Data quality was checked using spot-checks and back-checks.
The IE originally intended to look at child diarrhea prevalence as a secondary outcome. However due to the shift in the focus of the evaluation and the different methodology of collecting the child health roster in each survey round, diarrhea was dropped as an outcome of the study. At baseline households were asked about each child’s health status separately, while at endline households were asked about the aggregate incidence of diarrhea and other symptoms for all children under 5 years old in the household. This prevents us from supplementing the definition of child diarrhea using multiple symptoms, like blood or mucus in the stool. This is important as some primary caregivers are unable to accurately diagnose diarrhea in children. Additionally, our statistical power reduces by almost two thirds, because the number individual data points goes down from 2246 at baseline to 831 at endline.
Additional information on data appraisal is provided in the Study Design annex of the survey report provided under the resources tab.
