There is something irresistible about large international studies. Seventy countries. Three decades. Thousands of data points. The kind of research that immediately gives the impression that we are getting closer to answering a big question. In this case, what does cultural change do to the mental health of children and young people? Does a more individualistic society lead to more anxiety? The topic of individualism, anxiety, and children is increasingly important in the context of these cultural shifts. And what about the gradual decline, in many places, of a shared religious framework?

A recent study in Developmental Science by Kulisch and colleagues tries to do exactly that. The researchers examine how values of socialisation change over time. They also check whether those changes are associated with anxiety disorders in children and adolescents. The intuition is familiar. In many societies, there has been a growing emphasis on independence, autonomy and self-development, and less on obedience, religion or collective values. It is not uncommon to hear the suggestion that this shift may come at a cost.

One strength of this study is that it goes beyond simple snapshots. The authors combine data from the World Values Survey, health statistics and development indicators, and follow 70 countries over more than thirty years. That alone makes it more informative than many earlier studies that rely on cross-sectional correlations.

And yet, this is where the nuance begins. What the study does not find is just as important as what it does find. Globally, there is no clear relationship between a stronger emphasis on independence in upbringing and higher rates of anxiety disorders among young people. The simple story that “more individualism leads to more anxiety” does not hold when you look across countries.

The picture becomes more interesting when you zoom in. In so-called WEIRD countries – Western, educated, industrialised, rich and democratic — the researchers find a weak positive association. A stronger emphasis on independence is linked to slightly higher anxiety rates. In non-WEIRD countries, this pattern is absent and, in some cases, even reversed.

That in itself is an important signal. It suggests that cultural values do not operate in isolation but interact with context. What may be burdensome in one setting could be supportive in another.

Another finding stands out. Countries where “religious values” remain more prominent in child-rearing tend to show slightly lower levels of anxiety among young people. A similar pattern appears in the individual-level data from a US cohort. It is not so much the religiosity of parents, but the broader social norm that seems to be associated with lower anxiety.

This is also the point that attracted attention in press coverage. Some readers may see this as confirmation that something has been lost. Others may find it hard to believe.

This is precisely where it becomes important to slow down. First, the effects are small. Statistically significant, but small. That means they are sensitive to interpretation, context and noise. Second, the measures are indirect. “Culture” is derived from a single survey question about which qualities children should learn. “Anxiety disorders” are based on large international estimates that combine multiple data sources. That is not inherently wrong, but it is far from a direct measurement.

More importantly, these analyses operate at the country level. They show associations between countries that change over time, not between individual children. This brings us to the classic problem of the ecological fallacy: what holds at the group level does not necessarily hold at the individual level.

There are also alternative explanations to consider. Take the finding on religion. It is entirely plausible that more religious contexts offer greater social cohesion, structure and a shared sense of meaning. But it is equally plausible that in such contexts, anxiety is diagnosed less frequently or expressed differently. The study itself acknowledges that stigma and reporting differences may play a role.

The same applies to trends in anxiety disorders. Are we observing a real increase, or are we simply getting better at identifying, measuring and labelling these problems? The type of data used here cannot resolve this question.

To be clear, this does not make the study worthless. On the contrary, it is valuable precisely because it challenges an overly simple narrative and replaces it with something more complex. Cultural change does not seem to have uniform effects. Its impact appears to depend on context, economic development and social structures.

Based on this study, we cannot conclude that individualism is bad or that religion is good. What we can take away is that children grow up within a web of expectations, norms and structures. Changes in that web may matter, but rarely in a straightforward way.

Perhaps that is the most important lesson. When we look for simple explanations to complex problems, we are unlikely to find them. And when research seems to offer one, it is usually worth taking a closer look.