© 2014 Society for Adolescent Health and Medicine. Purpose The large contribution of inexperience to the high crash rate of newly licensed teens suggests that they enter licensure with insufficient skills. In a prior analysis, we found moderate support for a direct effect of a web-based intervention, the TeenDrivingPlan (TDP), on teens' driving performance. The purpose of the present study was to identify the mechanisms by which TDP may be effective and to extend our understanding of how teens learn to drive.Methods A randomized controlled trial conducted with teen permit holders and parent supervisors (N = 151 dyads) was used to determine if the effect of TDP on driver performance operated through five hypothesized mediators: (1) parent-perceived social support; (2) teen-perceived social support; (3) parent engagement; (4) practice quantity; and (5) practice diversity. Certified driving evaluators, blinded to teens' treatment allocation, assessed teens' driving performance 24 weeks after enrollment. Mediator variables were assessed on self-report surveys administered periodically over the study period.Results Exposure to TDP increased teen-perceived social support, parent engagement, and practice diversity. Both greater practice quantity and diversity were associated with better driving performance, but only practice diversity mediated the relationship between TDP and driver performance.Conclusions Practice diversity is feasible to change and increases teens' likelihood of completing a rigorous on-road driving assessment just before licensure. Future research should continue to identify mechanisms that diversify practice driving, explore complementary ways to help families optimize the time they spend on practice driving, and evaluate the long-term effectiveness of TDP.