Indirect Sampling for Hard-to-Reach PopulationsView Presentation
*Pierre Lavallée, Statistics Canada
Keywords: generalised weight share method, network sampling, adaptive cluster sampling, multiple frames
In survey sampling, some populations happen to be hard to survey. Their relative rareness and the absence of a suitable sampling frame are two main reasons for this. When no sampling frame is available for the desired target population, one can then choose a sampling frame that is indirectly related to this target population. We can then speak of two populations A and B that are related to one another. We wish to produce an estimate for B, by selecting a sample from A and using the existing links between the two populations. This is referred to as indirect sampling. Producing estimates in the context of indirect sampling can be difficult to achieve if the links between A and B are not one-to-one. A solution for this is to use the generalised weight share method (GWSM).
Using indirect sampling as described above is one way for surveying hard-to-reach populations. Other approaches exist, such as network sampling, adaptive cluster sampling, snowball sampling, and the use of multiple frames. Actually, all these approaches can be put into the context of indirect sampling. For example, in the case of adaptive cluster sampling, the population of interest B is one of clusters, while the sampling frame A contains the elements of these clusters from B. One can use the theory and developments surrounding indirect sampling and the GWSM to obtain a unified mathematical framework for the above approaches.
After an overview of indirect sampling and its readily application to hard-to-reach populations, the paper will describe indirect sampling and the GWSM in the context of network sampling, adaptive cluster sampling, snowball sampling and multiple frames.