Difference-in-differences and maternal health vouchers in Bangladesh

Difference-in-differences subtracts the comparison group's change from the treated group's change. The comparison group's trend stands in for what would have happened without the program, so the design handles both time trends and baseline gaps, provided the trends would have run parallel. Bangladesh's maternal health vouchers went to poorer subdistricts, where only 7.2% of women delivered in a facility. Event study estimates near zero before rollout supported parallel trends. Facility delivery rose about 6.5 percentage points, yet neonatal and infant mortality did not fall.

Also covered: parallel trends · event study · staggered rollout · utilization vs health outcomes

Alongside this step

Slides and transcript — the whole video
Read the chapter — Quasi-Experimental Designs

Lesson 2 · 6 steps

1What is your identification strategy?6:19 2Comparing groups without baseline data4:19 3Difference-in-differences and maternal health vouchers in Bangladesh4:43 4Interrupted time series and the kidney transplant wait-time policy6:02 5Regression discontinuity and HPV vaccination in Ontario5:36 6Instrumental variables, and choosing a design8:51

← Previous step

Research Designs · lesson 2 of 5