In ur data, bein unreprezentativ.
After two people recommended Ben Goldacre‘s Bad Science to me within the space of a few hours, I decided to spend some of my leftover book tokens on it. Having now finished it, I can wholeheartedly recommend it in turn: it’s very good, highly informative, and Goldacre does a nice line in withering put-downs.
On page 254, there is a footnote. It reads:
I’d be genuinely intrigued to know how long it takes to find someone who can tell you the difference between ‘median’, ‘mean’ and ‘mode’, from where you are sitting right now.
As it happens, I was sitting on the sofa in J’s front room, watching the rerun of Sky’s Going Postal adaptation with him and Rhiannon. A quick survey established that all three of us knew the difference, providing an answer to Goldacre’s question of ‘about four seconds’.
Naturally, based on this data I can only conclude that everyone in the world knows the difference between the three kinds of average. Because 100% of our sample did! (Extrapolating from the same sample, you could also conclude that one-third of the population is Quaker and that two-thirds will study to Masters level.)
Ridiculous, isn’t it? And yet things on a similar scale of silliness get into the papers every day.