Research has shown that females experience higher levels of math anxiety than males. However, Goetz et al. (2012) argued that gender disparities in math anxiety may be a consequence of gender stereotypes rather than actual differences in anxiety levels. Therefore, gender differences may be confined to trait measures which are influenced by stereotypes to a larger extent. In support of this view, Goetz et al. found a difference between girls and boys only on a trait measure, but not on a state measure of math anxiety. However, the levels of state anxiety in this study were low, and the lack of a gender difference may be due to a floor effect. We report two studies examining gender differences in 5th and 6th grade male and female students using both state and trait measures of math anxiety. State anxiety was measured before, during, and after a mental arithmetic test administered online (Study 1) or as a paper-and-pencil test (Study 2). To avoid floor effects, the arithmetic problems increased in difficulty across blocks of arithmetic problems, and the state anxiety scale was designed to capture even low levels of anxiety. In both studies, state anxiety was higher in girls than in boys, although absolute levels were low in both genders. Possible reasons for gender differences in math anxiety are discussed.