Correlation vs Regression
What you'll learn: You'll understand how correlation measures the strength of a relationship while regression predicts one variable from another.
The Core Distinction
Though correlation and regression both deal with relationships between two variables, they serve fundamentally different purposes.
Correlation is a single number (typically Pearson's r, ranging from -1 to +1) that quantifies how strongly two variables move together. It's symmetric—the correlation between height and weight is the same as the correlation between weight and height. Correlation answers: "Is there a linear association, and how strong is it?"
Regression goes further by building a predictive model. Simple linear regression creates an equation (ŷ = b₀ + b₁x) that lets you estimate one variable (the response) from another (the predictor). Regression is asymmetric—predicting weight from height gives you a different equation than predicting height from weight. Regression answers: "What value of Y should I expect for a given X?"
A Helpful Analogy
Think of correlation like asking "Are these two dancers moving in sync?" You're just observing their coordination. Regression is like saying "If dancer A moves 3 steps forward, how far will dancer B move?" You're using one to predict the other.
When to Use Each
Use correlation when you want to explore whether two variables are related, without assuming one causes or predicts the other (e.g., "Are test scores and study hours associated?").
Use regression when you want to make predictions or understand how changes in one variable affect another (e.g., "For each additional hour studied, how much does the test score increase?").
Key Takeaway: Correlation measures the strength and direction of association between variables; regression builds a predictive equation to estimate one variable from another.