Week 14 — Readings & Resources

Your primary reading is Chapter 14 (in this module). Everything below is the optional, go-deeper layer — pick what helps, skip the rest.

How to use this page

This page has 11 items (5 readings and 6 videos). Allow 92–107 minutes for everything. Each group points back to the matching chapter section and class-session segment, so use the full menu when you want more practice, or take the quick path below.

Every video here has captions: turn them on with the CC button in the player. The chapter covers the same ideas in text.

Pick-one quick path (18–22 min total)

• Hypothesis Test for a Mean — Stat Trek
• Welch (Unpooled Variance) t Tests and Confidence Intervals: An Example — jbstatistics

① Is the average really off the claim? The one-sample t test

Chapter 14, Section “Run the one-sample t test: count standard errors to the claim” · Session 1, Segment 1, “The one-sample t test” · Remember: t counts how many standard errors the sample mean sits from the null value, and a one-sample test uses n − 1 degrees of freedom.

• Hyp testing #3: Testing for μ when σ is unknown — zedstatistics — one full test on ten delivery times, checking a store's claim about its average delivery time at the 5% level. Its alternative is one-sided, because the claim is about staying under a limit; the chapter's penguin test is two-sided. (video, 10 min) (captions: automatic)
• Hypothesis Test for a Mean — Stat Trek — read when to use the test, then the steps from stating the hypotheses through interpreting the results. Follow the t-score branch, since the population standard deviation is unknown, and skip the calculator. (reading, 8–12 min)

② Two groups: the two-sample t test and its software output

Chapter 14, Section “Read the software output: Welch's t and its degrees of freedom” · Session 1, Segment 2, “Two means and Welch's t” · Remember: for two independent groups, divide the difference in sample means by its standard error, and let the software report Welch's degrees of freedom.

• T-Tests: A Matched Pair Made in Heaven: Crash Course Statistics #27 — CrashCourse — a lively build of a two-sample t test from a coffee tasting that compares two cafes. It reads the statistic as the number of standard errors between the observed difference and 0. Its later part on matched pairs goes beyond this course, so you can stop when that begins. (video, 11 min) (captions: creator)
• Welch (Unpooled Variance) t Tests and Confidence Intervals: An Example — jbstatistics — one careful worked example for two independent groups, with the test and the matching interval side by side. Notice where the degrees of freedom come from; in this course the software reports them. Use it while completing the Week-14 module overview. (video, 10 min) (captions: creator)

③ Shares, not averages: tests for one and two proportions

Chapter 14, Section “Compare two proportions: pool the samples under the null hypothesis” · Session 2, Segment 3, “Tests for one and two proportions” · Remember: a proportion test puts the null value in the standard error, and a two-proportion test pools both samples because H₀ says the two shares are equal.

• Hyp testing #4: testing for the population proportion, p (or π) — zedstatistics — a restaurant chain claims a share of happy customers, and a survey of 460 customers checks it. The same result is judged at the 5% level and at the 1% level, so you can see one z statistic reach two different decisions, as the chapter's poll does. (video, 11 min) (captions: automatic)
• Two-Sample Z Test for Difference in Proportions — Michael Porinchak — a slow, step-by-step two-proportion test: name both parameters, write the hypotheses, check the conditions, then compute the pooled proportion and the z statistic. It adds a 10% condition that the chapter does not use; the large-counts check is the same. (video, 14 min) (captions: automatic)
• Hypothesis Test: Difference Between Proportions — Stat Trek — read when to use the test, then the steps through interpreting the results. The test-statistic step shows the pooled proportion inside the standard error. Skip the practice problems at the end. (reading, 8–12 min)

④ Two views of one question: intervals, tests, and not significant

Chapter 14, Section “See tests and intervals as two views of one question” · Session 2, Segment 4, “Tests and intervals together” · Remember: a two-sided test at α = 0.05 rejects exactly the null values outside the 95% interval, and failing to reject leaves a range of values plausible.

• The Relationship Between Confidence Intervals and Hypothesis Tests — jbstatistics — a short, careful look at why a two-sided test and the matching confidence interval give the same verdict about a null value. (video, 6 min) (captions: creator)
• Failing to Reject the Null Hypothesis — Statistics By Jim — why a test that is not significant does not show there is no difference, told through a species presumed extinct and a criminal trial. This builds on Week 13's work with failing to reject versus accepting H₀. (reading, 5–7 min)
• Hypothesis Testing and Confidence Intervals — Statistics By Jim — read the parts on confidence intervals and statistical significance and on why the two methods always agree. A fuel-cost example reaches the same verdict twice: a p-value below 0.05, and an interval that leaves out the null value. (reading, 5–8 min)

Optional one-stop reference (free online reference)

• 1.3.5.3. Two-Sample t-Test for Equal Means — NIST/SEMATECH e-Handbook of Statistical Methods — a free government reference. Read the purpose and the definition, including the version for unequal standard deviations and its degrees of freedom. Skip the case study and the software notes. (reading, 4–6 min)

Heads-up (links rot): these point to outside sites that occasionally move or rename pages. If a link ever fails, tell your instructor and use the reference above in the meantime.

Provenance: Built with The Course Maker · samples.thecoursemaker.com