The Quincunx and Correlation
A man named Galton developed the apparatus in 1973, in which one may drop balls. The amazing thing is that every time you drop the balls, they will always land like that, forming a hill. In math terms, you might say that they distribute normally, or in hillocks (as Galton describes). Statistics is based on the idea that a large enough sample or representation of any population (not necessarily of people) will distribute in the same way, with the mean or average being the crest or maximum point on the graph. When a population is normally distributed, we can make inferences not only that two things are related, such as, for example, finger length and body height, but they are co-related with a specific index of how they relate based on how “average” they are, and in the graph, how far away they are from the crest. Adjusting for how outside of normal a certain measurement might be, we can make a much better “index” of relation or correlation.
What might this jargon have to do anything? Well, all of modern science and experimentation uses the idea of one thing depending on others to happen. It could be physics, where energy is dependent on mass and the speed of light, or human biology, where the age at death can be predicted by many factors, one such might be smoking, to the social sciences, where a country’s proneness to war could be explained by the status of women in that society. The thread is that everyone finds a ratio that describes behavior. Then we use that ratio to explain the past and make predictions about the future. Cool.