2014 Training Course - Impact Evaluation and Analysis of Development Interventions III

Event Date
IFPRI Dakar, Senegal


Over the course of the past ten years, the number of impact evaluations related to development interventions has increased substantially. In fact, under the leadership of leading academic institutions in the US and Europe, empirical studies based on robust comparisons of treatment and control groups have received considerable financial support from the donor community and led to dozens of publications in leading academic journals in the areas of agriculture, education and health.
At the heart of this movement has been a widely-shared idea that impact evaluations can contribute to development effectiveness by helping to determine “what works and what does not”, and that this research can be used to influence policy. Yet, despite an estimated 250 on-going or completed impact evaluation studies in Africa conducted by NGOs, the World Bank and universities, only a handful of these involve researchers from African universities. Why?  And how can this be changed?
This course intends to provide African-based researchers with knowledge of the latest developments regarding both technical and topical aspects of impact evaluation. The course is also meant to help build a high-level network of local impact evaluation practitioners and partners of within the region.


This course is a continuation of the two one-week courses held in Dakar, Senegal in 2013 and is by invitation only. This portion of the course will focus primarily on a set of statistical tools and research designs geared to conducting high-quality empirical research that are directly policy relevant. It will focus on methods for estimating causal effects, and aimed to develop a thorough understanding of causal identification.
The course will be organized as follows:
June 2nd - 3rd, 2014 (Michael Anderson)
  • Brief introduction
  • Introduction to causality and randomized trials
  • Regression: What does it do?
  • Real-world evidence on selection bias
  • Selection on observables research designs
  • Panel data and differences-in-differences
  • Instrumental Variables methods
June 4th, 2014
  • Morning session: Referee report and discussant workshop with structured lecture and discussion.
  • Afternoon session: Reading, commenting and compiling discussant presentation on fellow participant’s paper.
June 5th, 2014
  • All day: Presenter and discussant sessions.
June 6th, 2014
  • Morning session: Continuation of presenter and discussant sessions.
  • Afternoon session: Small group work to develop proposal or joint paper with fellow AGRODEP members.


Tanguy Bernard is a Research Fellow within the Markets, Trade and Institution Division of IFPRI, based in Senegal. Over the recent years, his research has mostly focused on producers’ organizations, their existence, their membership, their activities, and their performance in linking African smallholders to input and output markets. His works relies heavily on primary data collection and experimental tools, with a geographical focus on Senegal, Burkina Faso and Ethiopia. Prior to his current position, Tanguy Bernard was a research officer at the Agence Française de Développement, prior to which he was a post-doctoral fellow at IFPRI, based in Ethiopia.


Susan Godlonton is an Associate Research Fellow in the Markets, Trade and Institutions Division of IFPRI. She has a Ph.D. in Economics from the University of Michigan and a Masters in Economics from the University of Cape Town. Previously she lectured in the Economics department at the University of Cape Town. Her research focuses on job-seeker and employee decision-making and behavior in labor markets in developing countries mostly in urban areas. She also studies decision-making and information sharing in the context of health seeking behaviors with a particular focus on HIV prevention. She has conducted research in Malawi and South Africa, and is also currently working in Mozambique.


Michael Anderson is an Associate Professor of Agricultural and Resource Economic at the University of California, Berkeley. His research interests focus on the areas of health economics, environmental economics, and applied econometrics. Recent work includes papers on the impacts of long-term exposure to air pollution on mortality, the benefits of transit systems on reducing traffic congestion, optimal policies in the context of traffic safety, and the effects of health insurance on utilization of health care. His work appears in journals such as the American Economic Review, the Review of Economic Studies, the Journal of the American Statistical Association, the Review of Economics