10.4. Chapter Exercise#
Apply the key ideas from this chapter in the connected task below. Work in your own notebook or local project. Keep your implementation, reasoning, and verification results together.
10.4.1. Separate group differences from causes#
Use the supplied randomized experiment and observational comparison to define a group-difference statistic and the relevant assignment or sampling mechanism.
Run the taught permutation or re-randomization procedure for each case and report observed differences and simulated evidence.
Explain how confounding changes the interpretation of the observational result. Identify the population and causal claim each design can support and one uncertainty that remains after a small p-value.
Return to the chapter lessons as needed: A/B Testing, Comparing Groups Without Random Assignment, Causality.