91

Drawing on a case study (the randomized evaluation of the impact of the SKY microinsurance program in Cambodia), this paper examines the practicalities—and compromises—of combining, on the one hand, the constraints of academic rigor with operational demands and, on the other, different methods of knowledge production (quantitative/qualitative). In order to highlight the many challenges presented by these two types of combinations in relation to randomized controlled methods, we examine the interplay between actors. Compared with other quantitative methods, the narrowness of the research questions, the stringent constraints of the study protocol, and the essential collaboration of the field operators in implementing the study make this dual combination not only inevitable, but also particularly complex.