Correlation vs Regression
Linear Correlation — Chapter 6, GSEB Class 12 Statistics
Key Differences
| Aspect | Correlation | Regression |
|---|---|---|
| Purpose | Measures strength of association | Predicts value of one variable from another |
| Output | r (single number) | Y = a + bX (equation) |
| Direction | Symmetric — r(X,Y) = r(Y,X) | Not symmetric — Y on X ≠ X on Y |
| Cause-Effect | No assumption about cause | Implies X → Y (cause to effect) |
| Number of Results | One value (r) | Two lines (Y on X and X on Y) |
| Range | −1 to +1 | No fixed range for b (slope) |
Deep Dive
Correlation Answers: “How strongly related?”
- Correlation is symmetric— r(X,Y) = r(Y,X). It doesn't matter which is X and which is Y.
- It produces a single number (r) between −1 and +1.
- It does not assume cause and effect — just measures co-movement.
- Example: “height and weight have r = 0.85 — strongly related”.
When to Use Which?
Use Correlation When:
- →You want to know if two variables are related
- →You need a quick measure of association strength
- →You want to screen variables before building a model
Use Regression When:
- →You want to predict Y from X
- →You need the exact relationship (equation)
- →You want to estimate the effect of X on Y
Remember: Correlation does not imply causation. Two variables can be strongly correlated without one causing the other (e.g., ice cream sales and drowning rates are correlated because both increase in summer — not because one causes the other).
Key Takeaways
Key Takeaways
- Correlation = measure of association; Regression = method to predict.
- Correlation is symmetric (r(X,Y) = r(Y,X)); Regression is not (Y on X ≠ X on Y).
- Correlation gives a single number (r); Regression gives an equation (Y = a + bX).
- Correlation assumes no cause-effect; Regression implies X → Y.
- Correlation answers "how strongly?"; Regression answers "what value of Y for given X?".
- Correlation does not imply causation — two variables can be correlated without a causal link.