Worked examples: the reproducible analysis behind every lesson

Every lesson we publish has a chapter called "the worked example". This is that document. Real data, the analysis that produced every number on screen, the source you can re-run, and the limits of what each method can tell you.

• Your A/B test says the new page won. Did it?

  The worked example behind our A/B testing lesson: 2,300 visitors over four weeks, where the raw conversion count and the conversion rate name opposite winners, and the thing that separated them was the rollout schedule.

• Where is the biggest leak in your funnel?

  The worked example behind our funnel lesson: 2,500 accounts through five steps, where "biggest leak" turns out to be two questions naming different steps, and a channel that is best at one step and worst at the next.

• How long does a subscription account really last?

  The worked example behind our survival lesson: 320 B2B accounts, a third still running, and why averaging only the churned ones understates account lifetime by 44%. Three independent implementations agree.

• Which multiple-comparison correction do you actually need?

  The reproducible worked example behind our multiple-comparisons lesson: 66 pairwise correlations on 1,599 wines, corrected four ways, cross-checked against R's own p.adjust.

• McNemar's test or chi-square? Paired yes/no data, decided.

  The reproducible worked example behind our McNemar lesson: 200 paired approve/deny decisions, agreement, kappa, and the four variants of the test.

• Which ICC do you actually need?

  The reproducible worked example behind our ICC lesson: the same ratings scored by different ICC forms, with confidence intervals.