Study population, sample size and sampling

The study population comprised mothers with children aged 6–36 months who were randomly selected from five intervention districts (n 412) and one comparison district (n 300). The sample was estimated based on the primary outcome of child linear growth, which yielded the maximum sample size⁽¹⁰⁾. The sample size was thus calculated providing a statistical power of 80 % at 95 % confidence level to detect a reduction of 10-percentage point difference in the prevalence of stunting between the study groups. The sample size was estimated using OpenEpi software for epidemiologic statistics version 3.01.

In each selected cluster, a sample size of sixteen households with eligible mothers/caretakers and children was systematically selected. A complete list of all households was compiled, and systematic random sampling was used in selecting the households. All the households in each cluster were serially numbered. To get the sampling interval, the total number of households in a cluster was divided by the sample size of 16. The first household was then randomly selected by picking any number within the sample interval. Subsequent selections were made by adding the sampling interval to the selected number in order to locate the next household to visit. If the selected household does not have a target respondent, then the next household was selected using the systematic sampling procedure. This process continued until the required sample size was obtained. Only one eligible participant was selected from each household for household interview, using simple random sampling.