One-Sample t-Test: Confidence Intervals
What you'll learn: How to construct confidence intervals for a population mean when the population variance is unknown using the t-distribution.
Why Use the t-Distribution?
When you don't know the population standard deviation σ (which is almost always the case in real-world problems), you estimate it using the sample standard deviation s. This introduces extra uncertainty, so instead of the normal distribution, you use the t-distribution to build your confidence interval.
The Formula
For a population mean μ with unknown variance, the (1 - α)×100% confidence interval is:
x̄ ± t_(α/2, n-1) · (s/√n)
Where:
- x̄ = sample mean
- s = sample standard deviation
- n = sample size
- t_(α/2, n-1) = critical value from the t-distribution with (n-1) degrees of freedom
- s/√n = standard error of the mean
How It Works
Think of this interval as a "net" you cast around your sample mean. The t-distribution has heavier tails than the normal distribution (especially for small samples), so your net needs to be wider to maintain the same confidence level. As your sample size grows, the t-distribution approaches the normal distribution, and your interval naturally becomes more precise.
Example
Suppose you measure heights of 16 students: x̄ = 170 cm, s = 8 cm. For a 95% confidence interval (α = 0.05), you find t_(0.025, 15) ≈ 2.131 from a t-table.
Your interval: 170 ± 2.131 · (8/√16) = 170 ± 4.26 = [165.74, 174.26] cm
Key Takeaway: When constructing confidence intervals for a population mean with unknown variance, use the t-distribution with critical value t_(α/2, n-1) and standard error s/√n to account for the additional uncertainty from estimating σ.