Need evidence to prove that training pays off
By: Peter F. Lourens
Date: 1993-03-09
Dr. Peter Lourens is section coordinator at the Traffic Research Centre, University of Groningen, the Netherlands. His works include: "Drivers and children: a matter of education," "Theoretical perspectives on error analysis and traffic behaviour," and "Report on the training and certification of driving instructors in the European Community".
In Strassbourg, France, in May 1991, I presented a paper to the International Association of Driver Educators (IVV) that offered two conclusions:
- In far too many cases, basic driver training isn't even sufficient to enable learners to pass their road test.
- In far too many cases, testing standards are not adequate to ensure that drivers are reasonably safe on our roads.
These conclusions were based on the facts that, in many countries, only 50% of learner drivers pass the test on the first try; and that new drivers are over-represented in crash statistics during their first two years of driving. Without blaming any group, it seems modern society is doing a poor job and is unwilling to do anything about it.
As a researcher who is questioning the quality of driver education, I have been told repeatedly by superiors and colleagues that it's very difficult to put forward proposals in this field. Both the practitioners and the responsible authorities prefer not to invest in research that might show the weak points of the prevailing system to the world. You can understand that I'm unhappy about this situation and in this paper I will give it another try: Can driver education, in one form or another, be effective in terms of improving traffic safety?
There is very little research to answer this question. A recognized expert, Professor Ivan Brown of Britain's Cambridge University, has reached strong conclusions on this. Professor Brown and his colleagues (1987) found no experimental studies that provide sound evidence on the relationship between driver training and accident involvement. They also found that most earlier research on the topic is suspect on methodological grounds.
In later research (1989, 1992) Brown found that there's no reliable evidence on the benefits of training for road safety. He also found "there is little hard evidence on the desirable content or process of driving instruction." I appeal to those who are really interested in the question of the effectiveness of driver education to unite and produce the sound evidence that's lacking. To get this evidence we need:
- Insight into the psychology of driver training/education
- A research plan, agreed on by experts
- Time, places and money to carry out the plan.
The point of departure for our program is that psychology teaches us the necessity to train and educate (young) people in safe driving performance. In our society, any human being starting to engage in performing a new and complex task gets training and education. Driving can't be the exception. As a point of departure for our research program we need to know three things:
- How do people learn?
- What is safe driving?
- How can we best teach people?
A basic law of psychology is, "People learn by doing." At the most primitive level this involves a process of conditioning reflexes-like the dog that begins to slaver at the sound of the food tin, or the driver who turns his head in the direction of the horn. At a somewhat higher level, people learn by association (a red light triggers the impulse to stop, for example).
Typically human is learning by understanding. People learn best when they understand why something is worth doing in a certain situation.
In learning a complex psychomotor task such as driving a car, learning proceeds from higher levels to lower levels in what is commonly accepted to take place in three phases:
- The cognitive, or knowledge-based phase. People need to consciously process the knowledge which tells them what to do.
- The associative, or rule-based phase. After a while, knowledge is fully translated into rules, which are triggered quickly and smoothly, without thinking.
- Finally, the autonomous or skill-based phase. Actions follow situational stimuli automatically, and then we can say that people are skilled performers.
What is safe driving performance?
A thorough task analysis can determine which actions, in which circumstances, and how performed, will more or less guarantee safe outcomes in driving. In doing such analysis, researchers have come to recognize three different levels of the driving task which form a hierarchy. The three levels relate nicely to the three phases of learning behaviour described above.
Types of behaviour
Knowledge based
Rule-based
Skill-based
Levels of driving
Strategic
Tactical
Operational
The strategic level involves knowledge of things like available modes of transport, choice of routes, average time needed to get to destination, etc. This knowledge is used consciously in making driving decisions.
The tactical level involves the type of roadway, speed, and other relevant rules. Tactical tasks are performed by applying and routinely following rules.
The operational level involves appropriate use of the controls, and this is accomplished with skills we use automatically, without conscious thought.
We must keep in mind, however, that the relationships between task levels and behaviour types are not constant and, in particular, not the same for all drivers. A strategic route choice, for example, may be made totally automatically, and is thus skill-based. An operational act of braking may be carried out on a knowledge basis by a beginner driver.
To summarize:
In a driving task analysis, it is possible to describe exactly what safe performance is when the level of the task and the type of driving behaviour are exactly specified. Being a safe driver involves a process of inspecting the driving situation, detecting hazards, and taking appropriate action. But this inspection, detection, and action can only take place smoothly and effectively as long as there's no interference.
Interference comes from within the driver, so a driver must constantly be in a process of self-assessment and know how to avoid personal errors. The good, safe driver knows the errors and keeps them out. This means:
- No bad habits (such as drinking and driving).
- No automatization (absent-minded or subconscious driving) at the tactical level.
- No disturbances or distractions (keep your eyes on the road).
- No haste (take it easy).
The issue of what's the best way to teach driving is much discussed but not yet resolved. To me, it's absolutely clear that we need to send drivers to school. To driving schools, that is, which can give them a safety-based training course. But again, I come back to Professor Brown, who has stated, "We do not yet know how to train safe driving behaviour that will persist through the early years of driving experience." In other words, the traditional systems for driver training do not contribute enough to traffic safety.
As an alternative, Brown proposes "a graduated system of compulsory training and testing within a series of reducing legislative constraints on exposure to risk."
Conclusion
In conclusion, I would like to present a broad outline of a project that can produce the evidence we are waiting for. The project plan is to form an international research group, supervised by IVV, to do a comparative study of driver education systems. The goal is to prove that driver education is cost effective if it reaches a specified level of quality.
The plan:
This is built around four main tasks:
- Define what good driver education is (which components, which content, which teaching processes).
- Compose a number of different educational systems and apply these to groups of candidate drivers.
- Define and select evaluation criteria and measurement instruments.
- Do evaluation studies on all criteria, with all instruments.
The final things we need are the money to carry out this plan, and the people with relevant expertise. Who can help?
Comments to this article have been disabled.
