Lesson 2 · Quasi-Experimental Designs · Step 6 of 6
Instrumental variables, and choosing a design
An instrument shifts who gets treated and affects the outcome only through that treatment. Randomizing encouragement creates one: in Tangiers, door-to-door help with paperwork raised piped water connections from 10% to 69%. Connections didn't improve child health, because families already had clean public taps, but they freed over seven hours a week of water fetching and raised life satisfaction. The estimate applies to compliers, the households the help actually moved. Across all six designs, the question is the same: which assumption about the counterfactual can you defend here?
Also covered: exclusion restriction · encouragement designs · compliers and the LATE · choosing among quasi-experimental designs
Alongside this step
- Slides and transcript — the whole video
- Read the chapter — Quasi-Experimental Designs
Check yourself
5 questions, marked as you go. Nothing is recorded as a grade and nothing is locked behind it — if a question catches you out, that is the useful part.
Lesson 2 · 6 steps
Research Designs · lesson 2 of 5