Data is not information; it just a carrier of information
The Sunday Nation
Nairobi,
20 December 2015
A recent report in
the Daily Nation proclaimed that men are three times more likely to die
in a road accident than women. The supporting evidence for this
proclamation was that, in 2014, the National Transport and Safety
Authority (NTSA) recorded 1,091 men and 244 women who died on the road.
The question in my mind is whether this is enough evidence to support
the conclusion.
We can only conclude that men are more likely to die in road accidents if we were certain that the total number of men and of women “on the road” during the year were equal. But we know they were not for the report points out that there “are more men…working in the transport sector, which exposes them more to vehicles and increases their risk of being involved in a road crash”
So, the only conclusion we can make from the data presented is that more men than women died on the road in 2014. That does not mean that I am more likely to die in an accident than my wife. The reason is that we spend almost equal time on the road so our probabilities of getting involved in a fatal accident are the same.
This report reminded me of a fable that I read many years, nay, decades ago. It was written by mathematician George Polya in his book Induction and Analogy in Mathematics. The fable is called “The Logician, the Mathematician, the Physicist, and the Engineer”:
“ ‘Look at this mathematician,’ said the logician. ‘He observes that the 99 numbers are less than 100 and infers, hence, by what he calls induction, that all numbers are less than a hundred.’
“ ‘A physicist believes’, said the mathematician, ‘that 60 is divisible by 1, 2, 3, 4, 5, and 6. He examines a few more cases such as 10, 20, and 30, taken at random (he says). Since 60 is also divisible by these, he considers the experimental evidence sufficient [proof that 60 is divisible by all numbers]
“ ‘Yes, but look at the engineers’ said the physicist. ‘An engineer suspected that all odd numbers are prime numbers. At any rate, 1 can be considered a prime number, he argued. Then there comes 3, 5, and 7 all indubitably primes.
‘Then there comes 9; an awkward case; it does not seem to be a prime number. Yet 11 and 13 are certainly primes. “Coming back to 9” he said, “I conclude that 9 must be an experimental error”’
The moral of this fable is that we must always be careful about the conclusions we draw from any given data. After all, the numbers are not the information; they are only carriers of the information. The fact that 1,091 men and 244 women died on our roads in 2014 does not tell us that men are more likely to die in such accidents.
 |
 |