5 min read

Scatter Plot Correlation & Regression: Interpreting r, R², and 95% CI Bands

How to build honest regression scatter plots that communicate both trend strength and estimation uncertainty.

Scatter Plot Maker

Regression trendlines, 95% CI bands, R², Pearson r, and multi-group coloring.

Open Scatter Plot Maker →

Direction, form, and strength

Before fitting any equation, scan the raw point cloud for four properties: direction (positive or negative), form (linear or curved), strength (tight or dispersed), and leverage points. Two very different datasets can share an identical Pearson r, which is why plotting the raw points is non-negotiable.

Reading r, R², and the slope p-value

Pearson r ranges from −1 to +1 and captures linear association strength. R² (the square of r in simple linear regression) represents the fraction of Y variance explained by X. The slope p-value tests the null hypothesis that the true slope equals zero; always interpret it alongside effect magnitude and sample size.

What the shaded 95% confidence band shows

The shaded band around an OLS regression line is narrowest near the sample mean of X and flares slightly at both ends. It represents 95% confidence in the mean regression line—not a prediction interval for 95% of individual future observations.

Open Scatter Plot Maker →← All guides