AP Statistics / Unit 1: Exploring One-Variable Data and Collecting Data
1.13 Experimental Design
Notes
Principles: comparison (two or more treatments), random assignment, replication (enough units), and control of other variables.
- Completely randomized, randomized block (group similar units first), and matched pairs designs.
- Placebo and blinding reduce bias.
Practice
Why randomly assign treatments?
It balances confounding variables.
Grouping subjects by age before randomly assigning treatments within each age group is:
Blocks are similar groups.
Double-blind means:
Both sides are blind.
Ready for more? Take the Unit 1 test.