Confidence Interval for Difference in Means

Estimate how much two population means differ. The interval uses Welch’s method by default, which doesn’t assume equal variances. Tick “pooled” for the classic equal-variance interval.

(x̄₁ − x̄₂) ± t* × √(s₁²/n₁ + s₂²/n₂)

If the interval excludes 0, the difference is statistically significant at the matching α. Unlike a p-value, it also shows how big the difference could plausibly be.

Example

Group 1: x̄ = 82, s = 8.5, n = 20; group 2: x̄ = 76, s = 9.2, n = 18. The difference is 6 points with Welch SE 2.88 and df ≈ 34.8, giving a 95% interval of about [0.15, 11.85]. The groups differ significantly, but the true gap could be anywhere from almost nothing to about 12 points. That’s why reporting the interval is more informative than the p-value alone.