be understood through the culture perspective. This facet of the culture perspective has been elaborated by authors such as Rohlen (1974), Van Maanen (1979), Mohr (1982), Schein (1985; 1990), Kunda (1992), and Hatch (1993). The Lewinian perspective, in contrast, provides an awkward framework within which to understand the evolution of social process. By analytically separating the person from the social environment, it becomes quite difficult to devise a theory of how that social environment evolves. Despite several noteworthy attempts at conceptualizing the formation and evolution of climates (Schneider and Reichers, 1983; Ashforth, 1985) the Lewinian perspective still appears to be better 12

at framing the problem of the influence of context on human behavior than at understanding the process by which social context develops. Nonetheless, there are particular types of social processes -- mainly those that involve the influence of an established context on organizational members who are in subordinate positions of power -- that the Lewinian framework conceptualizes quite clearly. Thus, studying the impact of the system on individual members, even with a modest time lag between the systemic stimulus and the individual response, the Lewinian framework is a highly useful perspective. Two substantive examples help to illustrate the utility of this contrast. Research on socialization (Van Maanen, 1973; Van Maanen and Schein, 1979; Schneider and Reichers, 1987; Chatman, 1991; O'Reilly, Chatman, and Caldwell, 1991) illustrates a phenomenon in which the "agents" of socialization, who are representatives of the system, are quite distinct from the "subjects" of the system who newcomers being socialized. Within this context, Lewinian field theory with its core concept of separate individuals and environments provides a useful conceptual framework. Interestingly enough, even those authors who write about socialization from a clinical perspective (eg. Van Maanen, 1973; Van Maanen and Schein, 1979) and view socialization as a process of learning a culture, maintain a clear distinction between agents and subjects of socialization. At the beginning of this process the newcomer is highly distinct from other organizational members, while at the end, they are much less so. Innovation, in contrast, provides an example in which it may be less useful to separate the individual from the environment that they are a part of. For example, Kidder's (1981) analysis of a team that designed a new computer makes a good case for the co-evolution of team culture and individual identity. The innovation process thus becomes difficult to understand when studied within a framework that analytically separates the individual from the environment. Barley (1986) reaches a similar conclusion from his study of the adoption of a new form of technology. Thus, the evolutionary processes in innovation are difficult to understand unless there is a core concept of the co-evolution of the individual and the environment as suggested by the social constructionist perspective. 13

The Viability of Comparative Research The differences between these two theoretical perspectives also have strong implications for a second point, the viability of comparative research. If "environments" exist independently of individuals,, as they do in Lewinian field theory, then theey may be conceptualized, dimensionalized measured, and compared as social entities. In addition, the relationship between organizational contexts and individual perceptions can also be conceptualized and operationalized in a way that allows for generalization across social settings. This logic fits well with the climate metaphor, and is congruent with the idea that the social environments exist separately from the individuals that comprise them. In contrast, the idea of comparing, generalizing, and dimensionalizing cultures clashes quite badly with the concept of cultures as unique social construction that create unique meaning systems for their members. Thus, if all social action is situated, as suggested by the social construction and symbolic interaction perspectives (Lave and Wenger, 1991), then comparison across settings becomes a much more questionable enterprise. Some examples from the two literatures help to illustrate the different logics of comparison that derive from the Lewinian and social construction frameworks. For example, Joyce and Slocum (1984) present an argument that individuals who experience a similar set of social psychological conditions should be regarded as sharing the same "climate," even if they have no interaction, interdependence, collective history, or identity. In this example, climate is conceptualized as a characteristic of individual-organizational "dyads" that can be disaggregated or reaggregated with little attention to the original situation in which the "climate" originated. Not only can climate be generalized and dimensiionalized, but the climate itself can be detached from the social setting and reaggregated. In contrast, many culture researchers have argued that meaning and symbolic representation can only be understood with respect to specific settings. All cultures are thus unique, and all attempts at generalization are inherently futile. Thus the goal of research must be to understand and describe individual cultures at a level that allows for an understanding of individual meaning and organizational symbolism (i.e. "thick description," [Geertz, 1973]). In this 14

case, generalizations about the relationship between the individual and the organizational environment cannot be carried beyond the situations in which they arise. The recent turn of organizational culture research to the comparative and quantitative approaches noted earlier in this paper is interesting in this regard. The issues posed above raise the question, "what is the phenomenon that these studies are comparing and generalizing about? Each of these studies acknowledges the existence of "levels of culture," and also acknowledges the limitations of generalizations about deeper levels of culture such as assumptions and beliefs. Interestingly enough, however, each of these studies selects an "intermediate" level of culture, such as values or cultural traits about which to generalize. This approach does not deny the existence of either deeper level assumptions unique to a culture, or more surface level practices that may have situationally specific meaning. Instead, each of the studies has attempted to generalize about cultures at an intermediate level that examines values or traits, and perhaps represents an interesting middle ground between the culture and climate traditions. The Ideology of Managerialism A third ramification of the theoretical differences between these two perspectives is their contrasting approaches to managerialism. Once again, the key step of analytically separating the person from the environment has many implications. A "separate" environment, as suggested by the Lewinian framework, is more consistent with both the illusion and the reality of unidirectional managerial control. The complexities of the social construction and symbolic interaction perspectives suggest a much more indeterminant form of social process, and a host of challenges to both the illusion and the reality of unidirectional managerial control. Empirical climate research often has maintained a classic Lewinian distinction between the managerial creators of the organizational context and the survey respondents who are the perceivers of the context. This distinction may indeed reflect a managerial bias, but it is important to note that climate researchers often counter this bias by reflecting the interests of their nonmanagerial respondents. This often leaves climate researchers in the position of playing both sides 15

of the managerial issue. They seldom contest the managerial creation of organizational contexts, but they most often represent the perspective of non-managerial employees within that context. In contrast to the climate literature in which the issue of managerialism is seldom addressed, culture researchers frequently address the connection between their work and the ideology of managerialism. Because the social construction framework that serves as a foundation for most culture research presumes that the social environment is created through an emergent social process, managerialism is a much more salient issue which focuses on the central question, "Who controls the process?" Top management? Labor? Occupational cultures? National cultures? Women? Men? Ethnic, regional, or industry cultures? In short, with social construction as an organizing framework, competing cultural influences are engaged in a power struggle to define the "organizational culture." Thus, subcultures (Van Maanen and Barley, 1984; Martin, 1992) are of as much interest as organizational cultures and the value system espoused by the organizational elite is but one influence on the ultimate form of the organization. Thus, the political agendas of culture researchers range from a focus on the "emancipatory interests" of organizational members (Staeblein and Nord; 1985) to a focus on "building corporate character" (Wilkins, 1989). Culture researchers present a variety of viewpoints on the issue of managerialism, but these are most often rooted in the inherent pluralistic diversity of social construction instead of the tidy distinction between person and environment provided by the Lewinian framework. These ramifications of the two theoretical perspectives do not necessarily imply that there is a different phenomenon under investigation in the culture and climate literatures, or that only certain data can constitute evidence under each perspective. They do, however, make it quite clear that the culture and climate literatures provide two very different frameworks for the interpretation of the phenomenon and the evidence. 16

Some Undesirable Consequences The lack of integration between the culture and climate literatures and their research traditions has a number of consequences that deserve further consideration. In general, a tendency to view these two research areas in the context of paradigm wars allows for less integration than might otherwise occur based on the similarities of the substantive agenda. This section of the paper focuses on three consequences of the separation between these two research traditions. Overplaying the implications of each perspective. Traditional culture researchers have critiqued many aspects of positivist organizational research, including its comparative logic, quantitative methods, and managerial bias. This critique has occasionally generated some interesting epistemological epithets. A noteworthy example is provided by Siehl and Martin (1990) in their commentary on the "pernicious social effects" (Siehl and Martin, 1990, p.273) of research linking organizational culture and performance (eg. Denison, 1984; 1990; Gordon, 1985; Hansen and Wernerfelt, 1989; Gordon and DiTomaso, 1992; Kotter and Heskett, 1992; Calori and Sarnin, 1992; Denison and Mishra, 1993). Each of these studies, to varying degrees, is quantitative and comparative, and they are good examples of a type of culture research that bears some similarity to earlier research on organizational climate. A brief analysis of their critique shows that Siehl and Martin question the contribution of quantification and comparison and infer that research on the linkages between culture and performance is intended to legitimate the direct managerial control of organizational cultures (even though a more careful reading of this literature does not necessarily support this conclusion). Interestingly enough, however, very similar research (eg., Kochan, Gobeille, and Katz, 1983; Cutcher-Gershenfeld, 1991; Cooke, 1992; 1993) conducted from a labor perspective, reaches very similar conclusions about the relationship between employee involvement, cooperative labor management relations, and effectiveness criteria such as quality and productivity. Thus, Siehl and Martin tend to group all of the elements that they view as pernicious (managerial manipulation, comparitive research, survey data, and positivist 17

epistemology) into one convenient target. A more thoughtful and fine-grained analysis might reveal a more complex, yet more integrative set of dynamics underlying these issues. In short, simplistic concepts of the distinction between these two research approaches tend to create a contrast between the two literatures that may be more apparent than real. The inadequacies of one approach often become the justification for the other. The interests of researchers in each "camp" may in fact be served by maintaining the on-going paradigm wars, even though these dynamics may detract from progress in understanding the underlying phenomenon. Ignoring issues that require integration of the perspectives. Many of the critical issues in organizational culture research in fact appear to require the integration of these two perspectives. Lack of integration between the two perspectives often means that these critical issues are not addressed. To give but one example, consider the forementioned problem of the uniqueness of social settings. While some progress has been made on this issue (Martin, Feldman, Hatch, and Sitkin, 1983; Rentsch, 1990) there is still no real conceptual framework or theory about generalization that helps to identify the conditions under which generalization may be warranted. Progress on this issue would appear to require continued research that attempts to hold both perspectives in play at the same time, while maintaining a clear focus on the phenomenon. And yet, there is clearly more attention given to the logical extremities of this issue. On one hand, the literature presents a "radically de-situated" view of climate (Joyce and Slocum, 1984) which argues that individuals do not need to share the same social setting to experience the same perceived climate, while on the other hand it presents a "radically situated" view of culture (LaTour, 1988; Parker, 1992) which implys that no generalization can be made outside of a particular setting. Both of these extremes appear to receive more attention than does the central question of the relative uniqueness and generality of culture. As noted earlier, the work of Hofstede (1980; 1990) and his colleagues presents an interesting paradox in this regard. Researchers interested in the influence of national culture on 18

management and organization tended to accept Hofstede's comparative framework. Researchers interested in organizational culture, however, tended to reject comparative research with similar methods and measures for much of the 1980's. This was true even though even though nearly all of the empirical research on organizational culture took place in western, English-speaking, or even exclusively American organizations. Thus, comparative logic was rejected in a relatively homogeneous setting at the same time that it was accepted in a relatively heterogeneous setting. In either case, the issue of how and when researchers might generalize across settings remains relatively unexplored, despite the obvious theoretical and practical value. Unfortunately, this is unlikely to change until more researchers accept that the answer to the question of uniqueness and universality is neither a clear "yes," or "no," but rather an indeterminant "maybe," "sometimes," or "it depends." Distance from the phenomenon. Culture research originally gained great energy from the observation that organizational research had lost much of it fidelity -- the verisimilitude with respect to organizational life itself -- and suggested that the antidote to this problem was in description, ethnography, and an attempt to understand the native's point of view. But the epistemological nature of the culture literature has, over time, had a very different impact. Discussion of research methods and approaches often outweighs the discussion of the organizations themselves. Several authors even question whether organizational studies should be the study of organizations or the study of discourse about organizations (Smircich and Calas, 1987; Parker, 1992). At the same time, the field research required to do "thick description" represents a sizeable invesment of time. As Kunda (1993) points out, "thin description" may be a more accurate description of much of the culture research in organizational studies. Climate researchers, in contrast, seem inextricably wed to a limited form of contact with the organizations that they study: questionnaire data appear to be the sine qua non of culture research (Trice and Beyer, 1992). This approach may require contact with a research site (at least by mail) but it does not always require direct contact with the social psychological phenomena that are the primary objects of study. 19

Distance from the phenomenon itself tends to sustain paradigm wars and deter integration. The paradigm becomes the phenomenon rather than the organization. In contrast, a focus on the phenomenon, drawing in an eclectic manner from the various theories, methods, and perspectives. will make the greatest contribution to a substantive understanding of organizational cultures. Discussion This paper has attempted to address a remarkable paradox in the culture literature: With the recent appearance of culture studies based upon quantitative survey research methods, "culture" research has begun to emulate a substantive and epistemological research agenda that served as its antithesis less than a decade ago. Culture research is now being published in the leading organizational journals, but (ironically) only by emulating the same positivist research model that culture researchers originally deplored. Furthermore, a comparison of this recent culture research with the organizational "climate" literature of the 1960s and 1970s shows a growing similarity and suggests that it is becoming increasingly difficult to distinguish current culture research from the earlier climate paradigm on the basis of either the substantive phenomenon or the methods and epistemology. This paper has argued that one of the most enduring differences between culture and climate stems from their respective theoretical foundations. Both are rooted in dominant theoretical traditions of their time, climate research growing out of Lewinian field theory (Lewin, 1951) and culture research growing out of the social construction framework (Mead, 1934; Berger and Luckmann, 1966). Even this boundary is not always so clear. The research of Chatman (1991) and O'Reilly, Chatman, and Caldwell (1991) reflects many aspects of the Lewinian framework, and Ashforth (1985) or Schneider and Reichers (1983) can easily be viewed as describing the social construction of organizational contexts. Nonetheless, a focus on the theoretical perspectives and epistemological differences as differences in interpretation rather than differences in the underlying phenomenon may provide a stronger foundation for integration than the current assumption, frequently taken, that culture and 20

climate are fundamentally different and non-overlapping phenomenon. Taking the assumption that these two literatures primarily represent differences in interpretation has at least two interesting implications: First, both literatures can then be used to address the common phenomenon of the creation and influence of social contexts in organizations. The internal social psychological environment of organizations becomes a topic that spans epistemologies, methods, and theories, and advocates multi-faceted interpretations of eclectic assemblages of evidence. Second, all data become legitimate. Different researchers will, of course, generate different forms of evidence and different ways of interpreting each other's results, sustaining a rich source of diversity. But the debate about what constitutes the "right" kind of data can be given a decent burial. The debate over whether rituals or regressions, or surveys or semiotics constitute the best data can become subordinate to what these multiple data sources and strategies can reveal about social contexts and their influence on individuals and organizations. Finally, it worth returning to a question posed at the beginning of this paper; "What implications does the similarity of recent culture research with traditional climate research have for the future trajectory of research on organizational culture?" This discussion suggests at least three ways to interpret this evolution and three views of what culture research might look like in the future. Another victory for the logical positivists. Culture researchers have made a valiant effort to reform organizational studies, but it doesn't want to be reformed. In the end, the culture perspective cannot resist the dominant positivist tradition in organizational studies. "Culture" research will become increasingly traditional and comparative, and in fact will become "just another variable in existing models of... organizational performance" just as Siehl and Martin warned earlier in this paper. Epistemological reform continues to expand, but under another name. In this second scenario, the epistemological reform that characterized the early evolution of culture research continues, but it becomes progressively less connected with culture as a topic. Epistemological 21

reformists may "surrender" culture as their flagship, but the perspective will spread through a host of other topics in organizational studies. Culture was a "beachhead" in the paradigm wars of the 1980s, but the battle has now spread. Culture develops as a substantive topic area. The third scenario, and one to which much of the attention in this paper has been devoted, is that culture establishes a stronger identity as a topic area. The culture perspective subsumes and integrates the traditional climate literature and embraces both the comparative and interpretive perspectives as complementary approaches to understanding the complex and important topic of social contexts in organizations. 22

Table 1 A Comparison of Dimensions Used by Culture and Climate Researchers Hofstede O'Reilly & Chatman Litwin & Stringer Koys & DeCotiis Structure authority stability structure i Support power distance respect for people support support Risk security innovation teamwork risk innovation Cohesiveness collectivism identity cohesion Outcome Orientation results orientation outcome orientation standards pressure

References Ashforth, B. (1985) Climate formation: Issues and extensions. Academy of Management Review, 10(4), 837-847. Barley, S. (1986) Technology as an occasion for structuring: Observations on CT scanners and the social order of radiology departments. Administrative Science Quarterly, 31: 78 -108. Berger, P. and Luckmann, T. (1966) The social construction of reality. New York: Penguin. Bowen, D. and Schneider, B. (1988) Services marketing and management: Implications for organizational behavior. in B.M. Staw and L.L. Cummings (Eds.) Research in Organizational Behavior, Vol. 10. Greenwich, CT: JAI Press. Burrell G. and Morgan, G. (1989). Sociological paradigms and organizational analysis. London: Heinemann. Calori, R. and Sarnin, P. (1991) Corporate culture and economic performance: A French study. Organization Studies, 12(1), 49-74. Chatman,' J. (1991) Matching people and organizations: Selection arid socialization in public accounting firms. Administrative Science Quarterly, 36, 459-484. Cooke, W. (1992) Product quality improvement through employee participation: The effects of unionization and joint union-management administration. Industrial and Labor Relations Review. 46: 119-34. Cooke, W. (1993) Employee participation, group-based pay incentives, and company performance: A union-non-union comparison. Working Paper, Wayne State University, Detroit, MI. Cutcher-Gershenfeld, J. (1991). The impact on economic performance of a transformation in workplace relations. Industrial and Labor Relations Review, 44(2), 241-260. Cummings, L. L. (1965). Organizational climates for creativity. Academy of Management Journal 8, 837-847. Denison, D. (1984) Bringing corporate culture to the bottom line. Organizational Dynamics, 13(2), 4-22. Denison, D. (1990) Corporate culture and organizational effectiveness. New York: John Wiley. Denison, D. and Mishra, A. (1993) Toward a theory of organizational culture and effectiveness. Organization Science, in press. Drexter, J. (1977). Organizational climate: Its homogeneity within organizations. Journal of Applied Psychology, 62, 38-42. Eagleton, T. (1983). Literary theory. Oxford: Blackwell. 24

Ekvall, G. (1987) The climate metaphor in organizational theory. in B. Bass and P. Drenth (Eds.) Advances in Organizational Psychology. Beverly Hills, CA: Sage. Frost, P. (1985) Rekindling the flame. Symposium at the Academy of Management. Frost, P., Moore, L., Louis, M., Lundberg, C., and Martin, J. (1985) Organizational culture. Beverly Hills: Sage. Frost, P., Moore, L., Louis, M., Lundberg, C., and Martin, J. (1992) Reframing organizational culture. Beverly Hills: Sage. Gertz, C. (1973). The interpretation of cultures. New York: Basic books. Giddens, A. (1979) Central problems in social theory. Berkeley: University of California Press. Glick, W. (1985) Conceptualizing and measuring organization and psychological climate. Pitfalls in multilevel research. Academy of Management Review, 10, 601-616. Glick, W. (1988) Response: Organizations are not central tendencies: Shadowboxing in the dark, Round 2. Academy of Management Review, 13(1), 133-137. Gordon, G. (1985) The relationship of corporate culture to industry sector and corporate performance. in R.H. Kilman, M.J. Saxton, R. Serpa, and Associates (Eds.), Gaining control of the corporate culture. San Francisco: Jossey-Bass. Gordon, G. and DiTomaso, N. (1992) Predicting corporate performance from organizational culture. Journal of Management Studies, 29(6), 783-798. Gregory, K. (1983). Native view paradigms: Multiple cultures and culture conflict in organizations. Administrative Science Quarterly, 28, 359-376. Halpin, A. and Croft, D. (1962) The organizational climate of schools. St. Louis: Washington University Press. Hansen, G. and Wemerfelt, B. (1989) Determinants of firm performance: The relative importance of economic and organizational factors. Strategic Management Journal, 10(5), Hatch, M. J. (1993). The dynamics of culture. Academy of Management Review, in press. Hofstede, G. (1980a) Culture's consequences: International differences in work-related values. Beverly Hills:Sage. Hofstede, G. (1980b) Motivation, leadership and organization: Do American leadership theories apply abroad? Organizational Dynamics, 9, 42-63. Hofstede, G. (1991) Culture and organizations: Software of the mind. New York: McGrawHill. Hofstede, G., Neuijen, B., Ohayv, D., and Sanders, G. (1990) Measuring organizational cultures: A qualitative and quantitative study across twenty cases. Administrative Science Quarterly, 35(2), 286-316. 25

James, L. and Jones, A. (1974) Organizational climate: A review of theory and research. Journal of Applied Psychology, 76, 214-224. James, L., Joyce, W., and Slocum, J. (1988) Comment:: Organizations do not cognize. Academy of Management Review, 13(1), 129-132. Jermier, J., Slocum, J., Fry, L. and Gaines, J. (1991) Organizational subcultures in a soft bureaucracy: Resistance behind the myth and facade of an official culture. Organization Science, 2(2), 170-194. Joyce, L. and Slocum, J. (1984) Collective climate: Agreement as a basis for defining aggregate climate in organizations. Academy of Management Journal, 27(4), 721-742. Kidder, T. (1981) The soul of a new machine. Boston: Little, Brown, & Co. Kochan, T., Gobeille, K., and Katz, H. (1983). Industrial relations performance, economic performance, and QWL programs: An interplant analysis. Industrial and Labor Relations Review, 37: 3-17. Kotter, J. and Heskett, J. (1992) Corporate culture and performance. New York: Free Press. Koys, D. and Decotiis, T. (1991) Inductive measures of psychological climate. Human Relations, 44(3), 265-285. Kunda, G. (1992) Engineering culture. Philadelphia: Temple University Press. Kunda, G. (1993) Engineering culture. Presentation to the Academy of Management. LaTour, B. (1988). The Pasteurization of France. Cambridge: Harvard University Press. Lave, J. and Wenger, S. (1990) Situated learning. Cambridge: Cambridge University Press. Litwin, G. and Stringer, (1968) Motivation and organizational climate. Cambridge: Harvard University Press. Lewin, K. (1951) Field theory in social science. New York: Harper and Row. Lundberg, C. (1985) On the feasibility of cultural intervention in organizations. In Organizational culture. P. Frost, L. Moore, M. Louis, C. Lundberg, and J. Martin (Eds.), Beverly Hills, CA: Sage. Martin, J. (1992) Cultures in organizations: Three perspectives. New York: Oxford University Press. Martin, J., Feldman, M., Hatch, M., and Sitkin, S. (1983) The uniqueness paradox in organizational stories. Administrative Science Quarterly, 28, 438-453. Mead, G. (1934) Mind, self, and society. Chicago: University of Chicago Press. Mirvis, P. and Sales, A. Feeling the elephant: Culture consequences of a corporate acquisition and buy-back. In B. Schneider (Ed.) Organizational climate and culture. San Francisco: Jossey-Bass, 345-382. Mohr, L. (1982) Explaining organizational behavior. San Francisco, CA: Jossey-Bass. 26

Nicholson, N. and Johns, G. (1985). The absence culture and the psychological contract —who's in control of absence? Academy of Management Review, 10, 397-407. O'Reilly, C., Chatman, J., and Caldwell, D. (1991) People and organizational culture: A profile comparison approach to assessing person-environment fit. Academy of Management Journal, 34, 487-516. Ott, J. (1989) The organizational culture perspective. Chicago: Dorsey Press. Parker, M. (1992). Post-modem organizations or postmodern organization theory? Organizational Studies, 12(1): 1-17. Peters, T. and Waterman, R. (1982) In search of excellence: Lessons from America's best run companies. New York: Harper & Row. Pettigrew, A. (1979) On studying organizational cultures. Administrative Science Quarterly, 24, 570-581. Pettigrew, A. (1990) Organizational climate and culture: Two constructs in search of a role. In B. Schneider (Ed.) Organizational climate and culture. San Francisco, Jossey-Bass, 413-433. Pondy, L., Frost, P., Morgan, G. and Dandridge, T. (1983) Organizational symbolism. Greenwich, CT: JAI Press. Porter, L. and Lawler, E. (1973) Managerial attitudes and performance. Homewood, IL: Richard D. Irwin. Reichers, A. (1987) An interactionist perspective on newcomer socialization rates. Academy of Management Review, 12, 278-287. Reichers, A. and Schneider, B. (1990). Climate and culture: An evolution of constructs in B. Schneider (Ed.) Organizational climate and culture. San Francisco: Jossey-Bass, pp. 5-39. Rentsch, J. (1990) Climate and culture: Interaction and qualitative differences in organizational meanings. Journal of Applied Psychology, 75(6), 668-681. Riley, P. (1983) A structurationist account of political cultures. Administrative Science Quarterly, 28(3), 414-437. Rohlen, T. (1974) For harmony and strength: Japanese white collar organization in anthropological perspective. Berkeley: University of California Press. Rosenau, P. (1992) Postmodernism and the social sciences: Insights, inroads, and intrusions. Princeton, NJ: Princeton University Press. Rousseau, D. (1990) Assessing organizational culture: The case for multiple methods. In B. Schneider (Ed.) Organizational climate and culture. San Francisco: Jossey-Bass, 153 -192 Sackman, S. (1991) Cultural knowledge in organizations: Exploring the collective mind. Beverly Hills: Sage. 27

Schein, E. (1985) Organizational culture and leadership. San Francisco: Jossey-Bass. Schein, E. (1990) Organizational culture. American Psychologist, 45(2) 109-119. Schneider, B. (1975) Organizational climate: An essay. Personnel Psychology, 28, 447-479. Schneider, B. (1985) Organizational behavior. Annual Review of Psychology, 36, 573-611. Schneider, B. (1990) Organizational climate and culture. San Francisco: Jossey-Bass. Schwartz; H. and Davis, S. (1981) Matching corporate culture and business strategy. Organizational Dynamics, 10, 30-48. Schneider, B. and Reichers, A. (1983) On the etiology of climates. Personnel Psychology, 36, 19-39. Siehl, C. and Martin, J. (1990) Organizational culture: A key to financial performance? in B. Schneider (Ed.) Organizational climate and culture. San Francisco: Jossey-Bass. Smircich, L. and Calas, M. F. (1987) Organizational culture: A critical assessment. in F. Jablin, L. Putnam, K. Roberts, and L. Porter (Eds) Handbook of Organizational Communication. Beverly Hills, CA: Sage 228-263. Smircich, L. (1983). Concepts of culture and organizational analysis. Administrative Science Quarterly, 28(3), 339-358. Staeblein, R. and Nord, W. (1985). Practical and emancipatory interests in organizational symbolism: A review and evaluation. Journal of Management, 11, 13-28. Taylor, J. and Bowers, D. (1973) The survey of organizations. Ann Arbor: The Institute for Social Research. Trice, H. and Beyer, J. (1992) The cultures of work organizations. Englewood Cliffs, NJ: Prentice-Hall. Van Maanen, J. (1973) Observations on the making of policemen. Human Organization, 32, 407-17. Van Maanen, J. (1979) Qualitative research. Administrative Science Quarterly, 24: 570-581. Van Maanen, J, and Barley, S. (1984). Occupational communities: Cultural control in organizations. Research in Organizational Behavior, 6, 287-365. Van Maanen, J. and Schein, E. (1979) Toward a theory of organizational socialization. Research in Organizational Behavior 11: 1, 209-259. Weick, K. (1969) The social psychology of organizing. Reading, MA: Addison-Wesley. Weick, K. (1979) The social psychology of organizing (2ed). Reading, MA: Addison-Wesley. Weick, K. and Roberts, K. (1993) Collective mind and organizational reliability: The case of flight operations on a aircraft carrier deck. Working paper. University of Michigan. 28

Wilkins, A. (1989) Developing corporate character: How to successfully change an organization without destroying it. San Francisco: Jossey-Bass. Wilkins, A. (1978). Organizational stories as an expression of management philosophy. Unpublished Ph.D. thesis, Stanford University. Zohar, D. (1980). Safety climate in industrial organizations: Theoretical and applied implications. Journal of Applied Psychology, 65, 96-117. 29

I I I I i II \ l I i I i ~ ~ ~ ~ ~ ~ ~ ~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"