A lack of evidence

Most scientists do not believe in God for two simple reasons: they see no evidence for him; and they find no need to include supernatural elements in the models they build to describe their observations. Still, they are largely content to assign religion and science to separate domains, which Stephen Jay Gould calls "non-overlapping magisteria". This allows non-believing scientists to dodge the topic of religion altogether, thus avoiding jeopardizing what is most precious to them – their funding. It also permits scientists who do believe in God to compartmentalize their thinking and never discuss science and religion in the same breath.