Correlation is a statistical term that describes the relationship between two different, measurable factors. The relationship may be positive (same direction—one goes up, the other goes up, like temperature and the number of people on the beach), or negative (like temperature and the number of people wearing coats).
The relationship will have a mathematical formula associated with it, but it may not be a linear link. Changing one variable may do wacky things to the other. Some of the relationships can be rather complicated.
The strength of the relationship is described by the correlation coefficient. In practice, most people never have to calculate this, but you may hear a Black Belt use it at some point.
Correlation is often confused with causation. In the examples above, there is causation—one event drives the other event. In many cases, though, two factors are correlated only because they are both linked to a third variable.
An example of this might be the number of worms crawling on the sidewalk and the number of people with umbrellas. These factors might have a strong correlation, but one does not cause the other. Worms do not crawl out of their holes when they see an umbrella, and people don’t pop open umbrellas whenever they see a worm. Rain is obviously the cause of both.
The big warning here is not to immediately take action when you notice a correlation. Look for the cause first, or you will waste resources working on something that can’t do what you want.
Causation implies correlation. Correlation does not necessarily mean causation.