Published online by Cambridge University Press: 18 August 2015
We argue that personal religiosity and political religiosity are distinct attributes of a political candidate. Personal religiosity reflects a candidate's level of personal religious commitment and political religiosity reflects the candidate's policy regarding separating versus blending religion and politics. The paradoxical religiosity hypothesis predicts that, within a democracy, personal religiosity will increase voters' endorsement of a candidate whereas political religiosity will decrease voters' endorsement. We test this hypothesis comparatively in two experiments, one performed within a long-standing democracy containing predominantly Christian voters (the United States), and the other within a more recent democracy containing predominantly Muslim voters (Indonesia). We demonstrate the robustness of the paradoxical religiosity effect and its persistence across the two countries, suggesting that Muslim Indonesians are no less capable than Christian Americans in separating the sacred and the secular.