Correlation Coefficient — Pearson's r & Interpretation

The correlation coefficient usually means Pearson's correlation coefficient, written $r$. It measures the direction and strength of a linear relationship between two numerical variables, and you reach for it when data come in pairs, both variables are numerical, and a straight-line trend is the pattern you want to summarize.

If $r$ is positive, the variables tend to increase together. If $r$ is negative, one tends to decrease as the other increases. If $r$ is near $0$, Pearson's $r$ is saying there is little linear pattern, not necessarily no relationship at all.

When Pearson's r Is The Right Tool

Use Pearson's $r$ for paired numerical data when linear association is the question you want to summarize. It is a standardized measure of how two variables vary together. For a sample of paired data, the formula is

The numerator is positive when the variables tend to move in the same direction and negative when they move in opposite directions. The denominator rescales that joint movement using the spread of each variable. When defined, Pearson's $r$ satisfies

If one variable has no variation at all, the denominator becomes $0$, so $r$ is undefined.

The Steps To Compute And Read r

1. Check the setting

Confirm you have paired numerical data and that linear association is the question.

2. Center the data

Compute $\bar{x}$ and $\bar{y}$, then find each deviation from its mean.

3. Compare joint movement

Add the products $(x_i-\bar{x})(y_i-\bar{y})$ to see whether the variables rise and fall together.

4. Scale the result

Divide by the product of the two deviation-based spreads so the value stays between $-1$ and $1$, when defined.

5. Interpret carefully

Read the sign as direction: $r > 0$ is positive linear association, $r < 0$ is negative, $r = 0$ is no linear association. Then read the magnitude $|r|$: closer to $1$ means the points stay closer to a straight line, closer to $0$ means a weaker linear pattern. Be careful with labels like "weak," "moderate," or "strong," since those cutoffs depend on context. In one field $r = 0.3$ may matter; in another it may be too small to act on.

A Full Worked Example: Calculating $r = 0.9$

Suppose the paired data are

First compute the means:

List the deviations from the means:

• For $x$: $-2, -1, 0, 1, 2$
• For $y$: $-2, -1, 1, 0, 2$

Multiply the paired deviations and add:

Now the two sums of squares:

So

This is a strong positive linear association: as $x$ increases, $y$ usually increases too, and the points would sit fairly close to an upward-sloping line.

Where Each Step Goes Wrong, And A Self-Check

The interpretation step holds the most traps:

• Treating correlation as causation. A high $r$ does not prove one variable causes the other. A third factor may drive both.
• Forgetting that Pearson's $r$ is linear. A curved relationship can produce a small $r$ even when the variables are clearly related.
• Ignoring outliers. One unusual point can change $r$ a lot and tell a misleading story.
• Using $r$ when the setup does not fit. If one variable is categorical, or the pattern is clearly curved, $r$ may not answer your question.
• Overreading a near-zero value. It means "little linear association," not "no relationship of any kind."

The safest habit is to read $r$ alongside a scatter plot: the number summarizes the picture, it does not replace it. As a self-check, plot a small data set you understand and estimate whether the trend looks positive, negative, or unclear before computing $r$. To go further, fit a simple linear regression line on the same data and see how correlation and prediction relate, but are not identical.

Why It Matters

Pearson's $r$ is a quick summary of paired numerical data in statistics, science, economics, social research, and machine learning. It is most useful when you want to know whether a straight-line pattern is present before moving to a model such as linear regression. A scatter plot should come first; the coefficient is a summary, not a replacement for looking at the data.