- Docente: Denni Tommasi
- Credits: 5
- SSD: SECS-P/05
- Language: English
- Teaching Mode: Traditional lectures
- Campus: Bologna
- Corso: Second cycle degree programme (LM) in Economics and Economic Policy (cod. 8420)
-
from Sep 19, 2022 to Oct 24, 2022
Learning outcomes
We will build the statistical foundations for field experiments from the ground up. We will discuss the logic of randomized control trials (RCTs) conducted in real-world settings, their strengths and weaknesses compared to other methodologies, and the ways in which experimentation has been, and could be, used to investigate social phenomena. Students will learn how to interpret, design, and execute experiments.
Course contents
The course is targeted to students interested in understanding policy evaluations (i.e., public, health, labour, or development policies) as well as more business-oriented students interested in understanding business evaluations and A/B testing (i.e., useful for understanding user engagement and satisfaction of online features like a new feature or product).
The course is divided as follows:
Week 1: motivating examples of field experiments, potential outcomes framework, assumptions; ATEs, ITTs, LATEs; sampling distributions and randomization inference.
Week 2: blocking and covariate adjustment; balance checks; field experiments with one-sided noncompliance (failure-to-treat); field experiments with two-sided noncompliance (encouragement designs)
Week 3: peer effects; cluster designs; phase-in designs; determining optimal sample size; power calculations.
Week 4: sample attrition, interference between experimental units; heterogeneous treatment effects; mediation and causal mechanisms.
Week 5: Field experiments from A to Z: case studies and student presentations of the outline of the take home project, receiving feedback.
Readings/Bibliography
Gerber and Green (2012), “Field Experiments: Design, Analysis and Interpretation”.
Teaching methods
For each topic, we will first introduce the relevant statistical and econometric theory, and then move to Stata codes, examples and applications.
Assessment methods
Problem sets (20%): students will apply the concepts learnt in class through a set of applications using Stata.
Presentations (20%): students will be assigned a paper analysing the results of a field experiment and they will present it in class.
Take home exam (60%): students will be given a set of applied questions they need to answer within 7 days.
Teaching tools
Slides and Stata codes.
Office hours
See the website of Denni Tommasi