This study intends to compare and contrast student and school factors that are associated with students' mathematics self-efficacy in the United States and China. Using hierarchical linear regressions to analyze the Programme for International Student Assessment (PISA) 2012 data, this study compares math self-efficacy, achievement, and variables such as math teacher support and socioeconomic status (SES) between 15-year-old students in the U.S. and in Shanghai, China. The findings suggest that on average, students from Shanghai showed higher math self-efficacy and better achievement than those of American students. However, at the student level, similar positive relationships between math teacher support and math self-efficacy and between SES and math self-efficacy were found in both locations. That is, in the U.S. and Shanghai, an increase in math teacher support predicts an increase in math self-efficacy, also higher SES is significantly associated with higher math self-efficacy. In addition, at the school level, the smaller difference in American students' math self-efficacy between higher SES school and lower SES school indicates that the U.S. is more equitable between schools than Shanghai, China in terms of students' math self-efficacy. Implications from this study indicate that improving teacher support in math class and narrowing the gap in students' self-efficacy related to school-level SES is a significant issue for the U.S. and Shanghai, China respectively.