Project Sammaan is a randomized evaluation designed to answer two primary research questions. Random assignment of slums and households to various interventions will allow us to attribute any subsequent changes in health-related outcomes to Project Sammaan’s interventions, and not to any other observable or unobservable factors.
At the community-level – What is the impact of the popular community-led total sanitation (CLTS) approach on open defecation by slum residents?
- Project Sammaan, in collaboration with Feedback Foundation, is testing an urban version of CLTS. Randomly assigning communities to receive CLTS allows us to isolate the effect of CLTS on various health-related outcomes. Project communities will be either be allocated to “treatment” (i.e. site receives the CLTS intervention) or “control” (i.e. site does not receive the CLTS intervention).
At the household and individual-level – How are habits formed and sustained?
- Time discounted voucher – This intervention tests the repetition of subsidized use of Sammaan facilities during peak open defecation hours. Individuals are offered zero price discounts for forty-five days, and their usage patterns are observed through the discount period and after. This intervention has three groups: households who do not receive any discounts, households who may use the facility for free 7 days a week anytime during operational hours, and households who may use the facility for free on certain days of the week during peak hours only.
- Trigger-reward – This intervention tests the trigger-reward system as a means to create a habit loop. Households are provided with a Sammaan calendar that ‘triggers’ them to use a Sammaan facility instead of defecating in the open. They are then ‘rewarded’ for using Sammaan by access to a hygiene faucet in the toilet stall. At the household level, there are two treatment groups: households that receive the calendar, and households that do not receive the calendar. At the facility level, there are four groups: facilities with hygiene faucets on the male side, facilities with hygiene faucets on the female side, facilities with hygeine faucets on both sides, and facilities without hygiene faucets.
Keeping with the strategy of random assignment, households are randomly offered the discount vouchers and calendars via an in-field lottery.
The sample for this study is households without access to a private or shared toilet located within a 160-meter radius (5-minute walking distance) of the facility. Detailed baseline and endline data are collected from these households on demographics, and sanitation behaviours and preferences. The baseline is conducted prior to toilet construction and the endline is conducted 1 year after the opening of the Sammaan facility. This survey data is supplemented by real-time Sammaan facility usage data which is collected by each facility caretaker via a tablet-based application. One feature of this usage data is that for each individual included in the baseline and endline surveys, the project team will be able to link the survey data with that person’s actual usage of the Project Sammaan facility. Using this incredibly rich set of data, the research team will be able to compare intervention groups at the site and household level to ascertain whether these interventions have an impact on community toilet uptake and usage, and the magnitude of this impact.