2018/09/18 by Jatteau, Arthur
#Randomization #geometrical data analysis #hierarchical classification #multiple correspondence analysis #network analysis #prosopography #sociology of economists
paper · doi:10.12759/hsr.43.2018.3.94-119
Randomized controlled trials (RCTs) are a method to assess impact that has become increasingly popular over the last fifteen years, particularly as a result of the work done by Esther Duflo and her Poverty Action Lab (J-PAL), an organization devoted to the promotion of randomization. This article aims to explore and understand this success by using an in-depth sociographical study of the J-PAL and a network analysis of economists who use RCT. J-PAL appears to be a concentration of educational and academic capital that give great legitimacy to the RCT method. The network is controlled by certain leaders who are able to diffuse the J-PAL approach to RCTs. Furthermore, this article argues that it is necessary to go beyond the intrinsic quality of this method to explain how it became so popular.