Course contentsShow
← Data Science
Lesson 534 of 2,14514. Simple Linear RegressionFree lesson

R-Squared vs Correlation Squared

See that in simple linear regression, R² equals the squared Pearson correlation coefficient (r²).

R-Squared vs Correlation Squared

What you'll learn: You'll discover that in simple linear regression (one predictor), R² is exactly equal to the squared Pearson correlation coefficient (r²).


The Surprising Identity

Here's a beautiful mathematical fact: when you're running a simple linear regression with just one predictor variable, the R-squared value you calculate is exactly the same as taking the Pearson correlation coefficient between X and Y and squaring it.

In other words: R² = r² (in simple linear regression only!)

Why Does This Matter?

This relationship reveals that both metrics are measuring the same fundamental thing from different angles:

  • Pearson's r measures the strength and direction of the linear relationship between two variables
  • R² measures the proportion of variance in Y explained by your regression model

When you square Pearson's r, you remove the sign (direction) and focus purely on strength—which is exactly what R² does! Both tell you how tightly the points cluster around the regression line.

A Quick Example

If you calculate that the correlation between hours studied and exam scores is r = 0.9, then:

  • r² = (0.9)² = 0.81
  • Your regression R² will also equal 0.81

This means 81% of the variance in exam scores is explained by hours studied—whether you get there through correlation or regression.

Important Limitation

This equality only holds for simple linear regression (one predictor). With multiple predictors, R² no longer equals any single correlation squared—it becomes its own distinct measure of model fit.

Key Takeaway: In simple linear regression, R² and r² are identical, revealing that both capture how much variance in Y is explained by X.