Objetivos

• Learn methods for estimating causal effects using experimental and observational data
• Understand the assumptions of common methods for impact evaluation
• Demonstrate a sound understanding of the advantages and limitations of policy impact evaluation methods
• Interpret and critically analyse studies applying methods of impact evaluation.

Contenido

Week 1:
Part 1: The experimental design for evaluating policy impacts (RCT)
i) Experimentation and the potential outcome framework
ii) Threats to selection bias and causal inference
Part 2: Alternatives to the random allocation of treatment
i) Propensity Score Matching
ii) Difference-in-Difference
iii) Regression Discontinuity Design

Week 2:
Alternatives and complements to the random allocation of treatment
iv) Instrumental Variables in Policy Evaluation
Part 3: Unpacking the “black box” of causal impact:
i) Mediation analysis
ii) Structural modelling
ii) Measurement
Part 4: From design to scaling empirical results?
What can we learn from RCTs?
Threats to scalability

Tutorials:
The tutorials are generally hands-on lab sessions using STATA, involving small group work and discussions. In these sessions, students have an opportunity to reproduce parts of empirical papers using the method taught in the previous day’s lecture. The tutorials’ discussions focus on the interpretation of the results, and are opportunities to reflect upon the strengths and limitations of each method.
Week 1
Tutorial 1: Group discussion - analysis of an impact evaluation application, its strengths and weaknesses and familiarising with STATA commands.
Tutorial 2: Lab application on RCTs.
Tutorial 3: Lab application on cluster RCTs.
Tutorial 4: Lab application on Propensity Score Matching.
Tutorial 5: Lab application on Difference-in-Difference.
Week 2
Tutorial 6: Lab application on Regression Discontinuity Design.
Tutorial 7: Lab application on Instrumental Variables.
Tutorial 8: Lab application on mediation analysis.
Tutorial 9: Lab application on structural modelling
Tutorial 10: Formative exam and feedback
Assessment and assessment criteria:
The course is assessed by a 2 hours written examination. The exam consists of two parts. Part A includes exercises and problems where students need to show their understanding of the methods covered during the course. Part B focuses on the interpretation of results of empirical papers.

Assessment criteria:
i) Your understanding of the methodological assumptions. This means we assess your ability to evaluate key issues and evidence of critical approach to the evaluation question you are asked to address.
ii) Depth of analysis: this means we assess your ability to interpret evidence from impact evaluation studies, assess the evidence strengths and ability to make informed judgments.