Two-tailed hypothesis tests
Objective. Allocate significance and identify critical regions.
- 1. Learn
- 2. Worked example
- 3. Practice
- 4. Feedback
- 5. Continue
Learn: the key idea
A two-tailed hypothesis test checks for a difference in either direction, so the chosen significance level α is split equally between both tails of the distribution, giving critical regions at α/2 in each tail. This contrasts with a one-tailed test, where the alternative hypothesis specifies a direction and the whole α sits in one tail. For a 5% two-tailed test on a normal distribution, the critical z-values are ±1.96, found from the tables using 2.5% in each tail. Correctly identifying which type of test applies, from the wording of H1 (≠ for two-tailed, < or > for one-tailed), is essential for selecting the right critical region.
Worked example
A test uses H0: μ = 50, H1: μ ≠ 50 at the 5% significance level. State the two-tailed critical z-values and decide if z = −2.1 is significant.
- 1Two-tailed at 5% splits α as 2.5% per tail
- 2Critical values are z = −1.96 and z = 1.96
- 3Compare: −2.1 is more extreme than −1.96
- 4Since |−2.1| > 1.96, the result is significant: reject H0
Answer: Critical values ±1.96; z = −2.1 is significant, reject H0
Concept mastery
Based on your past attempts at this lesson — the question types to practise again come first.
Complete a practice set to see your concept breakdown here.
Practice
Practice happens on its own screen, one question at a time. Answers stay hidden until you submit yours, and you can stop and pick up at the same question later.
Start practisingCommon mistake. Students use a one-tailed critical value (e.g. 1.645) when H1 uses ≠, which requires splitting α between two tails. Always check whether H1 states a direction or simply "not equal" before choosing the critical value.
Your progress
Tick each step as you finish it. Your exact reading position is saved automatically, so Continue drops you back on the same line.
0 of 5 steps complete
Your place saves automatically as you read.