The primary objective of this Project was to determine the crash reduction potential of a quality, competency-based driver training program known as the Safe Performance Curriculum (SPC). The experimental design called for the random assignment of 18,000 volunteer high school students in DeKalb County Schools, Georgia, to one of the following: (1) Safe Performance Curriculum (SPC)--a 70-hour course including classroom, simulation, range, and on-street training; (2) Pre-Driver Licensing (PDL)--a modified curriculum containing only the minimum training required to obtain a license; (3) Control--no formal driver education in the secondary school. The sample of students were monitored for a period of 2 to 4 years after assignment to assess measures of intermediate and ultimate performance. The primary measures of ultimate performance analyzed were the numbers and types of crashes and violations the students experienced in this time frame. Comparative analyses of SPC vs. PLV vs. Control groups were then made in terms of these ultimate measures. Final detailed analyses showed a statistically significant short-term program effect, with SPC and PDL groups having significantly lower accident and violation means during the first 6 months of licensed driving. However, with an increasing time period of observation, the comparative relative differences between group means decreased and were not statistically significant. Moreover, the short-term effect was additionally offset or neutralized by the earlier licensing of SPC and PDL group students, yielding a net effect of no statistically reliable differences among SPC, PDL, and Control groups of students in overall accident and violation means.