Belief in God has accompanied humanity across cultures and through the centuries and is one of the strongest and most widespread phenomena in human society. From an evolutionary perspective, however, it is something of a mystery—practicing it requires time and effort, and yet it remains extremely widespread to this day. One explanation suggests that faith helps people cope with uncertainty, strengthens cooperation within a group, and provides a sense of psychological well-being in difficult life situations. This is confirmed by the fact that people tend to become more devout after disasters, wars, or crises.
At the same time, however, modern societies have been becoming increasingly secular over the long term, including the United States, which used to be considered an exception. The so-called theory of existential uncertainty explains this decline by suggesting that as material security increases, people no longer feel the need to turn to faith as a source of support in an uncertain world. A new study, in which our member Petr Palíšek participated, focused precisely on this topic: it examined whether a family’s material security during early adolescence can predict the strength of young Americans’ faith in God ten years later.
The study was conducted by a three-member team of researchers led by Martin Lang from the Laboratory for the Experimental Research of Religion (LEVYNA) at Masaryk University. Petr Palíšek from INPSY and Radim Chvaja also contributed to the study. The researchers used data from the long-term National Study of Youth and Religion, which tracked more than 3,000 American teenagers from 2002 until they turned 25. In the end, they included in the analysis only those participants who identified as Christian and believed in God at the start of the study—the final sample thus consisted of 2,832 people. The researchers used parental income at the start of data collection as an indicator of material well-being, while they tracked the development of faith based on repeated responses to the question of whether participants believed in God.
From Certainty to Doubt
The study’s results showed that a relationship between material well-being during adolescence and later belief in God does indeed exist—though it is a weaker, yet no less interesting, effect. In a model where the authors examined only the relationship between parental income and faith across time points, it turned out that the higher the material well-being during adolescence, the more likely participants were to stop believing in God ten years later. Among participants with low material security, only 5% of those who believed in God at the start of the study stopped believing in young adulthood, whereas in the group with high material security, the figure was 11%. When the authors added control variables such as parental education or ethnicity to the model, the effect weakened but remained significant.
Parental Role Models Carry More Weight Than Education
It is also interesting to note what the researchers found did not hold true. The participants’ level of education—which other studies had considered a possible predictor of loss of faith—did not, on its own, explain a decline in faith; people with a college degree did not drift away from their faith more often than those with less education. Conversely, a stronger predictor of faith than family income alone turned out to be the frequency with which parents practiced religious rituals—that is, the extent to which they credibly “demonstrated” their faith to their children. This aligns with the theory that children adopt faith primarily when their parents live it to the fullest themselves and thus serve as role models for their children. The authors thus conclude that material security in childhood plays a role in the loss of faith in adulthood, but it is certainly not the sole or primary driver of global secularization.
Recommended citation:
Lang, M., Palíšek, P., & Chvaja, R. (2026). The longitudinal associations of material security and belief in God in young Americans. Evolutionary Human Sciences, 8, e2. doi:10.1017/ehs.2025.10035
