Course contentsShow
← Mathematics
Lesson 3016 of 3,12159. Statistics — Descriptive and InferentialFree lesson

Confidence Interval for the Difference of Two Means

Construct intervals to estimate the difference between two population means from independent samples.

Confidence Interval for the Difference of Two Means

What you'll learn: You'll learn how to construct an interval that estimates the true difference between two population means using independent samples.

The Core Idea

Sometimes we want to compare two groups—does the new teaching method improve test scores more than the old one? Is the average height of men truly different from women? Instead of just comparing sample means, we build a confidence interval for the difference to quantify our uncertainty.

When you have two independent samples from two populations, you can estimate μ₁ - μ₂ (the difference of population means) using the difference of sample means: x̄₁ - x̄₂.

The Formula

The confidence interval takes the form:

(x̄₁ - x̄₂) ± (critical value) × SE(x̄₁ - x̄₂)

The standard error of the difference is:

SE = √(s₁²/n₁ + s₂²/n₂)

where s₁, s₂ are sample standard deviations and n₁, n₂ are sample sizes.

Which Critical Value?

  • If both population standard deviations are known (rare): use z* from the standard normal
  • If unknown (common): use t* from a t-distribution with degrees of freedom approximated by the complex Welch-Satterthwaite formula, or conservatively use min(n₁-1, n₂-1)

Interpretation

A 95% CI of (2, 8) for μ₁ - μ₂ means: "We're 95% confident the first population's mean exceeds the second's by between 2 and 8 units." If the interval contains zero, the difference might not be significant.

Key Takeaway: The confidence interval for the difference of two means combines the variability from both samples to create a range that likely captures the true population difference, allowing meaningful comparisons between two groups.