In: Statistics and Probability
Please make sure to display your thought process? It is imperative to be able to follow how the answer was deduced. Please be as thorough as possible. Please address all parts of this question.
A team of medical researchers wants to estimate the malaria infection rate in a difficult-to-access jungle region of [No Name Country]. There are 80 villages in the studied part of the jungle, which can be assumed to be rather similar to each other. Each village has at least 300 inhabitants. We want to test 200 individuals for malaria. Propose a cost-efficient sampling method.
In this case, a cost-efficient sampling method is "One - stage Cluster Sampling".
Justification: Cluster Sampling is a sampling method used when grouping is done in a mutually homogenous but internally heterogenous setup. The total population is divided into groups called clusters and a simple random sample from the clusters is selected and every member of the cluster is sampled.
The prime motivation for cluster sampling is to reduce the total cost given a certain accuracy.
In the given problem, there are 80 villages having 300 inhabitants each, then, the total number of inhabitants can be considered as the total population. Clusters can be formed with respect to each village. Suppose we form clusters from "N" villages (N<80) containing 200 inhabitants each and each of these 200 inhabitants are checked whether they are having malaria or not. Since the 80 villages can be rather similar to each other, as stated in the question, also, inhabitants differ among each other in terms of the presence of malaria or not, this corresponds to a setup of cluster sampling where there is homogeneity between the groups whereas heterogeneity is present within the members of the group.