Background: There is a negative relationship between education and religiosity and a positive relationship between education and acceptance of evolution, but how this manifests in college students who differ in degree of religiosity and prior educational experiences is unclear. We focused our study on the relative importance of education and religion on evolution understanding for college students at a large, public university in the Deep South. Methods: We used a structural equation model incorporating both acceptance and knowledge of evolution to evaluate the relative influence of religion and education on evolution understanding of 2,999 surveyed students. We further focused on acceptance of evolution and academic level, college major, high school experience, religion, and religiosity. We conducted pre and post course evaluations in three biology classes, and finally we tested the relationships between the quality of K-12 state science standards and states' religiosity and educational attainment. Results: We found that the degree of religiosity mattered significantly more than education when predicting students' understanding of evolution. When we focused on acceptance of evolution only, students taught evolution or neither evolution nor creationism in high school had significantly higher acceptance than those taught both evolution and creationism or just creationism. Science majors always outscored non-science majors, and not religious students significantly outperformed religious students. Highly religious students were more likely to reject evolution even though they understood that the scientific community accepted the theory of evolution. Overall, students in two of three biology classes increased their acceptance of evolution, but only those students that seldom/never attended religious services improved. K-12 state science standard grades were significantly and negatively correlated with measures of state religiosity and significantly and positively correlated with measures of state educational attainment.