The apparent lack of self-confidence in physical activity by females compared to males has been a recent concern of some researchers in sport psychology. Lenney (1977) suggested that females would be less confident than males when the task was male oriented or when the situation was competitive. This meta-analysis was conducted to examine the magnitude of gender differences in self-confidence in physical activity according to Lenney's assertions. An overall nonhomogeneous effect size of 0.40 favoring males was found. Although masculine tasks produced a larger effect-size difference than neutral tasks, it was also not homogeneous. Only one study employed a feminine task, resulting in a large effect size favoring females. However, the results of a regression analysis, which found that sex-type of task contributed to gender differences in self-confidence, did support Lenney's contention. Whether or not the task took place in a competitive situation did not differentially affect the magnitude of the gender differences. Age of subject and type of confidence measure employed are also discussed as possible variables contributing to gender differences in self-confidence.