As strange as it may sound, almost all driver education is well-intended training without a shred of evidence, testing or evaluation to prove students actually benefit from the training. We just hope teens do better after learning the “rules of the road”. Typically, there are no follow-up studies, no before and after evaluation of skills, no behind-the-wheel behavior change evaluation, and no crash frequency evaluation. In a word, traditional driver education is “pathetic” when it comes to the use of effective instructional design, instructional technology and impact evaluation.

teenSMART training was subjected to an extensive developmental evaluation conducted by world famous scientists from Carnegie Mellon University and the Traffic Injury Research Foundation. Teens’ knowledge, skills and driving behaviors were measured before and after taking teenSMART. The teens were assessed using multiple choice questions, computer-based driving simulation, and videotaping behind the wheel while driving. Subjects were randomly assigned to one of four groups to control for testing effects, and double-blind evaluations of video clips were used to determine pre- and post-driving behavior. In short, this was a legitimate third-party developmental evaluation of learning effectiveness

The evaluation findings indicated statistically significant improvement in collision reduction knowledge, skills and behind-the-wheel driving behavior. One finding ADEPT is particularly proud of is greatly improved visual search. We videotaped teen drivers before and after training. The camera was mounted in the back seat and positioned forward so as to see the road ahead and the rear view mirror. Unknown to the driver, we were counting eye movements to measure the visual search activity level. The pre-test subjects exhibited the typical target fixation on the road directly ahead associated with most new-to-the-road drivers. The post-test subjects showed statistically significant gains in eye movement and visual search. They were clearly seeing more of what was going on around the car after training.

teenSMART was the first, and remains the only, teen driver program to be subjected to this level of third -party evaluation of learning effectiveness.

Decades of well-conducted research has shown that traditional driver education has little or no collision reduction benefit. The famous DeKalb Georgia study conducted in the 1970’s was one of the first large scale studies to discover that there was no difference in collision frequency between teens with traditional driver education and those with no driver education at all. Collision rates were checked using DMV records at six months and at one year after completing driver education. More recently in 2004, Canadian researchers in British Columbia reported increased collision rates in teens that completed driver education when compared to those with no driver education, after controlling for driving exposure. These and other research studies underscore our assertion that traditional driver education is helpful in passing a DMV test, but not helpful in reducing collisions.

teenSMART was designed to change drivers’ knowledge, skills and behavior. teenSMART was beta tested in Elk Grove, California during the summer of 1998. One year after training, collision rates for teenSMART Beta graduates were 71% lower than the national collision rate for 16-19 year-old drivers during the same time frame. (as reported by the National Safety Council). Collision rates for teenSMART graduates were 30% lower than a highly safety conscious and matched control group that came from the same high schools that did not take the teenSMART training. California DMV accident reports were accessed one year after training to determine collision rates for the teenSMART and the matched control group subjects during the same time frame. Results indicated 30% fewer DMV-reported collisions in the teenSMART group.

The teenSMART Developmental Evaluation and Beta Test results were encouraging and unprecedented in the driver education industry. However, we needed to conduct further studies to answer the following questions (1) can the earlier collision reduction observation be replicated in future studies?, (2) Will examining large numbers of teens impact the observed collision reduction benefit? (3) will third party (independent) reporting of collisions produce different results than police reported collisions? In 2004, ADEPT examined third party insurance claims records of over 4700 teens that had completed teenSMART and passed the certification test during a three year period. ADEPT also examined third party claims records of over 250,000 control group teens that came from the same geographic areas as the teenSMART trained group during the same time frame. The age and gender proportions of the teenSMART trained and non trained groups were very similar. The two levels of the independent variable were (1)teenSMART trained and(2) non-teenSMART trained teens. The two levels of the dependent measure were (1) reported a claim as a result of a collision (2) did not report a claim. A two by two matrix was constructed and a Chi Square test of significance was performed along with descriptive statistics. The findings indicated a significant difference (p<.001) in the reported claims. teenSMART teens were observed to have 29.8% fewer reported car collisions claims fillings than the non-trained control group. These findings were consistent with earlier studies involving police-reported car collisions one year after teenSMART training.