BOSS blog 1: Do cultures change?

BOSS blogs

This is the first of a series of blogs I intend to publish here. I call them BOSS for Basics of Social Science. They are intended for a general audience of people who need science to help them solve real-world issues. The aim is to address topics that I think are important, with a view to painting the big picture. Expect occasional entries on topics such as  values, culture, or levels of analysis (individuals, groups, countries...).

I may talk about studies by others, such as the one I introduce here, or about the implications of my own work in artificial sociality. Enjoy!

Do cultures change? Yes and no

People change. The world changes. But does culture change? This is a contested issue. There is a good reason for this: it is hard to measure. Culture is defined, at least in most current definitions, as an attribute of a group or society. It has to do with the practices, values, or hidden rules, in that society. How do you measure such a thing?

Measuring the climate

Let me give an analogy. How do you measure the climate? You could measure the weather on many days on a number of dimensions: temperature, rainfall, solar radiation, and wind, for instance. Then you could aggregate the results, and you might come up with a limited number of possible climates: tropical, polar, Atlantic… You will need a lot of data. For instance, if you measure in Athens, Greece, only once and there happens to be snow cover, you might come to hasty conclusions. If you measure in Athens again the next year on the same date and it’s 35 degrees Centigrade, you might prematurely conclude that the climate has changed.

Likewise, one needs a lot of data from a lot of people to measure culture or to establish that cultures have changed.

Lots of data on culture

Fortunately, we have these lots of data today. One such source is the World Values Survey (WVS, and its European sister, the European Values Study (EVS, The WVS is a worldwide data collection effort that started in 1981 and is now in its seventh wave of data collection. Is surveys representative samples of the population of the world’s countries, and tries to use the same questions in all countries as much as possible. In 2019, its stated aim is “to help scientists and policy makers understand changes in the beliefs, values and motivations of people throughout the world.” It is a huge mine of publicly available data.

Beugelsdijk and Welzel 2018: cohort analysis

In a recent article in the Journal of Cross-Cultural Psychology, two cross-cultural scholars did a study based on WVS and EVS data that I want to discuss here. I believe it throws some light on the question “Do cultures change?”

Sjoerd Beugelsdijk and Chris Welzel had the clever idea of comparing the answers from people born in different decades. They defined five birth cohorts, cut off after twenty years: 1900 – 1920 – 1940 – 1960 – 1980 – 2000.

They selected a subset of questions on preferences and beliefs, and subjected it to an ecological factor analysis. This means they analyzed country averages, not individuals. That resulted in three factors. Here, I’ll give them a name in terms of Hofstede dimensions:  One for individualism and small power distance, one for long-term orientation and restraint, and one for something akin to uncertainty avoidance. They call these factors individualism / closeness, joy / duty, and distrust / trust. My purpose here is not to discuss these dimensions or how they were derived. Let’s remember that dimensions do not exist, but they can be useful analytical tools.

Culture in the 20th century: change and stability

Comparing across these cohorts, they find that there is substantial change on two dimensions:

  • Across the 20th century, individualism goes up (and power distance goes down)
  • Across the 20th century, long-term orientation goes down (and indulgence goes up)

This should not come as a surprise to those of a certain age that follow politics in countries across the world. The times have changed, and public attention focuses more and more on the individual and on hedonic interests.

One country changes more than the next, though. The authors also analyse the variation in change across countries. They find that economic development and generational shifts account for about half of the changes. The other half of national cultural differences, they say, can be accounted for by geography and history.

The take-home message: culture changes, yet matters

So yes, there is change, beyond reasonable doubt. But this does not mean that the structure of cross-national differences changes equally fast. Beugelsdijk and Welzel conclude that “… our finding on the stability of the countries’ relative position suggests that these measures will not be outdated any time soon and that findings using these measures will not be significantly affected by temporal variation, as long as the country scores are interpreted in a relative sense”.

This means that the study is a strong indication that dimensional models of culture have validity across the years. We should only use them to compare across countries though.

Finally, cultures are like individuals: they change, and they have continuity. We need to understand both in order to make sense of our world.


Sjoerd Beugelsdijk and Chris Welzel (2018) “Dimensions and Dynamics of National Culture: Synthesizing Hofstede With Inglehart”. Journal of Cross-Cultural Psychology 49(10) 1469-1505. DOI: 10.1177/00220221 18798508