Be careful what you ask for: Social desirability bias in self-reported data in India
Co-authored with Sangita Vyas.
Abstract: When assessing policy impact, how can one distinguish between genuine behavior change and reporting bias? We study this question in the context of a large-scale sanitation program in India. The association between child height and open defecation is attenuated after the sanitation program compared to before, which is consistent with greater under-reporting of open defecation after the program. Using a biased response model, our results imply that about half of rural Indian households still defecate in the open, versus 27 percent who self-report doing so. If so, widespread open defecation in rural India continues to harm children’s health and survival.