The Hofstede Model in Context
Other Applications of the Dimensional Paradigm
When Culture's Consequences appeared in 1980, it represented a new paradigm in social science research: analyzing survey-based values data at the national level and quantifying differences between national cultures by positions on these dimensions. Like other new paradigms, it initially met with rejection, criticism and ridicule next to enthusiasm. By the 1990s the paradigm had been taken over by many others, and discussions shifted to the content and number of dimensions. The paradigm inspired a number of other studies into dimensions of national cultures.
Many projects further explored the dimension of individualism versus collectivism. From all the Hofstede dimensions, this one met with the most positive reactions among psychologists, especially in the U.S.A. which happened to be the highest scoring country on it. Individualism/Collectivism scores were strongly correlated with national wealth which led some people to the conclusion that promoting individualism in other cultures would contribute to their economic development. In fact, data show that the causality is most probably reversed: wealth tends to lead to individualism. The individualism in U.S. culture also led people to studying it at the individual level (comparing one person to another), not at the level of societies. In this case it is no longer a dimension of culture but an aspect of personality. Also there is no more reason why individualism and collectivism need to be opposite; they should rather be considered separate features of personality. An extensive review of studies of individualism at the individual level was published by Oyserman, Coon and Kemmelmeier. Comparing these studies across societies they found a different ranking of countries from the Hofstede studies; but Schimmack, Oishi and Diener proved this was due to a methodological error: Oyserman et al. forgot to control for acquiescence (response set), and the acquiescence in their data was significantly negatively correlated with the object of their study which made their results random.
The cultural focus on the Individualism versus Collectivism dimension led Triandis to splitting it into horizontal and vertical individualism. This split overlooks the fact that the Hofstede dimension of large versus small Power Distance already covered the horizontal/vertical aspect quite satisfactorily. From my point of view the horizontal/ vertical distinction for Ind/Col as a dimension of culture is redundant. It may be useful at the individual level, but this is for others to decide.
Like individualism and collectivism, the terms masculinity and femininity have also been used for describing values at the individual level. Earlier studies by U.S. psychologist Sandra Bem showed already that in this case masculinity and femininity should again rather be treated as separate aspects than as opposite poles.
An important alternative application of the dimensional paradigm was developed by the Israeli psychologist Shalom Schwartz. Borrowing mainly from the work of U.S. psychologist Milton Rokeach who studied values of U.S. individuals, Schwartz composed a list of 56 values. Through a network of colleagues he collected scores from samples of elementary school teachers and of college students in over 50 countries. Respondents scored the importance of each value "as a guiding principle in my life". Schwartz at first assumed the same dimensions would apply to individuals and to countries, but his data showed he needed different classifications at different levels. At the country level he distinguished seven dimensions: Conservatism (later rebaptized "Embeddedness"), Hierarchy, Mastery, Affective autonomy, Intellectual autonomy, Egalitarianism and Harmony. Country scores for teachers published by Schwartz in 1994 were significantly correlated with the IBM scores for Individualism, Masculinity and Uncertainty Avoidance.
Another large scale application was the GLOBE (Global Leadership and Organizational Behaviour Effectiveness) project, conceived by US management scholar Robert J. House in 1991. At first House focused on leadership, but soon the study branched out into other aspects of national and organizational cultures. In the period 1994- 1997 some 170 voluntary collaborators collected data from about 17,000 managers in nearly 1,000 local (non-multinational) organizations belonging to one of three industries: food processing, financial services, and telecommunication services, in some 60 societies
throughout the world. In the preface to the book describing the project, House writes "We have a very adequate dataset to replicate Hofstede's landmark study and extend that study to test hypotheses relevant to relationships among societal- level variables, organizational practices, and leader attributes and behavior".
For conceptual reasons GLOBE expanded the five Hofstede dimensions to nine. They maintained the labels Power Distance and Uncertainty Avoidance (but not necessarily their meaning). They split Collectivism into Institutional Collectivism and In- Group Collectivism, and Masculinity-Femininity into Assertiveness and Gender Egalitarianism. Long Term Orientation became Future Orientation. They added two more dimensions: Humane Orientation and Performance Orientation. The nine dimensions were covered by 78 survey questions, half of them asking respondents to describe their culture ('as is') and the other half to judge it ('should be'). GLOBE thus produced 9 x 2 = 18 culture scores for each country: nine dimensions 'as is' and nine dimensions 'should be'.
In an evaluation of the GLOBE project, I re-factor analyzed the country scores on GLOBE's 18 dimensions. Five meta-factors emerged, of which the strongest, grouping seven of the 18 measures, was highly significantly correlated with GNP per capita and next with the Hofstede Power Distance dimension. Three more meta- factors were significantly correlated with respectively the Hofstede Uncertainty Avoidance, Individualism and Long Term Orientation dimensions. The GLOBE questionnaire contained very few items covering Masculinity in the Hofstede sense, but whatever there was belonged to the fifth meta-factor. The results show that in spite of a very different approach, the massive body of GLOBE data still reflected the structure of the original Hofstede model. The GLOBE research has provoked an extensive debate in the literature, but I have seen few applications relevant for practical use by cross-cultural practitioners. Minkov and Blagoev have tried to validate each of GLOBE's 18 dimensions by testing their nomological networks (correlation patterns with variables from other sources). The largest number of GLOBE's mutually correlated dimensions can be considered useful as facets of Hofstede's Individualism/Collectivism; some have enriched insights into Hofstede's Power Distance dimension, and GLOBE's Assertiveness "should be" provides some new elements. GLOBE's Humane Orientation and Performance Orientation, both "as is" and "should be" cannot be meaningfully validated at all.
An author sometimes cited as having researched dimensions of national culture is the Dutch management consultant Fons Trompenaars. He distinguished seven conceptual dimensions, the first five borrowed from Parsons and Shils and the last two from Kluckhohn and Strodtbeck which he applied to the level of nations. Trompenaars collected a database of survey items related to these dimensions, but in the only statistical analysis of his data published so far, applying Multidimensional Scaling to some 9,000 questionnaires, only two interpretable factors emerged, both correlated with Hofstede's Individualism, one of these also with Power Distance. The only country scores that could be based on Trompenaars' data refer to these two flavors of individualism. Trompenaars' claim to seven dimensions therefore lacks empirical support.
The World Values Survey has been described above. Although the search for dimensions was not a primary purpose of this study, WVS director Ronald Inglehart in an overall statistical analysis found two key country-level factors which he called: 'Well-being versus survival' and 'Secular-rational versus traditional authority'. Well-being versus survival correlated with a combination of Individualism and Masculinity; Secular-rational versus traditional authority negatively with Power Distance.
Michael Minkov issued an extended and updated version of his 2007 book in a new volume Cultural Differences in a Globalizing World. For the dimensions Exclusionism versus Universalism and Monumentalism versus Flexumility, country scores have been re-calculated from partly different sources, for 86 countries for exclusionism and for 43 countries for monumentalism. Indulgence versus Restraint has been reversed and renamed Industry versus Indulgence; scores for 43 countries have been based on a slightly different choice of WVS items. The old and new versions of these three dimensions are still strongly correlated, in the case of Indulgence obviously negatively.
A unique feature of the new book is the addition of a dimension not based on survey questions but on a statistically strong cluster of national statistics: murder rates, HIV (AIDS) rates, adolescent fertility rates and low average IQ (Intelligence Quotient, explainable from low education levels). This can be used for validation of dimensions based on survey items. Minkov called it Hypometropia versus Prudence; hypometropia is a medical term for short-sightedness, which he borrowed to avoid an a priori depreciating term. He calculated hypometropia scores for 80 countries. It correlates significantly with Minkov's Exclusionism and Monumentalism. From the six dimensions in Hofstede et al. only Individualism correlates significantly negatively with hypometropia, across 55 overlapping countries.