How many drivers are above average?
Discussion
Someone on R4 this morning was saying that 'Most drivers think they are above average but this is impossible'.
Of course for anything other than median average it's perfectly possible. Taking number of accidents as a very rough measure of driving ability, insurance companies must have some tables showing the distribution. If it's a bell shaped curve than only 50% of drivers are above average, but if 20% of drivers cause 80% of the accidents, then 80% of drivers could be above average.
Does anyone know what the distribution actually is?
Of course for anything other than median average it's perfectly possible. Taking number of accidents as a very rough measure of driving ability, insurance companies must have some tables showing the distribution. If it's a bell shaped curve than only 50% of drivers are above average, but if 20% of drivers cause 80% of the accidents, then 80% of drivers could be above average.
Does anyone know what the distribution actually is?
Ozzie Osmond said:
Fidgits said:
Eh?
Average by its very definition is the median of a group.
No it's not. Need some maths revision before resitting that GCSE!Average by its very definition is the median of a group.

Gcse's, that's what replaced the o levels right


My definition may be flawed (I'm not an accountant, sorry), but my point stands.
If you take all the drivers in the uk, and take an average of skill level, now if that skill level is terrible [on average] then the amount of skill needed to be "above average" isn't that high, but if the average skill is actually quite good, then to be above average, you would need to be much much more skilled.
Now, if we were to take the average of the entire worlds driving skill. Then it is fair to say, the majority of uk drivers could be considered above average..
edo said:
no would I. Not crashing doesnt make you an above average driver (or even average).
I did say a very rough measure, and ideally you would allow for mileage and road conditions. But the object of driving is to not crash, so I can't really think of a better measure.Whether you are looking at mean or modal, the question is the same.
Above average what?
If it's a driver's ability to not be involved in an accident then it will be a normal distribution, like most things. Therefore average (which when undefined is usually taken to be the mean) would be right in the centre of the distribution with 50% either side. So most drivers cannot be above average, and in fact the large majority are less than 1 standard deviation away so could only be described as average.
If it's a driver's ability to not be involved in an accident then it will be a normal distribution, like most things. Therefore average (which when undefined is usually taken to be the mean) would be right in the centre of the distribution with 50% either side. So most drivers cannot be above average, and in fact the large majority are less than 1 standard deviation away so could only be described as average.
Fidgits said:

Gcse's, that's what replaced the o levels right


My definition may be flawed (I'm not an accountant, sorry), but my point stands.
If you take all the drivers in the uk, and take an average of skill level, now if that skill level is terrible [on average] then the amount of skill needed to be "above average" isn't that high, but if the average skill is actually quite good, then to be above average, you would need to be much much more skilled.
Take these numbers: 1, 2, 4, 5, 7, 7, 8, 9.
The mean is 5.375 (sum them, divide by 8), the mode is 7 (the most common), the median is 6 (the centre of the sequence). Three very different results from the three simplest "averages", but which one is best? Complex question.
As Misery alludes to above, for a normal distribution of drivers there will be an equal split who are above and below average, gradually tailing off with skill as you go further away from the average (either very good or very bad drivers).
Statistics aren't helping - as 88.2% of people are below average at them 
Why don't we define a level of competence in driving, let's call it the PH standard, which most residents here think is acceptable.
Now I suspect this is quite a bit higher than the majority of "I drive to get from A to B and don't give a stuff about how" types.
I'd be surprised if 20% of drivers met the PH standard.

Why don't we define a level of competence in driving, let's call it the PH standard, which most residents here think is acceptable.
Now I suspect this is quite a bit higher than the majority of "I drive to get from A to B and don't give a stuff about how" types.
I'd be surprised if 20% of drivers met the PH standard.
Fidgits said:
Not at all.
Assuming most drivers will have 2 eyes, I would say the average number of eyes a driver has is 2... Therefore, unless you have 3 eyes, you won't have above average number of eyes...
My friend drives, he has one eye. I don't know anyone with three eyes therefore the average number of eyes per driver is less than two. I have two eyes, I'm above average. Assuming most drivers will have 2 eyes, I would say the average number of eyes a driver has is 2... Therefore, unless you have 3 eyes, you won't have above average number of eyes...

Fidgits said:
Dr Jekyll said:
Would you accept that most drivers have an above average number of eyes?
Not at all.Assuming most drivers will have 2 eyes, I would say the average number of eyes a driver has is 2... Therefore, unless you have 3 eyes, you won't have above average number of eyes...
Dr Jekyll said:
Someone on R4 this morning was saying that 'Most drivers think they are above average but this is impossible'.
Dunning kruger EffectNot only possible but inevitable. Part of incompetance is the lack of ability to correctly judge competance.
If you take away the people who think they are above average in driving skill, and the people who are above average in driving skill you'll have almost no one left.
Gassing Station | General Gassing | Top of Page | What's New | My Stuff


