The concept of culture
The word ‘culture’ has three meanings: Literally it means tilling the soil: cultivation. Metaphorically the word is used for the training or refining of the mind: civilization. However, in the past decades a broader metaphorical meaning has become popular, derived from anthropology: collective ways of acting, thinking, and feeling. ‘Culture’ in this sense is “the collective programming of the mind that distinguishes the members of one group or category of people from others” (Hofstede, Hofstede and Minkov 2010, 6). In the case of national culture, the category is the nation. In the case of organizational cultures, the category is the organization as opposed to other organizations—other things, like nationality, being equal. Next to national and organizational cultures one can distinguish regional cultures, occupational cultures, gender cultures and so on. However, the use of the word ‘culture’ for all these categories does not mean that they are identical phenomena. For different kinds of social systems, their ‘cultures’ are usually of a different nature. This is particularly the case for organizational cultures versus national cultures, if only because membership of an organization tends to be partial and more or less voluntary, while the ‘membership’ of a nation is permanent and usually established at birth.
‘Culture’ as thus defined is a construct, that is, a product of our imagination. We have defined it into existence: “A construct is not directly accessible to observation but inferable from verbal statements and other behaviors and useful in predicting still other observable and measurable verbal and nonverbal behaviors” (Levitin 1973).
Culture as collective programming of the mind manifests itself in several ways. From the many terms used to describe manifestations of culture, the following four together cover the total concept rather neatly: symbols, heroes, rituals and values. These can be imagined as the skins of an onion, symbols representing the most superficial, and values the deepest, layers of culture, with heroes and rituals in between (figure 1).
Symbols are words, gestures, pictures or objects which carry a particular meaning, only recognized as such by those who share the culture. The words in a language or jargon belong to this category, as do dress, hair-do, Coca-Cola, flags and status symbols. New symbols are easily developed and old ones disappear; symbols from one cultural group are regularly copied by others. This is why symbols represent the outer, most superficial layer of culture.
Heroes are persons, alive or dead, real or imaginary, who possess characteristics that are highly prized in a culture, and thus serve as models for behavior. Founders of companies often become cultural heroes. In this age of television, outward appearances have become more important in the choice of heroes than they were before.
Rituals are collective activities, technically superfluous to reach desired ends, but within a culture considered socially essential: they are therefore carried out for their own sake. Ways of greeting and paying respect to others, social and religious ceremonies are examples. Business and political meetings organized for seemingly rational reasons often serve mainly ritual purposes, like allowing the leaders to assert themselves.
Symbols, heroes and rituals together can be labeled ‘practices’. As such they are visible to an outside observer; their cultural meaning, however, is not necessarily visible and lies in the way these practices are interpreted by the insiders.
The core of culture is formed by values. Values are strong emotions with an arrow to it: a minus and a plus pole, such as evil versus good, abnormal versus normal, ugly versus beautiful, dangerous versus safe, immoral versus moral, indecent versus decent, unnatural versus natural, dirty versus clean, paradoxical versus logical, irrational versus rational.
Values are among the first things children learn—not consciously, but implicitly. Because they were acquired so early in our lives, many values remain unconscious to those who hold them. Therefore they can only rarely be discussed, or directly observed by outsiders. They can only be inferred from the way people act under various circumstances. This includes the way they answer questionnaires, although their answers should not always be taken literally. Interpreting answers to questionnaires is a main task of cross-cultural researchers who nowadays have many statistical tools at their disposal to help them.
National cultures and organizational cultures
Two large research projects into culture differences (Hofstede 1980; Hofstede et al. 1990) showed that national cultures differ mostly at the level of values, while organization cultures differ mostly at the level of the more superficial practices: symbols, heroes, and rituals. National cultures oppose otherwise similar individuals, institutions and organizations across countries; the pioneer study on national cultures was based on different national subsidiaries of one large international business company. Organizational (also called corporate) cultures oppose different organizations within the same countries. The path-breaking study in this field used different organizations or parts of organizations in two countries: Denmark and the Netherlands.
Figure 2 illustrates when and which of our mental programs were acquired. We humans are born incompletely programmed; during the first ten years of our lives we possess an amazing capacity for absorbing complex, diffuse and implicit mental programs. One example is learning a second language: when someone speaks another language accent-free, he or she almost surely learned it as a child. With the onset of puberty, our ways of learning become more explicit and focused; we can still learn foreign languages, but we will almost always retain an accent. As mentioned above, our early programming includes most of our basic values. We acquire these mental programs from our social environment, the family, the neighborhood, and early schooling. The right-hand column of figure 2 shows which levels of culture we acquire, and in which period. We are born boy or girl, and within a nation. Gender and nationality are therefore most decisive for our basic values. The school period mostly bridges puberty; the kind of school students attend relates to their social class, and influences their future occupation. Our school education mixes both values and practices. Cultures of work organizations are acquired through socialization at the work place, which most people enter as adults—that is, with their basic values firmly in place. A business culture (like the culture of banking, or of tourism) can be placed somewhere between the occupation and the organization level.
National cultures differ mostly at the level of values, while organization cultures differ mostly at the level of the more superficial practices: symbols, heroes, and rituals.
So national culture differences are rooted in values learned before age 10; children learn them from parents who also acquired them before age ten, so they are quite stable and take generations to be changed.
Organizational cultures are rooted in practices learned on the job, and they can change much faster. Their implications for management are quite different, as will be shown later.
The dimensions paradigm in studying the social world
My own cross-cultural research in the 1970s started from a large database of employee value statements (more than 100,000 questionnaires) collected in subsidiaries of the IBM corporation in 40 countries (Hofstede 1980). Being trained as a psychologist, I initially tried to analyze the data across individuals, but after a long struggle I discovered that they made much more sense if I compared mean answers across countries. When I did that, I could relate the differences between country cultures to basic dilemmas of human societies, which had been described 20 years earlier in a review of the anthropological and sociological literature (Inkeles and Levinson 1969 ). These dilemmas corresponded with dimensions on which each country could be scored. The dimension approach to culture research has since become a paradigm for empirical cross-cultural research. A paradigm is “a model from which spring particular coherent traditions of scientific research” (Kuhn 1970).
Dimensions are a conceptual way of dividing complex realities into separate basic elements. Many thinkers about the social world have divided it into categories, but the dimensions in this case are not born from armchair reflection but from empirical research, using the methods of modern statistical analysis. Like ‘culture’, ‘dimensions’ are constructs. According to the definition by Levitin cited above, they should be “useful in predicting … observable and measurable verbal and nonverbal behaviors.” Demonstrating these is called validation.
The social world is like a cake that can be cut in different ways: the way we divide it into dimensions depends on our intended purpose for them.
So there is not one set of ‘correct’ dimensions: different applications may need different models. Moreover, dimensional models depend on the level of analysis. As will be shown below, dimensions for comparing societies (countries) are completely different from dimensions for comparing organizations, and these are again different from dimensions for comparing individuals. Dimensions for comparing societies belong to anthropology, for comparing organizations to sociology, for comparing individuals to psychology. Applied disciplines like management studies, political science and economics that operate at more than one level of analysis run a danger of confusing dimensions from different levels.
Comparing national cultures
The most recent version of the Hofstede model for comparing national societies consists of six independent dimensions, rooted in differences between national cultural values. Scores on each dimension on a 0–100 scale are available for between 76 and 93 countries (Hofstede, Hofstede and Minkov 2010). The dimensions have been labeled:
- Power Distance (large versus small), related to solutions for the basic problem of human inequality;
- Uncertainty Avoidance (strong versus weak), related to the level of stress in a society in the face of an unknown future;
- Individualism versus Collectivism, related to the integration of individuals into primary groups;
- Masculinity versus Femininity, related to the division of emotional roles between women and men;
- Long-Term versus Short-Term Orientation, related to the choice of focus for people’s efforts: the future or the present and past;
- Indulgence versus Restraint, related to the extent of gratification of human desires for enjoying life.
Several of the dimensions have been replicated in major surveys by others among different kinds of respondents (Hofstede, Hofstede and Minkov, 2010, 35). The ranking of countries on the dimensions has been quite stable since the first data were collected. More important, the dimensions were validated against a large variety of cross-national data from other sources. Hofstede (2001, 520) lists more than 400 significant correlations for the first five dimensions. Examples are:
- Power distance correlates with: income inequality, respect for elders, polarization and violence in national politics.
- Uncertainty avoidance correlates with: number of laws and rules, belief in experts, xenophobia, faster driving.
- Individualism correlates with: national wealth, faster walking, weaker family ties, frequency of using the word “I.”
- Masculinity correlates with: a stress on growth as opposed to care for the weak and the environment, and negatively with the percentage of women elected in parliaments and governments.
- Long-Term Orientation correlates with: savings rates, economic growth of poor countries, and adapting to changed reality, as opposed to Short- Term Orientation which correlates with: concern for social obligations, national pride, and fundamentalisms.
- Indulgence correlates with: higher birthrates, more active sports, more obesity, more private Internet, smaller police force.
These societal-level dimensions have been applied in literature from a surprising number of different disciplines and areas, like:
- Cross-cultural psychology
- Structure of language, cognition, intelligence
- International and diversity management
- International business, acquisitions, alliances
- International marketing, advertising, consumer behavior, packaging
- International politics and economics
- International legislation, procedural justice, imprisonment, insurance
- Architecture, urban/country planning, office design, industrial design
- Health, medicine, psychiatry, medication, care
- Ethics, religion, spirituality, sexuality
Comparing organizational cultures
Our research project into organization culture differences (Hofstede et al. 1990; Hofstede 2001) was carried out on 20 units from ten different organizations, five in Denmark, five in the Netherlands. On the IBM national culture dimensions these two countries scored fairly similarly: both belong to the same Nordic-Dutch cluster. Units of study were both entire organizations and parts of organizations, which their management assumed to be culturally reasonably homogeneous (the research outcome later enabled this assumption to be tested). Unit sizes varied from 60 to 2,500 persons. The number of units was small enough to allow each unit to be studied in depth, qualitatively, as a separate case study. At the same time, it was large enough to permit statistical analysis of comparative quantitative data to cover all cases.
The first, qualitative phase of the research followed a classical anthropological approach. It consisted of in-depth person-to-person interviews of two to three hours duration each with nine informants per unit (thus, a total of 180 interviews). These interviews served both to get a qualitative feel for the whole (the ‘gestalt’) of the unit’s culture, and to collect issues to be included in the questionnaire for the ensuing survey.
The second, quantitative phase of the project consisted of a paper-and- pencil survey with pre-coded questions, administered to a strictly random sample from the unit, composed of about twenty-five managers (or as many as the unit counted), twenty-five college-level non-managers (‘professionals’) and twenty-five non-college-level non-managers. The questions in the survey included those used in the cross-national IBM study; most however were developed on the basis of the interviews of the first phase. Questions were formulated about all issues that the interviewers suspected differed substantially from one unit to another. These included in particular many perceptions of daily practices, which had not been covered in the cross-national studies.
A statistical (factor) analysis of the survey answers found only small differences in values between the units, but larger differences in practices. These could be divided into six dimensions of organizational cultures, and corresponded to distinctions well known from organization sociology and management studies. We labeled them as follows:
- Process-oriented versus results-oriented. Process-oriented cultures are dominated by technical and bureaucratic routines, results-oriented by a common concern for outcomes.
- Job-oriented versus employee-oriented. The former assume responsibility for the employees’ job performance only, and nothing more; employee-oriented cultures assume a broad responsibility for their members’ well-being.
- Professional versus parochial. In the former, the (usually highly educated) members identify primarily with their profession; in the latter, the members derive their identity from the organization for which they work.
- Open systems versus closed systems. This dimension refers to the common style of internal and external communication, and to the ease with which outsiders and newcomers are admitted.
- Tight versus loose control. This dimension deals with the degree of formality and punctuality within the organization; it is partly a function of the unit’s technology: banks and pharmaceutical companies can be expected to show tight control, research laboratories and advertising agencies loose control; but even with the same technology some units may still be tighter or looser than others.
- Pragmatic versus normative. The last dimension describes the prevailing way (flexible or rigid) of dealing with the environment, in particular with customers. Units selling services are likely to lean towards the pragmatic (flexible) side, units involved in the application of laws and rules towards the normative (rigid) side.
Just as the national culture dimensions were validated against cross-national data from other sources, we validated the organizational culture dimensions against information about the organizations from other sources. For each of the dimensions, we found one or more significant correlations with other data:
- Process rather than results orientation correlated with: material-versus labor-intensive processes, higher absenteeism, and steeper hierarchical structure.
- Employee rather than job orientation correlated with: higher invested capital, younger workforce, and more highly educated top managers.
- Parochial rather than professional correlated with: public ownership, smaller size, more unionization, and fewer meetings.
- Open rather than closed correlated with: higher percentage of female employees, higher average employee seniority, less formalized organization, freer expression of employee opinions in in-company personnel journal.
- Loose rather than tight correlated with: lower percentage of female managers, higher average employee education, recent growth, and lower absenteeism.
- Normative rather than pragmatic correlated with: public versus private ownership.
Our six dimensions are being used by management consultants in various countries as a framework to describe, measure and compare organization cultures, and to monitor cultural change processes. A caution is that their research base in 20 units from two North-West European countries is too narrow for declaring them universally valid and sufficient. For describing organization cultures in countries with markedly different national cultures and/or in kinds of organizations not covered in our initial study, additional dimensions of practices may be necessary and/or some of the present six may be less useful. In such cases, a similar research project should be carried out on a sufficient number of local organizations, and relevant dimensions should be extracted from that; they will probably partly overlap with the ones in our study.
Culture and the role of international management
Since the 1980s, ‘culture’ has become a main concern for management. It started when Japanese companies outperformed American companies, and ‘Japanese culture’ or ‘Japanese management’ was invoked as the mysterious recipe for their success. In 1982, two books, each by a Harvard professor and a McKinsey consultant (Deal and Kennedy 1982; Peters and Waterman 1982) introduced the concept of ‘corporate culture’ to explain why some American companies did so much better than others. The 1980 first edition of my book Culture’s Consequences owed part of its success to being the first data-based research report in an otherwise fuzzy and mysterious area. The distinction between national and organizational cultures was not clear to many readers; some tried applying my cross-national dimensions to corporate cultures.
The new interest in ‘corporate cultures’ was the immediate reason for designing our Danish-Dutch organizational cultures project. Setting it up and financing it took considerable time; the data were collected in 1985 and 1986 and the final article (Hofstede et al.) was published in 1990. In 1991 the first edition appeared of a student textbook Cultures and Organizations: Software of the Mind which integrated our findings about organizational cultures with our earlier findings about national cultures, and drew conclusions about their implications for management.
The fact that national cultures were found to differ primarily in their values, and organizational cultures in their practices, has profound implications for the management of culture. National cultures are rooted in values acquired in our childhood, they are passed on from generation to generation, and their study belongs to anthropology. They change over time because of outside influences, following their own logic; they cannot be changed according to anyone’s plan, neither by political nor by religious nor by business leaders. International managers should see them as the material they have to work with. From an organizational point of view, every national culture has its strengths and its weaknesses, and these should be taken into account when management sets international strategies. On the other hand, organizational cultures, rooted in practices that can be learned and unlearned throughout people’s lives, are basically changeable, and their study belongs to organization sociology. They were created, usually unconsciously, by the organizations’ founders and early members; their development was influenced by managers and other significant members; they can be changed and monitored, given enough time, money and management attention.
International companies and international organizations always consist of members with different national values. The way they function is through a shared company or organization culture based on common practices. Establishing, monitoring and adapting corporate or organizational practices is a core strategic task for international management. Proper practices are what keeps multinationals together.
Some authors refer to national ‘management’ and ‘leadership’ cultures. However, in national cultures, all spheres of life and society are interrelated: family, school, job, religious practice, economic behavior, health, crime, punishment, art, science, literature, management and leadership. So there is no separate national management or leadership culture—management and leadership can only be understood as part of a larger national culture, and international managers should understand the cultures of the nations their organization operates in.
Companies and organizations function through a shared company or organization culture based on common practices
Comparing cultures or comparing individuals
Dimensions of national cultures describe national societies; dimensions of organizational cultures describe organizations. A common error is to apply these dimensions to the individuals within these societies and/or these organizations. A society is a symbiosis of very different individuals; so is an organization. Applying characteristics of societies or of organizations to the individuals within them is known in sociology as the ‘ecological fallacy’ (Robinson 1950). National cultures are the result of the interaction of different individuals. Statistically, national culture dimensions are calculated from questions that correlate at the national level (which means national mean scores or national percentages of answers on these questions are strongly correlated), but the same questions usually do not correlate across individuals; they may even show a reverse relationship, as the individuals in a society often supplement each other. The same is true for organizational culture dimensions.
Comparing mental programs of individuals is the subject of personality psychology. Historically, many armchair theories of personality dimensions have competed, but empirical research across a large number of countries since the early 1990s (Mc Crae and John 1992) has found five universal personality dimensions, known as the “Big Five”:
- Openness to experience
Their first letters produce the acronym OCEAN as an aid to memory.
In the early 2000s, Big Five author Robert McCrae compared national standards on his five dimensions (mean scores for standard samples from the national populations) for more than 30 countries and found these to be significantly correlated with the Hofstede dimensions of national culture (Hofstede and McCrae 2004). For example, 39% of the differences in national standards for “Extraversion” were explained by the national culture dimension of Individualism; 31% of the differences in Neuroticism were explained by Uncertainty Avoidance, and 55% by a combination of Uncertainty Avoidance and Masculinity.
A statistical link between the results of our organizational culture study and the Big Five personality dimensions was demonstrated in Hofstede, Bond and Luk (1993). In the organizational culture study, different individuals within the same organizational unit did not necessarily give identical answers to questions about how they saw their organization’s practices. The organizational culture study had not looked at these differences between individuals: its concern was with differences between organizational units. Michael Bond and Chung-Leung Luk re-analyzed the data to find out in what ways individuals’ answers differed after organization culture differences were eliminated. They showed that the answers of individuals in this case differed along six dimensions of individual personality, and that five of these closely resembled the Big Five. The sixth had no equivalent, but in later years extensions of personality research to Asia suggested that for true universality the Big Five should be extended with a sixth: Dependence on others (Hofstede 2007), and this supplies the missing equivalent for the sixth individual dimension from the organizational cultures study.
In spite of evidence and reason, quite a few articles are still published in which cross-societal dimensions, especially Individualism versus Collectivism, are applied to describe individuals. A review article “of empirical research incorporating Hofstede’s cultural values framework” by Kirkman, Lowe and Gibson (2006) listed 180 studies, of which no less than 100 had applied societal culture dimensions to individuals. Nearly all of these appeared in the United States, where the concept of the free individual is strong and the concept of society is weak. But confusing societies with individuals does not make sense, neither conceptually nor statistically. It can also lead to unwanted stereotyping. Individuals have no personal culture but do have individual personalities, partly influenced by the culture in which they grew up, but with a large range of personal variance due to many other factors.
Conclusion: Gardens, bouquets, and flowers
The two research projects described illustrate the rich possibilities of empirical multilevel research. The national culture project started from what had been supposed to be individual psychological data and aggregated them to the society level. Related to basic anthropological dilemmas, they caused a paradigm shift for cross-cultural research and proved relevant to a variety of other social science fields. A quarter of a century later they even supplied new insights into personality psychology, so that the project had come full circle. In the second project, a study started anthropologically led into organization sociology; and when re-analyzed across individuals it reconfirmed dimensions from personality psychology.
The social sciences have compartmentalized the study of the social world, and visits to neighboring disciplines or even sub-disciplines are too seldom encouraged. Some social scientists are even unaware of the level at which they operate. But whether we like it or not, we live in national societies, belong to organizations, and have our own personalities, and these relate to each other. Disciplinary parochialism and level myopia do not only make the social sciences sterile; they also make them dull. It is exciting to explore more than one level of the social reality. Societies are the gardens of the social world, organizations the bouquets, and individuals the flowers; a complete social gardener should be able to deal with all three.
Deal, T. E. and A. A. Kennedy. 1982. Corporate Cultures: The Rites and Rituals of Corporate Life. Reading, MA: Addison-Wesley.
Hofstede, Geert. 1980. Culture’s Consequences: International Differences in Work- Related Values. London: Sage.
Hofstede, Geert. 1991. Cultures and Organizations: Software of the Mind. London: McGraw-Hill.
Hofstede, Geert. 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations. Thousand Oaks, CA: Sage.
Hofstede, Geert. 2007. “A European in Asia.” Asian Journal of Social Psychology 10: 16–21.
Hofstede, Geert, and R. R. McCrae. 2004. “Personality and Culture Revisited: Linking Traits and Dimensions of Culture.” Cross-Cultural Research 38, 1: 52–88.
Hofstede, Geert, M. H. Bond and C. L. Luk. 1993. “Individual Perceptions of Organizational Cultures: a Methodological Treatise on Levels of Analysis.” Organizational Studies 14: 483–503.
Hofstede, Geert, G. J. Hosfstede and M. Minkov. 2010. Cultures and Organizations: Software of the Mind. New York: McGraw-Hill.
Hofstede, Geert, B. Neuijen, D. D. Ohayv, and G. Sanders. 1990. “Measuring Organizational Cultures: a Qualitative and Quantitative Study across Twenty Cases.” Administrative Science Quarterly 35: 286–316.
Inkeles, A., and D. J. Levinson. 1969 . “National Character: The Study of Modal Personality and Socio-Cultural Systems.” In The Handbook of Social Psychology, edited by G. Lindzey and E. Aronson. Reading, MA: Addison-Wesley.
Kirkman, B. L., K. B. Lowe and C. B. Gibson. 2006. “A Quarter Century of Culture’s Consequences: a Review of Empirical Research Incorporating Hofstede’s Cultural Values Framework.” Journal of International Business Studies 37: 285–320.
Kuhn, T. S. 1970. The Structure of Scientific Revolutions. Chicago, IL: University of Chicago Press.
Levitin, T. 1973. “Values.” In Measures of Social Psychological Attitudes, edited by J. P. Robinson and P. R. Shaver. Ann Arbor, MI: Survey Research Center, Institute for Social Research, University of Michigan.
McCrae, R. R., and O. P. John. 1992. “An Introduction to the Five-Factor Model and its Applications.” Journal of Personality and Social Psychology 60: 175–215.
Peters, T. J., and R. H. Waterman Jr. 1982. In Search of Excellence: Lessons from America’s Best-Run Companies. New York, NY: Harper and Row.
Robinson, W. S. 1950. “Ecological Correlations and the Behavior of Individuals.” American Sociological Review 15: 351–357.