Lucky I don’t have to take the Driver Qualification Test

In a couple of weeks Andrew will make his first and I hope only attempt at the NSW Driver Qualification Test, the computer exam that separates provisional drivers from fully licenced drivers. (Here’s the NSW car licence pathway.) And may I say that it’s a good thing that I got a full licence back when there was only one test (I did the on-road driving test in 1998, held what is now the P1 licence for twelve months and then just got handed the full licence when it expired) because although it did take me four attempts to pass the on-road test that’s nothing on what it would take to get me through the DQT.

Why? Well, the problem would be that the DQT asks a lot of questions based on the test handbook, and I wouldn’t be able to read the test handbook without smashing a wall down. With my teeth.

Let’s have a look. First we have some dodgy handwaving in which the authors of the handbook are astonished that 90% of their data fits into only five categories:

About 90 per cent of all crashes in NSW involving full licence drivers in their first year fall within only five crash types…

What’s that only doing there, huh? How many crash types are there? Six? Three hundred and twelve? For reference the five crash common types are: rear-ending someone, colliding with a vehicle travelling perpendicular to you (at an intersection), colliding with a vehicle coming from the opposite direction, running off a straight section of road and running off a bend. That does seem to cover a lot of possibilities doesn’t it? Don’t divide your population into five obvious segments plus a catch-all and then act all surprised that most of the population ends up in the five categories.

But the much more annoying problem is this. They note that 25% of provisional driver accidents are rear-enders, 34% of first year full licence accidents are rear-enders and 40% of experienced full licence accidents (‘experienced’ is drivers in the fifth and all subsequent years) are rear-enders. From these numbers they draw this conclusion:

Researchers think that these differences are due to novice drivers getting better at staying on the road but also getting into the habit of driving too close behind other vehicles in traffic. This bad habit seems to continue for full licence holders. As you can see from [a colourful pie chart], full licence drivers with more than five years’ experience have even more rear end crashes. However, they are much less likely to run off the road and hit an object.

Uh-huh. And what an interesting conclusion that would be, if they had first established that they were all having the same number of accidents. If they’ve done this, they do not indicate it, it’s all percentages of crashes within each population. (‘The same number of accidents’ is a little hard to define, but I believe it’s usually number of accidents per driver, number of accidents per driven kilometre, or number of accidents per passenger kilometre. For various reasons, mostly because of a subsequent section about how women have less accidents partly because they drive about half the kilometres men do — half! — I think this booklet is measuring on a per driver basis.) If they’re not, the higher proportion of rear-end accidents among experienced drivers doesn’t suggest that they’re risk-taking: only a higher absolute number of accidents does that. It could be, for example, that rear-end accidents are harder for even experienced drivers to avoid, and thus rise as a proportion of their total accidents just because they don’t drop as much.

In fact, we certainly hope that the same-number-of-accidents premise is false here, because the whole point of the insanely complicated and steadily getting worse NSW licencing system (in which learner drivers now must log one hundred and twenty supervised driving hours before taking the on road test, ie, about twenty minutes every single day for a year) is to reduce the number of accidents among novice solo drivers by making them less novice (and also older and less prone to risk-taking, although NSW hasn’t made this as obvious as in, say, Queensland where novice drivers over 24 get to skip a significant part of the pathway).

So far so good. That brings us to page 12 of the 94 page Driver qualification handbook. I think it’s fairly clear that I’d struggle to get through it. Especially since I can’t argue with a computer examiner.