The sampling was designed to generate representative findings at the camp level, allowing WFP and UNHCR to use the data to make conclusions that could be generalized for the refugee population in each camp. Hence, relevant statistical parameters including a 95 percent confidence level and a margin of error of +-4 percent were applied to calculate the sample size.
The sample size was calculated based on the total number of refugee groups (as the proxy to the total number of households) registered in UNHCR ProGres database as of November 2020.
Systematic random sampling with a fixed periodic interval was conducted in each camps to select the households.
An alternative list of households was prepared before going to the field. The enumerators visited the households on this list if the interview with the originally sampled households could not be undertaken with justified reasons.
At the household level, in principle, either the household head or spouse (in the absence of the head) or any adult member of the household (in the absence of both the household head and the spouse) were interviewed. Enumerator did not exclude the child-headed households if sampled.
In total 2,501 households were randomly assessed, with 2,499 consenting to being interviewed. In the analysis, post-stratification weighting was applied to align the sample with the known population proportion for each camp and the sex of the head of household.
