One-Sample t-Test: Testing a Population Mean
What you'll learn: How to test hypotheses about a population mean when you don't know the population standard deviation, using the t-distribution.
The Core Idea
In the real world, we rarely know the population standard deviation σ. The one-sample t-test lets you test claims about a population mean μ when σ is unknown by using the sample standard deviation s instead. Because this introduces extra uncertainty, we use the t-distribution (which has heavier tails than the normal distribution) rather than the z-distribution.
The Test Procedure
You follow the same hypothesis testing logic you learned with z-tests, but with a crucial swap:
- State hypotheses about the population mean μ (null: μ = μ₀; alternative: μ ≠ μ₀, μ > μ₀, or μ < μ₀)
- Calculate the t-statistic:
t = (x̄ - μ₀) / (s/√n)where x̄ is the sample mean, s is the sample standard deviation, and n is the sample size - Find degrees of freedom: df = n - 1
- Compare your t-statistic to the t-distribution with df degrees of freedom to get a p-value or check against critical values
- Decide: reject or fail to reject the null based on your significance level
When to Use It
Use the one-sample t-test when:
- You're testing a claim about one population mean
- The population standard deviation is unknown
- Either the population is normally distributed OR your sample size is large enough (typically n ≥ 30) for the CLT to apply
Key Takeaway: The one-sample t-test extends hypothesis testing to realistic scenarios where σ is unknown, using the sample standard deviation and the t-distribution to account for the additional uncertainty this creates.