Spotlight on EGAP Learning Days  at African School of Economics

As part of the event, the ASE community welcomed on Monday, June 11st, 2019 Dr. Nahomi Ichino from the University of Michigan, Ms. Tara Slough and Mr. Salif Jaiteh from Columbia University, and from Columbia University.

Dr. Nahomi Ichino, who is an assistant professor in the Department of Political Science at the University of Michigan, introduced the Counterfactual Causal Inference in order to answer some core causal questions. She  explained the counterfactual approach, which views that a causal effect is the difference between a potential outcome under treatment and a potential outcome under control. The fundamental problem of causal inference is the fact that one cannot simultaneously observe the potential outcome under treatment and the potential outcome under control for a subject. Notwithstanding, she pointed out that randomization can allow us to estimate the average causal effect under three core assumptions: Random assignment of subjects to treatment, Non-inference, and Excludability. Random assignment of subjects to treatment is a powerful tool to estimate the average difference between the two potential outcomes.

Ms. Tara Slough is a Ph.D. Candidate in Political Science at Columbia University. She shared her major finding on the study titled: Bureaucrats driving inequalities: evidence from Columbia.  As part of that study, she ran an experiment aiming at untangling how bureaucrats’ actions condition access to social programs. She pointed out the ethical issues related to such an experiment, which are related to fake requests for service, impossibility of asking for bureaucrats’ consent, and the innovation required through partnerships with government organizations overseeing the audited programs. Her results suggest important biases  against the lower class, relative to lower middle class petitioners in information provision. Analyzing the mechanisms, she argues biases are driven by the threat of citizen complaints: According to her conclusion, substantial preparations are needed in order to execute a successful experiment, it is important to study the context and work with people that are familiar with it, and ethical consideration should be of paramount importance in any intervention in the world.   

As a Ph.D. Student in Political Science at Columbia University, Mr. Salif Jaiteh presented the main elements of a research design and enabled participants to develop their research design. Besides the introductory section, he substantially developed the components of the other sections which are Treatment; Outcome; Identification strategy; Sample, Data, & Implementation Strategies; Power; and Analysis & Threats. He observed that a clear and good research design is critical in answering research questions convincingly.