Lesson 2 · Quasi-Experimental Designs · Step 1 of 6
What is your identification strategy?
In 1854 John Snow couldn't randomize which Londoners drank which water, but two companies served the same streets from different parts of the Thames, and households drinking the contaminated supply died at roughly eight times the rate. The credibility revolution asked every study the same question: what feature of your design, not your statistical model, makes the estimate causal? The simplest design, measuring before and after in one group, assumes nothing else changed. A South African parenting study used it deliberately, as a safe first step that sized a future cluster trial.
Also covered: identification strategy · credibility revolution · pre-post designs · history and maturation
Alongside this step
- Slides and transcript — the whole video
- Read the chapter — Quasi-Experimental Designs
Lesson 2 · 6 steps
Research Designs · lesson 2 of 5