📊 Chapter 9: Hypothesis Testing and Inference

🎮 Interactive Apps (HTML/JS Simulations)

Note: Apps 1–3 cover the essential content clearly. Older apps remain for reference.

• 9.1.1 – Intro to Hypothesis Testing
• 9.1 – Full Hypothesis Testing Walkthrough
• 9.1.2 – Type I & Type II Errors — ICE
• 9.1.3 – One-sided vs Two-sided Tests (FYI)
• 9.1.4 – P-values (FYI)
• 9.1.6 – Testing Procedure Summary
• 9.2 – Normal Distribution in Testing (FYI)
• 9.3 – t-Test Visualization (FYI)
• Chapter 9 Review: Quiz & Recap

📽️ PPT Slides

Download Chapter 9 Slides (ch09.pptx)

📝 Quiz

• Quiz 1 — Hypothesis Basics (z/t, p-values, one- vs two-tail)

  15 T/F. Covers H₀/H₁, μ₀/μ₁, α, z vs t, p-values, and reject vs. fail to reject.

  Open Quiz 1

• Quiz 2 — Type I, Type II, Power & Choosing μ₁

  15 T/F. α, β, Power, effect size δ, and practical μ₁ selection.

  Open Quiz 2

📚 Homework

1. Two-Tailed z-Test Homework (Dorm Commute)
2. One-Tailed t-Test Homework (Dorm Commute)

💬 Excel

• Hypothesis Testing with Excel

🙋 Student Q&A

• What is a hypothesis test?
  A procedure to decide if data provide enough evidence to reject a claim (H₀) about a population parameter.
• When do I use a z-test vs t-test?
  z: σ known (or large n with reliable historical σ). t: σ unknown and you estimate with s (especially small/medium n).
• What does α (e.g., 0.05) mean?
  Your tolerance for a false alarm (Type I error): rejecting a true H₀ about 5% of the time.
• Define Type I and Type II errors.
  Type I (α): reject true H₀. Type II (β): fail to reject false H₀.
• What is Power?
  1 − β: probability you correctly detect a real effect (reject H₀ when H₁ is true).
• How does sample size n affect Power?
  Bigger n ↓SE, so β goes down and Power goes up.
• One-sided vs Two-sided?
  Two-sided checks for any difference (≠). One-sided tests for a specific direction (> or <).
• What’s μ₁ and δ?
  μ₁ is a specific true mean to evaluate β/Power. δ = |μ₁ − μ₀| is the effect size you care to detect.
• How do p-values relate to α?
  If p ≤ α → reject H₀; if p > α → fail to reject. p is computed assuming H₀ is true.
• CI vs hypothesis test?
  For two-sided tests, if μ₀ lies outside the (1−α) CI, you would reject H₀ at level α.
• Common mistakes?
  Interpreting p as “probability H₀ is true”; choosing μ₁ to inflate Power; ignoring practical δ.
• Reporting template?
  “Tested H₀: μ=μ₀ at α=0.05 with n=…; z/t=…, p=… → [reject/fail to reject] H₀. CI: […, …]. Power (δ=…): …”.