Advanced level · Module 6: Data Science Capstone · Lesson 2 of 5
Impact Analysis
Estimate the programme's effect on food consumption with a credible design and sensitivity checks.
In this lesson you will
- Apply a causal design to the programme data
- Report effects with assumptions and sensitivity
Key concepts
design choiceassumption checkseffect estimatesensitivity analysis
The lesson
Choose the design that fits the data — likely difference-in-differences using rollout timing — and state its assumptions.
Check assumptions with pre-programme trends and run sensitivity analyses with alternative comparison groups.
Report the effect with its interval and translate it into meaningful terms, such as additional meals per week.
Step by step
- Choose and justify the design.
- Check assumptions.
- Estimate the effect.
- Run sensitivity analyses.
Practice activity
Complete milestone 2: the impact analysis.
Expected result: A credible, checked effect estimate with sensitivity results.
Check your understanding
-
How can the parallel-trends assumption be checked?
Show answer
By comparing groups' trends before the programme started.
-
Why translate effects into meaningful terms?
Show answer
So decision-makers understand what the effect means for people.
Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.