We use a dataset for a demographically representative sample of the Dutch population that contains a revealed preference risk attitude measure, as well as detailed information about participants’ religious background, to study three issues. First, we find strong confirmatory evidence that more...