One-Sample z-Test for Means
What You'll Learn
You'll learn how to formally test a claim about a population mean when you know the population standard deviation σ, using the standard normal distribution.
The Core Idea
The one-sample z-test is your tool when you want to test whether a population mean μ equals some specific value, and you happen to know the population standard deviation σ. This is the first "complete" hypothesis test you'll perform with all the pieces in place.
Here's how it works:
-
Set up hypotheses: Your null hypothesis typically states H₀: μ = μ₀ (some claimed value), while your alternative might be H₁: μ ≠ μ₀ (two-sided), H₁: μ > μ₀, or H₁: μ < μ₀
-
Calculate the test statistic: You compute the z-statistic:
z = (x̄ - μ₀) / (σ/√n)where x̄ is your sample mean, μ₀ is the claimed population mean, σ is the known population standard deviation, and n is your sample size
-
Find the p-value or critical region: Under H₀, this z-statistic follows a standard normal distribution (thanks to the Central Limit Theorem for reasonable sample sizes). You compare your calculated z to the standard normal distribution
-
Make a decision: If your p-value is less than your significance level α, or if z falls in the critical region, you reject H₀
Analogy: Think of it like a "how unusual is this?" meter. If your sample mean is so far from the claimed value that it would be very unlikely (under a normal curve), you have evidence against the claim.
Key Takeaway: The one-sample z-test uses the standardized difference (x̄ - μ₀)/(σ/√n) as a test statistic that follows N(0,1) under H₀, allowing you to assess whether sample data contradicts a claim about the population mean when σ is known.