Where exactly is the problem? Critically evaluate Hofstede and his national culture theory
National culture has been defined in hundreds of ways (Erez and Early, 1993), but the most widely utilized dimensions of culture are the five presented by Hofstede (Crotts and Erdmann, 2000). Hofstede is the pioneer in this field and his national cultural theory is so famous that it is a milestone in this field.
Hofstede’s national culture theory has been developed and perfected though these 30 years. However, ever since the theory was published, the critiques of his theory have never stopped.
This article will critically evaluate Hofstede’s national culture theory in three areas: Hofstede’s personal experience, data collection from IBM and his theory’s findings on national culture. In the personal experiences section I explore how the limitation of his experience could affect the results of his study and also include the example of the Cold War and the reunification of Germany. In the data collection section I critique the questionnaire on professional (originally for IBM), time limitation (the data was ‘old-fashioned’) and sampling (only one company’s male employees, small number of sample) grounds. In the section on theory building I argue that the theory may not be reliable due to the insufficient data. He also made many of other assumptions. Finally I give several examples of change: namely that our life has been influenced by the internet in different ways, that China has become an economic centre, and that many change in government policies have affected people’s way of life. Furthermore, culture change never stopped in China.
Limitation of personal experiences:
Born in Haarlem, the Netherlands in 1928, Geert Hofstede earned his Masters degree in Mechanical Engineering from Delft Technical University. During the period between 1965 and 1971, he worked for IBM and launched a questionnaire about its different subsidiaries all over the world. This formed the basis for his national culture theory. After Hofstede left IBM, he became a professor at various international management schools including IMD, Lausanne, INSEAD Fontainebleau, and the European Institute for Advanced Studies in Management, Brussels. It was at this time that he completed his analysis of the IBM employee survey database, leading to his definition of four dimensions of national culture and publication of his (1980 a) influential book ‘Culture’s Consequences’. (ITAP international; Hofstede, 1980 a; Hofstede Homepage; Powell, 2006.)
Hofstede had plenty of experience in both working and studying. However, this is not enough if one wants to create a theory that is suitable for every national culture all over the world. First, he had a Masters degree in Mechanical Engineering, not anthropology or a related field. It is different field and may impact or mislead his thinking about culture. Second, he was born in the Netherlands, and before 1980 when Hofstede was doing research for his book; he only stayed and worked entirely in Europe. How can he totally understand the culture of other counties on different continents? How can he distinguish between capitalism and communism? He can only truly understand culture in Europe because he was born there and worked there. Whereas, he would have difficulty understanding other cultures because he had not been there. This could be a big gap if he only analysis the questionnaire not been to other countries himself. And even now, he only has been to several countries outside of Europe. What about the rest of countries? Can he totally understand the culture in Africa and Latin America? The answer is obviously not. Even after staying in a country for several years, people may still be unable to fully understand its national culture.
Cold War and Germany:
The Cold War may have had special influence on European countries’ culture. During the Cold War, Western European countries only had a good relationship with US, and Eastern European countries had a good relationship with USSR. Culture could have been more stable at that period because countries could not communicate with each other continually.
Cold War Germany may provide an especially good example. Political activities caused Germany to be divided into two countries after World War II – Western Germany and East Germany. The Soviet Union instituted a blockade of East Berlin in 1948 to protest Western’s countries’ policies. From that point on, one country became two, and one culture was divided into two, because of the influence of capitalism and communism. People in each country had totally different ways of life. West Germany developed at a faster rate than East. So when the Berlin Wall collapsed, and West and East unite as one, West Germany was not willing to accept East Germany because of their lower level of development. (US Department of State)
When the Soviet Union crumbled, the world structure was changed. After the Cold War, countries were able to communicate frequently and this started to drive culture change, especially during this past decade. For some of the countries that belonged to the Soviet Union but joined the European Union, the changes from communism to capitalism definitely change their national culture. The PROUT Institute made the following comment about communism and capitalism. For example: human liberty under communism was limited by the primacy of the interests of the state; under capitalism, a licentious freedom of expression is permitted. Culture under communism is compelled to be consistent with the state ideology; under capitalism, mass culture serves commercial interests; it is creative but not authentic, energetic but destructive of higher values. Communism’s command economy emphasised production; capitalism’s free market economy is motivated by profit.
Data collection (Research Methodology)
The research was originally conducted by IBM, and used for IBM purpose, not academic research. Hofstede just ‘borrowed’ the data and analysed it. So the fact is undeniable; the questionnaire was designed for business purposes not for Hofstede. McSweeney (Spring, 2002) critically argued that the questions asked may not have been wide-ranging and deep enough. The consequences of not having comprehensively ‘identified’ national value sets are not merely incomplete descriptions, but more importantly inaccurate descriptions. Restricted questions and answers could miss influential values that might counterbalance or outweigh the values that were measured, so the resulting depictions of national cultures would be distorted. As the questionnaires were not designed to identify national cultures it is likely that the questions were not adequate for Hofstede purpose.
McSweeney (2002a) argued that Hofstede’s primary data was extracted from a pre-existing bank of employee attitude surveys undertaken in 1967 and 1973 within IBM subsidiaries in 66 countries. Two surveys were undertaken – around 1968-69 and repeated around 1971-73. We can see the data was old-fashioned, and it was behind the times for approximate forty years. Such as how can we debate that when the British army invited into India, and make it as a colony of ‘The Empire on which the sun never sets’; it is unlikely that India’s national culture has not changed in any form.. We should take note of the change of national culture due to both internal and external influence. Outdated data cannot give a clear picture of current culture and current international situations.
The problem for Hofstede’s analysis is that most, if not all, of these stratifications would produce response differences (Schwarz, 1999). The small sample does not present enough information. McSweeney (2002 a) indicated that the figure of 117,000 questionnaires is the combined number for both (1969 and 1971) surveys. The survey covered 66 countries, but the data used only came from 40 countries. In only six of the included countries were the number of respondents more than 1,000 in both surveys viz. Belgium, France, Great Britain, Germany, Japan, and Sweden. In fifteen countries the number was less than 200 viz. Chile, Columbia, Greece, Hong Kong, Iran, Ireland, Israel, New Zealand, Pakistan, Peru, Philippines, Singapore, Taiwan, Thailand, and Turkey. The first survey in Pakistan was of 37 IBM employees, while the second was of 70 employees (Hofstede, 1980a: 73). The only surveys in Hong Kong and Singapore included 88, 71 and 58 respondents respectively (1980a: 411).
Obviously, one or two hundred of people cannot represent a whole country or even a city. In response to criticisms of the small number of respondents in some countries (Goodstein, 1981). Hofstede has stated that: ‘if a sample is really homogeneous with regard to the criteria under study, there is very little to gain in reliability over an absolute sample size of 50. … I could therefore have done my research on 40 (countries) x 50 (respondents per country) x 2 (survey rounds) – or 4,000 respondents in total – and obtained almost equally reliable results’ (1981:65).
However, how could it possible that those out of season data can represent different countries? For example, suppose the population of the research is one hundred thousand, then 100 respondents only make up 0.1% of the population; if the population is one million, the sample size for 100 respondents is merely 0.01%. How could these results convince people? Especially this is an academic article, it has to convince people that the theory is based on reasonable data.
Just like I mentioned before, hundreds of people cannot represent whole country, just as one person within an organisation cannot represent the whole. So many problems exist: people are characterised by their different occupation, different industry, different generation, different agenda, different religion different stratum etc. Hofstede ignored those factors when he did the research. McSweeney (Spring, 2002) also indicates that people may often be individualistic within office politics, but usually they act as a collectivist in the organisations. The IBM data was effectively restricted to the workplace. Other sections of national populations – the unemployed, full-time students, the self-employed, the retired, home workers, and others – were ignored. The questions were almost exclusively about workplace issues, were completed in the workplace and not replicated in other types of locations.
Moulettes (2007) argues that embedded in Western scientific rationality, Hofstede’s model on national culture relies on a quantitative methodology. The sample consists of a group of well educated white ‘men’ from the middle classes working for the same company and sharing identical or similar occupations. Taking the sample as the norm for national culture might mislead the readers that Hofstede perceives of culture as equally distributed among men and women and thinks that there are no differences with regard to the possession of power. Women are not the only party Hofstede ignored. There are many of others. How can IBM employees represent peasants who worked in the farmland? How can IBM employees represent students who study in primary schools or universities? They cannot, because middle class employees working in IBM are quite different from both peasants and students.
Problems of analysis data
Blodgett et al (2008) stated that the original purpose of the IBM research was to create a ‘job attitudes’ instrument that could be used for ‘organisational development’. It focused on uncovering differences among IBM employees in various countries. Hofstede attributed some of the findings to cultural differences after analysing the data. Whereas the development of the job attitudes instrument was based on within-culture analyses, it occurred to Hofstede that between- cultures analyses might reveal additional insights. He began to conceive a framework. But the purpose is not practical, how can Hofstede get right results from analysis?
Over the past two decades, the economy has become global, resulting in greater heterogeneity of markets and consumers. The heterogeneity cause differences within national cultures in large part (Blodgett et al, 2008). With the change of time, the world has changed a lot. People changed their lifestyle, taste, and the perceptions of the world. Those go hand in hand with cultural change. Laroche et al. (2005, p. 282) assert that: ‘A serious limitation of national cultural indices is their high level of aggregation, which may hide important variations, including regional … and individual differences and experiences.’ Blodgett et al (2008) suggested that such indices should also be applicable when comparing diverse sub-groups within a particular country. Such as: Hispanic-Americans, Asian-Americans, African-Americans and Muslim-Americans. Those people are Americans but they have different cultures because they are totally different groups who come from different original culture.
As concluded by Riddle (2008) that national cultural models also ignore the importance of economic and social class. For many Americans, in particular, it is even more difficult to think and talk about class than about race, religion, and gender. It is dangerous to use only one class (middle class) in one company (IBM) to present all kinds of classes and companies.
McSweeney (Spring, 2002) critically state that Hofstede generalises the entire national population in each country based on a few questionnaire responses’ results. The respondents were simply certain categories of employees in the subsidiaries of a single company: IBM. There is no evidence to show that they were nationally representative. He just assumes it. Both his assumption of individuals shares a common national culture in a nation; and findings of ‘national norm’ or ‘central tendency’, or ‘average [national] tendency’ are all problematic. For example, a statement that Chinese people are old-fashioned, because there are some older Chinese people who are old-fashioned, but can this small amount of people represent all Chinese people and Chinese culture? This will be ridiculous, because Chinese are now quite fashionable, especially in big city such as Beijing and Shanghai. Generalising an entire national population on the basis of small number of questionnaire responses is equally absurd. IBM surveys within each country revealed radical differences in the answers to the same questions. There are no evidence-based reasons for assuming that the average IBM responses reflected ‘the’ national average. Hofstede’s assumptions are a mere leap of faith (McSweeney Spring, 2002).
Hofstede assumed that the attitudes expressed were not specific to workplace. But one’s attitudes may differ depending on whether they are in the courtroom, on the sports field, in the bedroom, or somewhere else. But same situation cannot exist in everywhere (McSweeney, Spring, 2002). For example, in the football yard, football fans are madness because the exciting football match, but can they do the same in their office? Of course they cannot.
Problems with the theory
As identified by McSweeney (2002), the notion of national culture in the work of Hofstede (1980a, 1983) which claim one territorially unique. Possible examples to challenge this are the United Kingdom and U.S. There are four nations within the United Kingdom – England, Scotland, Wales and Northern Ireland. Of course different nations will have different custom and traditional culture, language, food etc. the same goes the US, where fifty states compose the whole nation. Each state has its own law, different age for marriage, for drinking, for driving etc. However, Hofstede consider all of this as a single culture.
This is also defeated Hofstede’s another concept – national is uniform (1980a, 1980b); which presupposed national uniformity, and people in the nation share a common national culture so called ‘subculturally heterogeneous’ (Hofstede, 1980a). National culture is common to each individual within the country. But the example of ‘GB’ and ‘USA’ are already proved this is an unreal concept which builds on conjecture.
Denial of agency
Later, Hofstede added the theories: culture as coherent, national culture as stable, national culture as pure. (Hofstede 2001, Hofstede 2005). But McSweeney’s (2009) critical analysis of Hofstede’s theories in his article in the first par bring up an important objective: the denial of agency. For example: in culture as coherent, McSweeney (2009) argued that combinations of cultures will not be coherent; cultural coherence allows no gaps, no ambiguities for individuals, to engage with or exploit; after all, culture is not a pre-established monolith (see more detail at McSweeney (2009).
Three discrete components
Hofstede’s (1991, 1980 a) concept of the relationship among ‘organizational’, ‘occupational’, and ‘national’ culture is another fallacy. He concluded that employees from different nations in IBM have the same organisational culture and occupation culture, which allows one to analyse the differences due to national culture. That’s why he think only IBM is appropriate for his research. McSweeney (2002 a) defeated this theory just by a easy example: US accountants might be influenced by the short-termism of the US capital market compared with the German accountants possible effected by longer-termism of the German capital market. This theory ignored distinct of entry requirements, regulations, social status, structure etc. There are many competing factors that influence the same organisation or same occupation in different places, both internal and external, cultural and non-cultural influence. Hofstede treats analysing culture just he is doing mathematics. Detail of this model is in appendix 1
Examples of Change
The Internet affects culture
The Internet emerged and has developed quickly in the past thirty years. Our live have been changed by the internet and it has affected our culture in different ways. Zahir et al (2002) sew that the internet change the way we do business, obtain an education and learn other skills, gather information, bank and invest, pay bills, listen to music, see films, buy and sell things, exchange greetings and communicate with others, express views, participate in debates and are entertained. These changes are likely to affect cultures.
Segel-Brown (2007) noted that the internet can influence culture through both individuals and groups. The internet influences individuals’ daily lives by affecting individual thinking, individual behaviour, and how human needs are satisfied. According to Wallace (1999), people use different attributes judge people online. Also, as internet use has increased, people have begun not to feel guilty about downloading illegally. The internet has changed moral consciousness of people. It also allows groups to have more influence over others. For example, it is easy for one group to inexpensively publish knowledge online to influence other groups. Additionally, groups can easily find out information about others online, which can impact decision-making. The Internet has had a large effect on the culture of groups because group behaviour has changed; group formation has changed, and group dynamics are different because online identities are disposable and anonymous.
Chinese special culture:
Liu (2005) demonstrate the change in religion and culture in China as three phases: phase 1 (pre-1949): Utilitarianism and Indifference; phase 2 (post-1949): Surviving Governmental Suppression; and phase 3 (post-1979): Revival and Competition. During phase 1, in ancient China, ordinary people, believers and non-believers, held utilitarian views of religion; intellectuals followed lofty Confucianism. During phase 2, after 1949, the Communist Party’s policy was to minimize religious activity while, at the same time, claiming that Marxism is the only valid ideology. During the Cultural Revolution, the major religions were almost destroyed but did not die out completely. And after Mao became a “living Buddha figure,” this cult of personality replaced all religions and philosophies. As for phase 3, with the end of the Mao era, religions started to revive. During 1990s, economic growth improved the material life of the people, but the natural desire to seek more prosperity introduced a new idolatry: the worship of money, or materialism. The acquisition of wealth replaced moral progress as the primary criterion for judging social standing among ordinary people. The influence of materialism can still said to be a problem in China because it is undermining traditional values. The change of Chinese culture is not only on the surface but reaches the core of culture and impacts on traditional Chinese culture in a totally different way.
Chinese culture change in 1966 – the Great Proletarian Cultural Revolution
It narrowed ‘the differences between town and country, between worker and peasant, and between manual and mental labour’ (Spence, 1990, p. 644). People could feel more proud to be a worker and farmer than a student or writer. Traditional Chinese culture changed a lot during this period.
Contemporary China is the fastest developing nation on the earth, and after two decades of rapid economic development, China has become an important economic centre, especially in 2009 during the economic crises. In large cities, you can see high-rise buildings and people wearing western-style modern clothing, while consumers’ product choices match those from the most developed economies such as United States or Europe. In contrast, you will also see traditional festivals being celebrated. In rural areas and villages traditional culture is even more important, as farmers and village people still lead simple lives without modern technology, has less marketing support systems (Piron, 2006; China, an inner realm).
This assignment critically analysed Hofstede’s national Culture theory in three areas: personal experiences, data collection (IBM questionnaire) and theory finding. Each of the arguments is reasonable with support. If somebody wants to have a clear sense about a nation’s culture, he/ she should go there and make real judgement, not just analyse several unreliable questionnaires in the office. Because there are so many variables that could mislead and make people misjudge the data, leading them to make the wrong conclusion. Once again, ‘middle class’ employees in IBM can not represent everyone in a nation; they do not even represent all of IBM well. In addition, the theory is much more hypothetic and unrealistic. However, this assignment is too short to fully defeated Hofstede’s national culture theory. There are a host of articles on the topic that give unique arguments that readers can make audience judge for themselves.
- Blodgett, J. Bakir, A. and Rose, G. (2008). A test of the validity of Hofstede’s cultural framework. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/0770250602.pdf [Accessed: 11 January 2010]
- China an inner realm. The culture of China. [Online]. Available from: //library.thinkquest.org/20443/old_new.html [Accessed: 11 January 2010]
- Crotts, J. and Erdmann, R. (2000). Does national culture influence consumers’ evaluation of travel services? A test of Hofstede’s model of cross-cultural differences. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/1080100608.pdf [Accessed: 11 January 2010]
- Erez, M, Earley, P.C (1993), Culture, Self Identity and Work, Oxford University Press, New York, NY, .
- Goodstein, L. D. (1981). Commentary: Do American theories apply abroad? Organizational Dynamics. 1981, Summer, 49-54.
- Hofstede, G. (1980 a). Culture’s consequences: international differences in work-related values, Beverly Hills, CA: Sage.
- Hofstede, G. (1980 b) Motivation, leadership and organization: Do American theories apply abroad? Organizational Dynamics, 1980 b, Summer, 42-63
- Hofstede, G. (1981). Do American theories apply abroad? A reply to Goodstein and Hunt Organizational Dynamics. 1981, Summer, 63-68.
- Hofstede, G. (1983) The cultural relativity of organizational practices and theories. Journal of International Business Studies, 1983, Fall, 75-90
- Hofstede, G. & Franke, R. H. (1991). Cultural roots of economic performance: a research note. Strategic management journal. 1991. 12. pp. 165-173
- Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviours, institutions and organizations across nations, 2nd edition. Thousand Oaks, CA: Sage
- Hofstede, G. and Gert Jan Hofstede (2005) Cultures and organizations: Software of the mind, 2nd edition. New York: McGraw-Hill
- Hofstede, G.. Homepage. [Online]. Available from: //stuwww.uvt.nl/~csmeets/index.html [Accessed: 11 January 2010]
- ITAP international. Geert Hofstede – originator and licenser of the CW. [Online]. Available from: //www.itapintl.com/tools/culture-in-the-workplace-questionnaire-cw/geerthofstede.html [Accessed: 11 January 2010]
- Laroche, M., Kalamas, M. and Cleveland, M. (2005). ‘I versus “we”: how individualists and collectivists use information sources to formulate their service expectations’, International Marketing Review, Vol. 22 No.3, pp.279-308.
- Liu, P. (2005). Changing Chinese Attitudes Toward Religion and Culture: A Comparative Perspective. [Online]. February, 2005 Available from: //www.brookings.edu/fp/cnaps/events/liupeng20050201.pdf [Accessed: 11 January 2010]
- McSweeney, B. (Spring, 2002).Fundamental flaws in Hofstede’s research. European Business Forum. [Online]. Spring, 2002. Available from: //findarticles.com/p/articles/mi_hb4779/is_9/ai_n28975104/pg_4/?tag=content;col1 [Accessed: 11 January 2010]
- McSweeney, B. (2002 a) ‘Hofstede’s Model of National Cultural Differences and Their Consequences: A Triumph of Faith – A Failure of Analysis’. Human Relations, 55(1): 89-118. [Online]. November 2003. Available from: //www.rhul.ac.uk/Management/News-and-Events/seminars/McSweeney12-11-03.pdf [Accessed: 11 January 2010]
- McSweeney, B. (2009). Dynamic Diversity: Variety and Variation within Countries. Organization Studies, Vol. 30, No. 9, 933-957
- Moulettes, A. (2007). The absence of women’s voices in Hofstede’s Cultural Consequences. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/0530220601.pdf [Accessed: 11 January 2010]
- Piron, F. (2006). China’s changing culture: rural and urban consumers’ favourite things. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/0770230603.pdf [Accessed: 11 January 2010]
- Powell, S. (2006). Geert Hofstede: challenges of cultural diversity. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/0440140304.pdf [Accessed: 11 January 2010]
- PROUT Institute. PROUT Compared to Capitalism and Communism. [Online]. Available from: //www.proutinstitute.org/pdfs/PROUT_Compared_to_Cap_and_Com.pdf [Accessed: 11 January 2010]
- Riddle, P. (2008) Cultural models considered harmful. [Online]. March, 2008. Available from: //www.prentissriddle.com/talks/riddle-2008-sxsw-CulturalModels.pdf [Accessed: 11 January 2010]
- Segel-Brown, A. (2007). Effect of the Internet on Culture and Daily Living. [Online]. June 2005. Available from: //hot.carleton.ca/hot-topics/articles/internet-and-culture/ [Accessed: 11 January 2010]
- Schwarz, N. (1999) Self-reports: How the questions shape the answers. American Psychologist, 54, 93-105.
- Spence, J. (1990), The Search of Modern China, W.W. Norton & Co., New York, NY, .
- US Department of State. Berlin Crises. [Online]. Available from: //www.state.gov/r/pa/ho/time/cwr/17378.htm [Accessed: 11 January 2010]
- Wallace, P. (1999). The Psychology of the Internet. United States: Cambridge University Press.
- Zahir, S. Dobing, B. and Hunter, M. (2002). Cross- cultural dimensions of Internet portals. [Online]. Available from: //www.emeraldinsight.com.ezproxy.staffs.ac.uk/Insight/viewPDF.jsp?contentType=Article&Filename=html/Output/Published/EmeraldFullTextArticle/Pdf/1720120301.pdf [Accessed: 11 January 2010]