I recently took a test, labelled “Driver Risk Index“, developed by University of Cranfield, as part of my summer sabbatical job as a tourist coach driver in Stockholm. The purpose of the test is to establish, by answering a hundred or so multiple choice questions, whether a driver is a safe driver or not.
This test is an excellent illustration of the increasingly popular “academization” of any and all professions today (cmp the “certification” madness of the software industry, where you after a day or two of listening to a “guru” – and paying a hefty fee – can become a Certified Master of <insert your favorite discipline here> !) including those professions that are fundamentally and primarily skill and practice-based, such a driving a coach, doing brain surgery, or for that matter, playing soccer, violin or tennis, or being a programmer.
My point is that to become a “Master” of anything that involves skill, you need to do your 10000h of dedicated homework (practice), and to assess anything skill based, you need to observe the test subject in realtime!
Take any fundamentally skill based endeavor involving risk, such as car or MC racing, downhill skiing, or sail racing: I doubt there’s any coach who would base their views on the risk level of each practitioner on a survey: instead, they will use observation and analysis of the data obtained by observation to make their assessment.
The belief behind this academization seems to be that everything, even endeavors which are fundamentally skill based, can be assessed without actual observation of the practice; if we just can come up with a set of smart questions to ask, we will be able to know whether the professional is good or not….!
Let me make my point clear: I doubt the validity of this approach… deeply!
Taking the Driver Risk Index test, triggered my curiosity, in particular about the foundation and validity for making a purely theoretical – online questionary based – assessment of something (driving) that is deeply practical and skill and experience based, with inherent cognitive complexity, which for mastery demands the 10000 hours of practice that Malcolm Gladwell and others have written extensively about, to safely, comfortably and timely drive a vechile from point A to point B, in very varying traffic situations and conditions.
Doing the test, it was not very hard for me (an understatement!) to figure out which of the multiselection answers would gain me “good” ratings as a “safe” driver, and thus I decided to play the test a bit by providing “provocative” answers which I believed would put my assessment into the “red”, i.e. “high risk” sector… And quite so it did:
For instance, questions like:
– “have you ever driven faster than the speed limit ?”
– “do yo enjoy driving fast ?”,
– “have you ever crossed the solid line ?”
– “have you ever run a red light ?”
when given positive answers immediately places the respondent in the “red” or “high risk” category.
To me, using answers to these types of questions to infer that the driver exhibits risky behavior appears very simplistic, naive and misleading, in fact, I would argue that this type of “science” is totally ignorant about what driving – particularly safe driving – implies and demands, and what type of behavior in traffic generates the lowest level of risk.
For sure, it goes without saying that the less speed you carry, the less severe impact there will be if you have an accident, but I’d argue that low speed in general is not always the answer to increased safety – there are situations in traffic where a higher speed, even a speed above the speed limit, results in less risk for an accident than the legal speed. In other words, there are situations in traffic where low speed can increase risk, and higher speed can decrease it. Similarly, there are situations in traffic where crossing a solid line, or running a red light – particularly if you drive large and heavy vechiles – can actually lower the risk for an accident.
But in order to understand why, you need deep experience of traffic, experience that you can only obtain by having done your 10000 hours of dedicated practice. Purely academic analysis, like that of “Driver Risk Index”, will simpy not do – on the contrary, such a theoretical approach can very well provide very false results.
To understand why risk in some situations can be decreased by faster driving, or by crossing the solid line, it is helpful to think about traffic in terms of fluid dynamics: what you want to strive for in traffic is the equivalent of laminar flow, i.e. a dynamic situation where each of the fluid molecules – say water molecules – flow in a steady streamline. What you want to avoid in traffic is turbulence, i.e. the chaotic behavior the results when the molecules are exhibiting different velocities, e.g. due to disturbances.
For a concrete example of where (temporarily) increasing speed, even over and above the limit, will actually reduce risk is when merging into a crowded lane, say from a freeway entrance: the mistake many inexperienced drives commit is to drive far to close to the car in front on the entrance, and then keeping too low a speed for seamless merging. It’s much safer – less risky – and easier to leave a significant gap, and then accelerate – even if that acceleration results in temporary speeding – into a suitable gap in the lane you want to merge to.
But, according to the “traffic experts” at University of Cranfield, this type of behavior places you in the “high risk” category.
The same faulty thinking from the academics goes for the question
“do you enjoy driving fast?”
Any driver, professional or not, who really wants to improve their driving skills, *must* learn the limits of the vechicle and their driving skills in order to be safe. For sure, you should not practice this in your everyday driving, you need to find a time and place to do so, ideally on a closed track, but unless you “experiment” with your driving, you will never improve.
Since driving is skill based, it obeys the same “laws” as any skill based endeavor, e.g. alpine skiing, playing tennis, soccer or violin, and I doubt that the “Masters” of those activities have never pushed their practice close to or even beyond their limits. The important thing is to do this pushing in a controlled environment. But again, this seems to be far from how the academic “experts” think about traffic and driving….
I’m therefore questioning the accuracy, precision and validity of the approach taken by Cranfield University on their very academic but from reality disconnected approach for assessing driving patterns.