Correlation

A correlation is a link of some kind between two variables.

Most often the correlation is measured as part of a linear relationship between the two variables. Mathematically, correlation is the measure of how close the points on a scatterplot are to the line of best fit calculated using linear regression. The closer the points are to the line, the higher the correlation.

Correlations can be either positive (where, in general, both variables increase or decrease at the same time) or negative (where, in general, as one variable increases, the other decreases). A correlation of any kind does not, however, imply that the changes in one variable cause changes in the other to happen. We just know the variables that are correlated are linked somehow. We just don’t know exactly...how.

For example, ice cream sales and drowning deaths are highly, positively correlated. It’s pretty clear ice cream vendors aren’t murdering their clients once they get in the water, and in fact, both of these variables are very summer-dependent. In other words, in the summer more people buy ice cream, and more people are near water, cooling off and, sadly, drowning. Still, there is a definite correlation between increased ice cream sales and increased deaths by drowning.