Cluster Sampling and EPI Methods Cluster sampling Cluster

Cluster Sampling and EPI Methods

Cluster sampling Cluster: A group of sampling units close to each other i. e. crowding together in the same area or neighborhood § Used when "natural" but relatively homogeneous groupings are evident in a statistical population. § In this technique, the total population is divided into these groups (or clusters) and a simple random sample of the groups is selected. 2

Cluster sampling 1. The population within a cluster should ideally be as heterogeneous as possible. 2. There should be homogeneity between cluster means. 3. Each cluster should be a small scale representation of the total population. 4. The clusters should be mutually exclusive and collectively exhaustive.

Cluster sampling Household Village So all the sampling units within the cluster is selected This is particularly useful in population based studies 4

Cluster sampling Block 1 Block 2 Block 3 Block 5 Block 4

EPI cluster survey design § Select a central location in the village or town (e. g. market, church, tree) § Randomly select a direction to walk in 6

EPI cluster survey design § Walk to the edge of the village in the selected direction and count the number of houses § Select a random number between 1 and the total number of houses and return to this house § This is the first household § The second household should be the one whose front door is closest to the first 7

Cluster Sampling Pros Cons § Most economical way of sampling § Sampling frame may exist at cluster level (e. g. districts/ boroughs/ postcode) § Efficient (less time to list and implement) § May not reflect diversity of community § Other elements in cluster may share similar characteristics and therefore provides less information than simple random sampling § More variation around the estimate (bigger confidence intervals) 8

Example: GPS use § What about an area which isn’t mapped? à E. g. refugee camp § Use of GPS or drones to map an area

Cluster vs. Stratified Sampling • Cluster sampling is often confused with stratified sampling, because they both involve "groups". • In stratified sampling, the population is split into groups (strata) based on some characteristic. • In cluster sampling, the population is already broken into groups (clusters), and each cluster represents the population. • Cluster sampling is appropriate when the clusters are approximately the same size.

Multi-stage sampling In a two-stage sampling design, a sample of is primary units is selected and then a sample of secondary units is selected within each primary unit Classes Students 11

Multi-stage sampling Two-stage cluster sampling aims at minimizing survey costs and at the same time controlling the uncertainty related to estimates of interest. It is used frequently in health and social sciences.

Summary

Summary • Often a non-random selection of basic sampling frame (city, organization etc. ) • Fit between sampling frame and research goals must be evaluated • Sampling frame as a concept is relevant to all kinds of research (including nonprobability) • Nonprobability sampling means you cannot generalize beyond the sample • Probability sampling means you can generalize to the population defined by the sampling frame