Lesson 2 · Quasi-Experimental Designs · Step 2 of 6
Comparing groups without baseline data
Comparing a treated group with an untreated one at the same moment removes the history problem of a pre-post design, because both groups lived through the same events. It replaces it with selection: why did one group get the intervention? The Millennium Villages Project ran for ten years before its endline evaluation, so comparison villages had to be matched retrospectively. Thirty of 40 outcomes favored the project villages, including 23 fewer child deaths per 1,000 live births. Without baseline data, though, no amount of matching can show the villages were comparable when the project began.
Also covered: post-test only comparisons · selection bias · retrospective matching · Millennium Villages Project
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