Correlation vs. Causation

In psychology, a correlation is a statistical measurement of the relationship between two variables. We describe this relationship using a correlation coefficient, represented by the letter $r$. This value ranges from $-1.00$ to $+1.00$. A positive correlation means both variables move in the same direction (as $x$ increases, $y$ increases). A negative correlation means they move in opposite directions. However, even a 'perfect' correlation of $r = 1.0$ does not prove that $x$ produced a change in $y$. Correlation simply tells us that two things vary together, not why they do so.

The most common reason people mistake correlation for causation is the Third Variable Problem. This occurs when an unmeasured outside factor—a confounding variable—is actually responsible for the changes in both things we are studying. In our ice cream and shark example, the third variable is Temperature. High heat causes more people to buy ice cream and more people to go swimming in the ocean. The ice cream and the sharks have no direct effect on each other; they are both reacting to the sun.

Another hurdle is the Directionality Problem. Even if a causal link exists, we might not know which way it flows. Does $A$ cause $B$, or does $B$ cause $A$? For example, a study might show that people who exercise more are less depressed. Does exercise cause happiness, or are happy people simply more motivated to exercise? Media headlines often ignore this nuance, using 'power verbs' like boosts, triggers, or destroys to make a simple correlation sound like a definitive cause-and-effect discovery.