National Culture: Hofstede’s Dimensions
Hofstede’s model scores a country — not a person — on six value dimensions, each normalised to 0–100 [1]. They come from a 1960s–70s survey of IBM employees: workplace-values questionnaires aggregated by country, factor-analysed into four dimensions, later extended to six.
1. The six dimensions
US and France show how far comparable engineering cultures sit apart.
| Dimension | A high score means | US | France |
|---|---|---|---|
| Power Distance (PDI) | hierarchy and formal approval accepted; juniors defer publicly | 40 | 68 |
| Individualism (IDV) | personal goals and individual credit over group loyalty | 60 | 74 |
| Motivation toward achievement (MAS) | competition and visible success over consensus and balance | 62 | 43 |
| Uncertainty Avoidance (UAI) | rules, specs and formal procedure over learn-as-you-go | 48 | 76 |
| Long-term Orientation (LTO) | deferred payoff — architecture, debt repayment — over quick wins | 50 | 60 |
| Indulgence (IVR) | leisure and expressiveness over restraint and entitlement | 68 | 48 |
Hofstede called the third dimension Masculinity/Femininity; The Culture Factor now calls it motivation towards achievement. The dimensions are complementary lenses: high UAI with high PDI amplifies the preference for formal approval beyond what either predicts alone. Read the scores relatively — France at 76 on UAI against the US 48 says the French half of a team wants the test plan written down, not that any French engineer does.
Source: Hofstede Insights.
2. What to do with them
Use them to anticipate friction and design practices that serve both ends. Advance agendas, architecture decision records, runbooks and post-incident reports reassure high-UAI and high-LTO colleagues at no cost to the others; naming who decides, and by when, settles the PDI gap; individual objectives paired with team OKRs keep credit attributable without a leaderboard. Where public dissent is unlikely, add an anonymous channel.
3. A worked example
Example — a mid-sprint scope change. A product manager asks a mixed US–French squad to swap two stories mid-sprint for a market window. The American engineers start that afternoon; the French engineers want an impact assessment and a formal schedule change first, and the sprint stalls in an argument about process. France scores 76 on Uncertainty Avoidance against the US 48 and 68 on Power Distance against 40: a change with no written impact and no named approver reads as responsiveness to one half of the team and as an unmanaged risk to the other.
The fix was two lanes. A change inside a squad’s own code takes the fast lane, priced at a rollback plan and a test; anything touching a cross-team API needs a one-page change request and a named approver. The same logic fixed code review: open discussion on the pull request, every comment written down, module owner decides ties.
How solid is this?
- The IDV and LTO scores above are the post-2023 revision. In October 2023 The Culture Factor replaced its Individualism and Long-Term Orientation scores with figures from Michael Minkov’s work, citing the age and representativeness of the original IBM data [2]. The new index rests on a 2015 survey of 55 countries plus World Values Survey data for 47 more and correlates with the original at r = .75. The United States scored 91 on Individualism originally and 60 after the revision; Japan now scores 62, above the US, inverting the most-repeated contrast in cross-cultural teaching. If a textbook disagrees, check the vintage.
- The items do not hold together at the individual level. Venaik and Brewer [3] show Hofstede’s three Uncertainty Avoidance items correlating 0.58, 0.46 and 0.44 between countries but 0.14, 0.00 and −0.11 between individuals; the same holds for GLOBE. A score built from unrelated items, they conclude, is “as meaningless in describing individuals as would be a scale … created by adding an individual’s age, height and income.” Country scores are valid between countries and invalid within them.
- The current instrument performs poorly. Gerlach and Eriksson [4] tested the VSM 2013 scales across 22,863 people in 57 countries and found Cronbach’s alpha of 0.04, −0.71 and 0.31 against the conventional 0.70 threshold — a negative alpha means the items pull against each other. Hofstede’s own 1980 figures were 0.84 and 0.77. Their country scores correlated only 0.14–0.28 with the official ones, while the same samples reproduced nationally representative World Values Survey data at r = 0.65 — sound samples, faulty instrument.
- What follows. National scores are statistical, not deterministic: variation within a country is wide and occupational cultures differ from the average. As heuristics for anticipating friction they stay useful; as measurements of a person they do not work.
Acknowledgments
This page adapts material from lectures by Eduardo Miranda and David Root [5] on software project management.
References
- G. Hofstede,“Dimensionalizing cultures: The Hofstede model in context,” Online readings in psychology and culture, vol. 2, no. 1, p. 8, 2011.
- A. Komisarof and P. Akaliyski,“New Developments in Hofstede’s Individualism-Collectivism: A Guide for Scholars, Educators, Trainers, and Other Practitioners,” International Journal of Intercultural Relations, vol. 107, p. 102200, 2025, doi: 10.1016/j.ijintrel.2025.102200.
- S. Venaik and P. Brewer,“Critical Issues in the Hofstede and GLOBE National Culture Models,” International Marketing Review, vol. 30, no. 5, pp. 469–482, 2013, doi: 10.1108/IMR-03-2013-0058.
- P. Gerlach and K. Eriksson,“Measuring Cultural Dimensions: External Validity and Internal Consistency of Hofstede’s VSM 2013 Scales,” Frontiers in Psychology, vol. 12, p. 662604, 2021, doi: 10.3389/fpsyg.2021.662604.
- D. Root,Managing Software Development. Carnegie Mellon University / Innopolis University lecture series, 2014.
Disclaimer: AI is used for text summarization, polishing and explaining. Authors have verified all facts and claims. In case of an error, feel free to file an issue.