why teen drivers go splat so much

http://www.sciencedaily.com/releases/2011/04/110419190032.htm

Scanning involves observing the surroundings far ahead of the vehicle and side-to-side, not just immediately in front of the hood. It is a higher-level skill that experienced drivers develop over time. The study authors note that developing effective ways to teach this skill sooner in the learning-to-drive process could reduce teen crash risk. Pilot tests of this type of training have shown promise in increasing hazard detection and response skills among novice drivers.

By getting very specific about the types of teen driver errors that are most likely to precede a crash, this study makes it possible to target policies, programs, driver education and other strategies to reduce those critical errors and prevent crashes from happening.

Among crashes with a teen driver error:

• Twenty-one percent occurred due to lack of scanning that is needed to detect and respond to hazards.
• Twenty-one percent occurred due to going too fast for road conditions, (for example, driving too fast to respond to others, or to successfully navigate a curve).
• Twenty percent occurred due to being distracted by something inside or outside the vehicle.

-------------------------
http://www.sciencedaily.com/releases/2010/06/100609083225.htm

D; and another reason to start high school student after the primary school kids.
Besides grades.

Results indicate that in 2008 the teen crash rate was about 41 percent higher in Virginia Beach, Va., where high school classes began at 7:20 a.m., than in adjacent Chesapeake, Va., where classes started more than an hour later at 8:40 a.m. There were 65.4 automobile crashes for every 1,000 teen drivers in Virginia Beach, and 46.2 crashes for every 1,000 teen drivers in Chesapeake.

"We were concerned that Virginia Beach teens might be sleep restricted due to their early rise times and that this could eventuate in an increased crash rate," said lead author Robert Vorona, MD, associate professor of internal medicine at Eastern Virginia Medical School in Norfolk, Va. "The study supported our hypothesis, but it is important to note that this is an association study and does not prove cause and effect."