What Is a Correlation Coefficient?
The short answer is that a correlation coefficient squeezes the relationship between two sets of numbers into one value between -1 and 1. It is usually called r, and it only measures straight-line relationships, so a curve can fool it.
What Does a Correlation Coefficient Tell You?
It tells you two things at once: direction and strength. The sign is the direction. A positive r means both variables tend to go up together. A negative r means one tends to go up while the other goes down. The size is the strength, and the closer to 1 or -1, the tighter the pattern.
| r value | What it roughly means |
|---|---|
| 1 | Perfect positive line |
| 0.7 | Strong positive |
| 0.3 | Weak positive |
| 0 | No straight-line pattern |
| -0.7 | Strong negative |
| -1 | Perfect negative line |
What Is the Correlation Coefficient Formula?
The Pearson formula looks scary and is not. In plain words: take each pair of values, see how far each one sits from its own average, multiply those two gaps, and add up those products. Then divide by how spread out both variables are. That is it. The result is r.
Written out, r equals the sum of each pair's deviation products, divided by the square root of the summed squared deviations of x times the summed squared deviations of y.
Can You Show a Worked Example?
Take five students with hours studied (x) and test score (y): (1, 50), (2, 60), (3, 65), (4, 80), (5, 90). Both climb together, so r should be positive and high. Run the formula and you get about 0.98, which is a very strong positive correlation. More study hours went with higher scores in this tiny sample. That does not prove studying caused the scores, only that the two moved together.
Does Correlation Mean Causation?
No, and this is the classic trap. Ice cream sales and drowning rates rise together, but ice cream does not drown anyone. Hot weather drives both. A lurking third variable can create a correlation out of thin air, so treat r as a clue, not a verdict.
One fun detail: the letter r for this statistic traces back to Francis Galton's late 1800s work on heredity, and Karl Pearson turned it into the formula we use today.
Do the math: open the correlation calculator