What actually works in driver education

In July 2024 the US National Highway Traffic Safety Administration published "Driver Education and Training Promising Practices: A Systemic Literature Review" (report DOT HS 813 566), written for it by Daniel Mayhew, Ward Vanlaar and Robyn Robertson of the Traffic Injury Research Foundation. Its job was to take the standards that guide teen driver education in the United States and pressure-test them against the most recent research, filling the gaps where the evidence had moved on.

Those standards are known as the NTDETAS, the Novice Teen Driver Education and Training Administrative Standards. As the review puts it, they "reflect current and recommended practice in driver education and are based on a foundation of expert opinion, experience, and consensus backed by available scientific evidence." The word that matters in the title is promising. This is a map of what looks most likely to work, drawn carefully and without overselling.

The review identified three priority questions: the value of blended learning, the use of deliberate practice, and the application of technology in driver education. Each one maps onto a decision every driving school and every parent quietly makes.

The thread running through the standards the report reviews is structure. Skills are meant to be introduced in an order, each one secure before the next is added, so a learner is never asked to juggle a new demand while still fighting the basics. Car control comes before reading traffic; reading traffic comes before the harder work of judgement, anticipation and self-awareness. Get the order wrong and a learner spends the whole lesson overloaded, which is where mistakes and bad habits set in.

This is where deliberate practice earns its place. A structured programme knows exactly which skill today's session is building, drills it deliberately, and only moves on when it is solid. That is very different from the common pattern of simply driving around and hoping the hours add up. The wider evidence is blunt on this point: simply accumulating lessons or time does not, on its own, reliably lower a new driver's crash risk. What you practise, and how, matters more than how much.

The NHTSA review puts a sharp point on this. It cautions that instructors often focus on hitting an hourly standard rather than on purposeful practice during the lesson. Teaching to the clock is not the same as teaching to the skill, and the hours are only worth what you deliberately do with them.

The instructor's job is more than demonstrating manoeuvres. Modern driver education leans on coaching: asking before telling, so the learner does the thinking and builds judgement they can carry after the test, rather than following instructions they forget the moment they are alone in the car. The instructor sets the standard and the plan.

The accompanying driver then multiplies it. Ireland's EDT system exists precisely because supervised practice between lessons is where skills bed in. But that practice is only as good as its direction. When a parent knows what this week's lesson was working on and reinforces the same habits, the two teachers compound each other. When they contradict each other, the learner is caught in the middle. The report's emphasis on the supervising adult is a reminder that the hours at home are part of the programme, not a break from it.

Assessing the learner

Passing the test is one measure, and a narrow one. Good assessment checks the things the test barely touches: whether a learner can read a developing hazard, judge their own driving honestly, and hold it together under pressure. It should run through the course, not just at the end, so teaching can respond to what a learner actually needs rather than what the syllabus assumes. The most useful feedback is objective, something a learner cannot wave away as just an opinion.

Keeping them engaged

This is where the Gen Alpha point bites. Teenagers raised on responsive, interactive screens do not learn well from a lecture. The report's interest in blended learning and technology is partly about engagement: short interactive modules, hazard-perception clips, simulators and apps can hold attention and let a learner practise safely before the real road. Used well, technology is not a shortcut, it is a way to keep a young driver doing the work instead of tuning out.

The report even suggests matching each tool to a job: a simulator for junctions, faster roads and night driving; simulation techniques for hazard perception and other thinking skills; virtual reality for the effects of distraction and impairment; and augmented reality for basic vehicle control and the rules of the road.

Transfer is the whole game. A learner can ace a theory module, tap every hazard in a clip and drive a flawless simulator lap, and none of it counts until it shows up as safer behaviour on a wet Tuesday with friends in the back. This is exactly why the report cares about structure, deliberate practice and honest assessment: they are the parts most likely to stick, because the learner built the skill themselves rather than borrowing it for a test.

It is also why the honest framing matters. The wider research on driver training is famously humbling. Large reviews have found that formal training aimed at passing the test does not, by itself, reliably reduce crashes compared with informal practice. The promise in "promising practices" is real, but it lives in the details: the sequence, the quality of practice, the coaching, the reinforcement at home, and the hard skills of hazard perception and self-awareness that the road actually demands.

The review is honest about its own limits, too. It set out to answer three questions and deliberately left others open, naming parent involvement, what it takes to train a good instructor, and graduated licensing as gaps still waiting for solid evidence.