A simulation study of sampling in difficult settings: Statistical superiority of a little-used method

What is it about?

Authorities, policy makers, and researchers often want to do population surveys. There are many reasons for the surveys; for example, to see if vaccination programs have been successful by estimating the proportion of people (or children) vaccinated. When there is little information available on the population, especially in refugee camps or in war zones, getting a valid estimate is not easy. There are several different methods of choosing a sample for the survey. We did a large computer simulation to find out which method is ‘best.’ We found that the best is one that has very rarely been applied. It uses satellite images of the area being surveyed. People doing surveys should consider using it.

Featured Image

Why is it important?

The best method is practical and can reduce the resources needed to do surveys.