Track 8 · Hypothesis Testing · Lesson 12 of 14
Checking assumptions
Check the independence, normality and equal-variance conditions of t-tests with plots and judgement, understand how robust the tests are, and choose a sensible alternative when a condition fails.
35 minIntermediate7 graded exercisesMembers
Every p-value in this track was computed under assumptions about how the data arose. If they fail badly, the p-value and interval can be wrong without any warning sign in the output. Knowing which assumptions matter, how to check them and what to do when they fail is what separates running a test from defending one.
Included with a Formula Quest membership
It has 7 graded exercises with feedback, hints and worked solutions. After it you will be able to:
- Rank the t-test conditions: independence first, then normality of the mean, then equal variances
- Read normal Q-Q plots for skew, heavy tails and outliers
- Interpret normality and variance tests in light of the sample size
- Choose and report a robust alternative (transformation, rank test, Welch) when a condition fails