New Delhi, Delhi, India: Penn Social Norms Group (PennSoNG), a research group at the University of Pennsylvania, has conducted a formative research study to understand the social determinants of toilet construction and use in India. More specifically, the study focuses on social norms and networks in 14 districts in the states of Bihar and Tamil Nadu. This research was conducted in two waves (fall 2017 and spring 2018), and it included two large surveys (n=3370 and n=5052) and rounds of focus group discussions.

Our results are in line with recent government data about toilet use conditional on ownership of functional toilets. We also observed an increase in ownership and use across the two waves of our study. Though previous studies have highlighted the importance of gender, caste, and purity considerations as barriers to improved sanitation behavior, we did not find evidence to support those results. We find that use across genders is similar (and high), and that purity and pollution considerations are not meaningfully factored into the decision of where and whether to build a toilet. Although specific social groups primarily empty toilets, when asked about the appropriateness of others emptying a toilet, our data did not find clear evidence that higher castes’ concerns about “pollution” is a barrier to toilet construction and use.

Our research suggests that the primary determinant of changes in sanitation behavior and the sustainability of such changes are the beliefs that people have about others’ toilet use in their social network. Surprisingly, we do not find any evidence that social approval or disapproval of open defecation or toilet use predict the decision to use a toilet. Our conclusion is that the sustainability of the positive changes we have observed in our data is linked to the existence of a descriptive norm of toilet use. By this, we mean that sanitation behavior is in large part determined by beliefs about relevant others’ behavior. This finding implies that as individuals’ expectations about others’ use rises, their likelihood to use a toilet rises as well.

To continue towards the goal of improved sanitation across India, sustaining the positive changes that have been made will be a critical step. Leveraging our critical observation that expectations of others’ behavior are an important determinant of toilet use, any intervention should aim at broadly disseminating information about positive behavior. As changes in expectations lead to changes in behavior, future expectations will also grow. This predicted multiplier effect should ensure stable and lasting changes.