Imagine a house where footsteps supposedly echo through the hallways at night and doors close in empty rooms.
If you had to spend a night there, would explanations involving old floorboards and drafts be enough to settle your doubts?
As it happens, trusting a scientific explanation doesn’t necessarily close the door on ghosts.
A US study found that, among people who retained religious ties, an outlook combining support for science and reason with independence from religious authority was associated with stronger belief in magic, including ghosts and curses.
The researchers wanted to understand what happens to supernatural beliefs as religious institutions lose influence. When people question church teachings, do they also leave ghosts, curses, and unseen forces behind? Or do they begin choosing their beliefs for themselves?
Sociologists Paul Froese of Baylor University and Isaiah King of Carson-Newman University, together with Ruiqian Li of American University, investigated these questions by analyzing survey responses from 1,812 adults in the 2025 Baylor Religion Survey.
Gallup collected the answers online and by mail between January and March. Responses were weighted to better reflect the US adult population.
The team examined an outlook they called “secular rationalism”.
This measure combines the view that science is the only way to understand all aspects of the world with an emphasis on reason and human experience in moral decisions, independent interpretation of spiritual teachings, and disinterest in religious discussions.
Scoring highly didn’t necessarily mean cutting all religious ties. Some participants still believed in God, identified with a religion, or attended religious services.
The researchers compared these views with three categories of supernatural belief. They assessed miracles, magic, and everyday superstitions separately, distinguishing beliefs involving divine intervention from those that require no god or shared religious doctrine.
Their magic score included belief in ghosts, curses, spirits inhabiting objects, communicating with the dead, predicting the future, and magic itself. Superstitions included habits such as avoiding words that might jinx a hoped-for outcome.
The responses showed that people with higher secular rationalism scores were less likely to believe in miracles. Their scores for magic and superstition, however, were higher.
Looking more closely, the researchers found that the association between secular rationalism and stronger magical belief appeared among people who kept these religious ties. The same association wasn’t clearly evident among nonbelievers, people without a religious affiliation, and those who never attended services.
In his account of the findings, Li described the differences among people who believed in God. Around 40 percent of those embracing secular rationalism leaned toward magical belief, compared with 19 percent of those rejecting that outlook.
The researchers’ explanation centers on the distinction between personal faith and institutional authority. A religious community might accept miracles while rejecting fortune-telling or attempts to communicate with ghosts. A member who trusts their own judgment on spiritual matters might embrace those ideas anyway.
This independence could allow supernatural beliefs to persist as trust in religious authority weakens. The survey responses are consistent with this possibility, although the survey didn’t directly test how people developed their beliefs.
For sociologists, the findings suggest that questioning religious authority doesn’t always mean abandoning supernatural beliefs.
To understand what people believe, researchers may need to look beyond religious affiliation and consider how people combine faith, trust in science, and their own spiritual judgments. Two people attending the same church might have very different ideas about which unseen forces exist.
The results don’t establish that trusting science causes magical thinking. The rationalism measure wasn’t a test of scientific knowledge or reasoning ability, either.
The survey also covered only US adults and captured their views during one period. It cannot reveal whether magical beliefs are becoming more or less common. Li noted that some comparison groups were small.
Following the same people over time could show how their supernatural beliefs change alongside their relationship with religion.
The house we imagined offers no evidence for ghosts. The study explores the beliefs someone might bring through its door: confidence in scientific explanations can still leave room for the possibility of an unseen presence.
The research has been published in Social Forces.
This article was fact-checked by Peter Dockrill and edited by Peter Dockrill. While we pride ourselves on our process, we are only human. If you spot a mistake, please let us know.
