What Q-Q Plots Show: Comparing Residual Distribution to Normal
You'll learn what Q-Q plots visualize and why they're a critical tool for checking the normality assumption in regression diagnostics.
What Is a Q-Q Plot?
A Q-Q plot (quantile-quantile plot) is a graphical tool that compares two probability distributions by plotting their quantiles against each other. In regression diagnostics, we use it specifically to check if our residuals follow a normal distribution.
Think of it like a "matching game" for distributions. If your residuals are normally distributed, the Q-Q plot will show points falling almost perfectly along a straight diagonal line. Any deviation from that line signals that your residuals deviate from normality.
How It Works
The plot places:
- Theoretical quantiles (what we'd expect from a perfect normal distribution) on the x-axis
- Sample quantiles (your actual residual values, sorted) on the y-axis
If your residuals are normal, each quantile from your data should match the corresponding theoretical quantile from the normal distribution. Points line up straight—like two synchronized dancers.
What Deviations Mean
- S-shaped curve: Heavy-tailed distribution (more extreme values than normal)
- Inverted S-shape: Light-tailed distribution (fewer extreme values)
- Points curving upward at ends: Right skew
- Points curving downward at ends: Left skew
- Straight line: Normality assumption holds ✓
This matters because regression inference (confidence intervals, hypothesis tests) relies on the normality of residuals assumption.
Key Takeaway: Q-Q plots compare your residuals' distribution to a theoretical normal distribution—if points follow a straight diagonal line, the normality assumption is satisfied; deviations indicate non-normality that may require transformation or alternative methods.