Even on the fringes of road safety, most people now understand that we should use the word ‘crashes’, not accidents, but there are many other important topics where there is still confusion and inaccuracy. This includes the proportion of crashes that are — quote — ’caused’ by driver error, or less accurately, human error.
To avoid this erroneous, unfortunate and misleading statement, it is necessary to understand that crashes are random events which, in layman’s terms, often involve simply being in the wrong place at the wrong time.
Beyond that, however, crashes are almost always caused by more than one factor.
Let’s consider a hypothetical incident: Imagine a crossroads where traffic on the main road does not have to stop and the two side roads each have a stop sign. Imagine, too, a car coming along the main road at night, in excess of the speed limit and with one set of front lights not working because of prior damage. So in the dark it now potentially looks like a motorcycle because it only has one front light and for the same reason its approach speed is harder to estimate.
At the same time (and this is the ‘wrong place/wrong time’ factor at play), a drunk driver is approaching the main road from one of the side roads but fails to see the stop sign or the single headlight approaching, and drives into the path of the first car. The two collide and both drivers are killed.
Whose fault was it?
How is driver error quantified in this case?
The minimum number of actual factors in the above incident were:
- Speeding
- Defective vehicle (the lights)
- Drunk-driving
- Failing to stop for the stop sign
Did it involve ‘driver error’ per se? Yes, but there are different ways of quantifying such human errors so if you will forgive the pun, I will go no further down that road in this article.
However, compare the above list of four specific wrong-doings with another crash at the same theoretical intersection — this time in daylight — when the driver of a car that is emerging after stopping at one of the stop signs simply does not look properly and starts to turn left, across the path of a motorcycle coming from his left, right on the 55mph speed limit. The motorcycle hits the car and the rider is killed. (This, incidentally, is a heartbreakingly common type of crash and cause of death.)
In the latter incident, the only explanation is that despite a good view to his left, the car driver either failed to see the approaching motorcycle or completely misjudged its speed and/or the distance. In either case, it is blatant driver error, best described more specifically as inadequate observations.
According to British research, the average is 2.5 causal factors per crash.
According to NHTSA: “The critical reason is the immediate reason for the critical event and is often the last failure in the causal chain (i.e., closest in time to the critical pre-crash event.) The critical reason can be attributed to the driver, vehicle, roadway, or atmospheric condition.”
In other words, even with that brief description, NHTSA has identified four critical reasons, only one of which is the driver’s actions or lack of actions.
In the same NHTSA document linked above, these are examples of the ‘critical reason’ figures quoted:
- Atmospheric: 8.4%
- At least one vehicle deficiency: 6.8%
- Roadway related feature: 16.3%
From these figures alone, it is easy to see that the claim that “over 90 percent of crashes are caused by driver error” simply cannot be correct. The >90% figure is the number of crashes that include driver error as one of the factors, which is a very different thing.
A comprehensive study of road safety (Treat et al., 1977) found that human error was the sole cause in 57% of all accidents and was a contributing factor in over 90%. Figures very close to these have been used by NHTSA in recent years.
A similar misquoting was prevalent a few years ago when people were quoting the total number of alcohol-related crashes (which includes pedestrians and bicyclists) as though they were all drunk drivers. It was patently inaccurate and untrue.
The point is that if we — as road safety advocates — are going to do the best possible job and get the best possible improvements to the system, then it is absolutely incumbent upon us to get our facts and figures totally correct. Any misrepresentation, whether accidental, or — heaven forbid — deliberate, will undermine our case dreadfully.
One of our self-selected tasks at Road Safety USA is to help ensure that victims, relatives and other road safety advocates are given very precise and accurate information with which to work.
We hope that you will find some of our work in this area to be useful.