### Make No Assumptions

I once took a course on statistics taught by a man from Denmark.

In the lecture on ‘correlation is not causation’ he asked the question, “Do storks bring babies?

If the answer is yes, then presumably a graph of the stork population vs. the human birth rate should be positively correlated (by a straight line, if each stork brings a certain number of babies per month).

He showed a graph similar to the one below (which I have stolen from Daled Amos’s blog), which is based on real data collected in Germany.

Of course, the joke is that the existence of a straight-line relationship does not necessarily prove that storks do bring babies. There could be many other reasons why the observed stork population would go up as the human population goes up (for example, if storks like to eat food scraps people throw away, then the supply of stork food would go up linearly as the human population did).

As he was talking, I was a bit ruffled by the use of human population as one axis of the graph. If we want to know if storks bring babies, we should look at the birth rate, not the total population.

As I was mulling over the idea of raising this point, a Korean student sitting next to me turned to me and asked, “What is a stork?”

I guess the only point of this post is: If you’re giving a lecture on any topic, we have reached the point in globalization now where it is unwise to assume that your audience knows the things you take for granted. The person who will be best understood is the one who makes no assumptions about what their audience knows and is still able to get their point across succinctly.