Posted by: Didapper PJ10 APR 2014
You may not be aware of this, but I have been blessed with more than the average number of lower limbs.
No, I am not Jake the Peg, deedle eedle eedle um, with an extra leg, deedle eedle eedle um. I admit to having the customary pair. But since a small proportion of people — through accident, warfare, disease or birth defect — have fewer than the standard two legs, then the average, or mean, must be slightly less than two. In fact, roughly one person in a hundred is missing a leg, so that the mean leg count is about 1.99.
The statistical terms mean, median and mode can all be useful when considering a range of values but they must all be used sensibly. In the case of leg counts, the significant term is the mode, which is the number that occurs most frequently.
The median is the middle value in a list, and since people can have two legs, one leg or no legs, the median must be one — which makes it as useless as a mean leg count of 1.99.
I cite all this to demonstrate that we should be cautious in using or interpreting statistics. As Mark Twain, Benjamin Disraeli and various other figures in history are alleged to have said: “There are three kinds of lies: lies, damned lies and statistics.”
As a retired journalist who has edited many research papers (mainly written by pharmacists), I am saddened by the sloppy way in which statistics are often used.
Researchers often manage to draw concrete conclusions from studies based on inadequate sample sizes. And they may also contrive to calculate percentages to two or more decimal places when the sample size cannot justify such precision. For example, a response rate of two out of seven is not 28.57 per cent. It’s just two out of seven.
Of course, since readers of the PJ are all in the top 100 per cent for intelligence, I trust they will spot such misleading statistics when they appear in print.