Improving the quality of development assistance

While randomized experiments can be valuable tools in evaluating aid effectiveness, research designs limit the role of qualitative methods to ‘field visits’ or description of contexts. This article suggests expanding the role of qualitative methods and highlights their advantages and limitations relative to survey methods. It reviews a range of qualitative methods and suggests that life histories are compatible with the internal and external validity criteria of randomized experiments. It illustrates this with a case study of their proposed use in an evaluation of the promotion of Jatropha curcas, a second-generation biofuel, in Malawi.