Globally, men, and particularly young men, are involved in more road traffic crashes than women, which may be due to a greater tendency to engage in risky behaviors. Understanding and explaining this "gender specificity" in risky behaviors, specifically among young drivers, has become a major public-health issue. The present study extends research on the effect of gender identity on risky driving behaviors by investigating the effect of sex, sex-stereotype conformity and gender group identification on self-reported driving behaviors among young drivers. 75 young drivers (28 males, 47 females) filled in a form including a series of scales assessing gender group identification by measuring three components of gender identity (typicality, contentedness and centrality), a French version of the Bem Sex-Role Inventory, an extended version of the Driver Behaviour Questionnaire (assessing violations, dangerous and inattention errors and positive behaviors) and questions about mobility and accident history. The effects were tested on outcome variables by using hierarchical regression analysis. It was found that sex (being female) only predicted the inexperience errors, while the femininity score negatively predicted the number of accidents. No effects of maleness and masculinity were observed in other driver behaviors, contrary to what was expected. Gender identity variables only had an effect among males, with typicality positively predicting dangerous errors and contentedness negatively predicting positive behaviors. Lastly, results showed that gender identification appears to be associated with low conformity to feminine stereotypes among males. Hypotheses were not confirmed but results underlined the importance of taking gender identity variables into account when explaining risk-taking differences between and within gender groups. Implications of these results are discussed.