The country told you nothing

Imagine two made-up countries: Northland and Southland.

We have measured one trait in each: how much people there like to make their own choices, rather than go along with the group. Everyone gets a score from 0 to 100.

On average, Northland comes out a bit lower, and Southland a bit higher. That gap between the two national averages is real, and researchers can measure it.

Now let us meet one real person from each country, and see what that average can, and cannot, tell us about them.

Northland44average, out of 100
Southland56average, out of 100
Nils grew up in Northland. Sofia grew up in Southland. Going on their countries alone, who prizes making their own choices more?
Draw some random pairs and watch how often the person from lower-scoring Northland still comes out higher.

This is the ecological fallacy. Northland and Southland really do differ on average, but the spread within each country is so wide that the two almost entirely overlap. Draw one person from each at random and, roughly a third of the time, the person from the lower-scoring country comes out higher.

So the national average is close to useless for predicting the individual in front of you (Brewer and Venaik, 2014). And remember: even averages like these were mostly measured on a narrow, unusual slice of humanity, wealthy Western samples, so "what people are like" rewards caution (Henrich, Heine and Norenzayan, 2010).

Use a cultural score as a weak hint about a group, never as a label for a person.