Week 11 · Suggested 90 minutes · First-pass draft
An exact test and a computational interval
Demonstrate two inferential ideas that can be understood without memorizing a reference-distribution table.
Applied assignment, not a stored grade. Instructions and model checks are provided for self-study or instructor adaptation. Work is not transmitted to this site.
What you will do
- Enumerate the 70 balanced assignments of four of eight participants and calculate the exact two-sided randomization p-value for the given outcomes.
- Compare the one-sided and two-sided alternatives, explaining why they must be specified before seeing the results.
- Bootstrap the fictional waiting-time data, changing the number of replicates while preserving the original observations.
- Compare Monte Carlo variability with uncertainty inherent in the original sample and identify design limits.
What to submit
One exact randomization table, one bootstrap summary, and a plain-language limitations memo.
Self-check before finishing
Under the stated sharp null, the two-sided exact randomization p-value is 2/70 ≈ .02857.
Assigned reading
Enumerate assignments and resample empirical data.
Instructor discussion guidance
Require the sharp-null and randomized-assignment assumptions; label synthetic data accurately.
Print a predict–experiment–explain worksheet → · Open browser R →