Data Science
Correlation Vs Causation
May 11, 2017 Fred Schwaner

Machine Learning Definitions with Fred

The relationship between correlation and causation is a common trap that a data analysts can fall into. When we see that event A is correlated to event B we are trained to believe that either event A caused event B or event B caused event A, but there are other possibilities that are less frequently considered.

The first possibility is that the two events are happening at the same time just purely out of coincidence. The second possibility is that both event A and event B are caused by a third event, C. One example where this phenomenon may occur is when you notice an increase in ice cream sales leads to an increase in people drowning. You could conclude that eating more ice cream causes you to not swim as well or you could consider the fact that both are increasing because it’s summertime and the weather is warmer.

Fred Schwaner
Machine Learning Engineer