Lesson 2 · Quasi-Experimental Designs · Step 4 of 6
Interrupted time series and the kidney transplant wait-time policy
When a policy reaches everyone at once there is no comparison group, so an interrupted time series uses the unit's own past. Segmented regression asks whether the outcome jumped at the intervention and whether its slope changed. The design needs many points on each side, ideally 12 or more, and its central threat is anything else that changed at the same moment. When the transplant network restored wait time to Black candidates harmed by a race-based kidney equation, transplants rose by 5.3 per 1,000 listings, but only for patients the system could already see.
Also covered: level and slope change · segmented regression · concurrent events · autocorrelation and seasonality
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