Social Science Comes to the Village : Assessing Behavioral Research in Developing Countries

Survey and lab-in-the-field experiments have become popular methods of social science research in the developing world, yet we know little about how subjects understand these projects. We explore this issue with a two-stage research design: one coauthor used a reputable survey firm to conduct vignette experiments and behavioral games in rural India, following which other coauthors claiming to be unconnected with the original project conducted qualitative interviews. Our fieldwork shows that 1) there were major and hard-to-avoid departures from random sampling, reflecting actions by both the firm and the subjects, 2) there was widespread lack of informed consent among subjects, and 3) there was considerable confusion about the purpose and structure of the experiments. These findings have implications for the ethics, validity, and interpretation of a broad set of social scientific interventions routinely carried out in the Global South