Cluster sampling to assess immunization coverage: a review of experience with a simplified sampling method.

Vaccination coverage survey - Publication abstract: A simplified cluster sampling method, involving the random selection of 210 children in 30 clusters of 7 each, has been used by the Expanded Programme on Immunization to estimate immunization coverage. This paper analysis the performance of the method in 60 actual surveys and 1500 computer simulated surveys. Although the method gives a proportion of results with confidence limits exceeding the desired maximum of 10 absolute percentage points, it is concluded that is performs satisfactorily.