Accueil > Evaluating social programs: randomized experiments


Type d'enseignement : Seminar

Semester : Spring 2017-2018

Number of hours : 24

Language of tuition : English

Voir les plans de cours et bibliographies


No prerequisites. This course is not designed for economists already having a good knowledge of randomized impact evaluations, but for a general audience wanting to understand how they work and what they can bring to the public debate.

Course Description

Randomized experiments are getting more and more popular to measure the impact of social policies and programs. The goal of this course is to provide a detailed understanding of this methodology and of its results. The first part of the course will address the methodology: Why and when is a rigorous evaluation of social impact needed? What are the common pitfalls of evaluations, and why does randomization help? What are the key components of a good randomized evaluation design? How do you determine the appropriate sample size, measure outcomes, and manage data? What kind of results do you get? The second part of the course will present examples of completed field experiments in the areas of education and private sector development and the last part will discuss advantages and limitations of randomized evaluation and how they can be used to better understand global poverty. Throughout the sessions, students will be prepared to conceive a group presentation giving the main lines of a possible randomized evaluation.


POULIQUEN, Victor (Chercheur)

Pedagogical format

12 weeks.

Course validation

50% individual work (to be delivered at the end of February) ; 50% group project.


Some readings, a midterm essay, a group presentation.

Required reading

Abhijit V. Banerjee and Esther Duflo (2011): “Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty”. (Version Française disponible sous le titre « Repenser la pauvreté »)

Additional required reading

  • Glennerster, Takavarasha: “Running Randomized Evaluations: a Practical Guide”, Princeton University Press, 2013
  • Gertler, Martinez, Premand, Rawlings, Vermeersch : « Impact Evaluation in Practice , Second Edition », The World Bank, 2016 (PdF available on the Internet)
  • Duflo, E.: “Human values and the design of the fight against poverty”, Tanner Lectures, May 2012

Plans de cours et bibliographies

  • Session 1: Why randomize? Introduction to Randomized Impact Evaluations
  • Session 2: Theory of change and theory of measurement
  • Session 3: How to randomize?
  • Session 4: How to calculate sample size?
  • Session 5: Managing threats to an Impact Evaluation
  • Session 6: An evaluation from A to Z:  A Cash Transfer Program for Education in Morocco
  • Session 7: Preparing group works
  • Session 8: An evaluation from A to Z: Improving microbusinesses regulation in Benin
  • Session 9: Advantages and limitations of RCTs to fight poverty
  • Session 10: Looking for the « big pictures »
  • Session 11: Group presentations
  • Session 12: Group presentations

Biographical Information

Victor Pouliquen joined J-PAL in 2008 and has worked on numerous randomized trials relating to health, education, private sector development and governance. He is currently working with J-PAL as a principal investigator on a family planning project in Burkina Faso and on a public finance project in Senegal. Since 2013, he is also working as a consultant for the World Bank on randomized evaluations measuring the effects of business regulation reforms in Africa and in Asia. Victor holds a Master Degree in Economics from the Paris School of Economics and has working experience in several countries in Africa and Asia.