Sampling Methods PDF

Methods of allocation of sample size Equal Allocation where the sample is allocated equally over strata Proportional allocation where the sample is allocated over strata proportional to stratum size Optimum allocation where allocation over strata depends on stratum size and degree of variability wit...

Methods of allocation of sample size Equal Allocation where the sample is allocated equally over strata Proportional allocation where the sample is allocated over strata proportional to stratum size Optimum allocation where allocation over strata depends on stratum size and degree of variability within strata Advantages The stratified sample ensures the appropriate representation of the various population groups in the smple The degree of precision of the results of stratified samples is always higher than those of any other probability sample of the same size Disadvantage Difficulty in constructing the frame when stratum sizes are large (sampling frame has to be prepared separately for each stratum Cluster Sampling Used if no sampling frame is available A cluster ➔ is a group of sampling units close to each others i.e. crowding together in the same area or neighborhood. Example…. Villages, districts, institutions. It is applied by dividing the population into a number of clusters then a certain number of clusters are selected by SRS. For each selected cluster, either all the elements are included in the sample or a sample of elements is selected. Cluster Sampling • Sometimes if clusters are big and of equal size the sample is divided equally over selected clusters. If clusters are unequal the sample is divided over clusters with a probability proportional to size (PPS) Cluster Sampling The more clusters selected and the fewer individuals in each cluster, the more representative the sample The more internally homogenous the clusters, the fewer elements needed per cluster Sampling error is greater than in simple random sampling Cluster Sampling Reduces travel by interviewing many respondents within the same geographic area Greatly reduces survey field cost Easy construction of frame Lower degree of precision compared to other types of samples Multi-stage Random Sample The idea of this sample is to proceed in the selection of sample in stages and for each stage the construction of the frame for that particular stage is always an easy task. The multi-stage random methods are rarely used alone, usually a combination of different sampling methods are used such as multistage and stratified random samples. Example Suppose that we want to select a representative sample from the Egyptian population as a whole ⚫ 1st stage: select governorates ⚫ 2nd stage: select districts within selected governorates ⚫ 3rd stage: select villages within selected districts Multi-stage Random Sample No difficulty in constructing the frame Particularly useful when we want to select a sample representing the whole nation (National household survey) Reduces cost, minimizes travel Degree of precision is usually low Selecting a sampling method Population to be studied ⚫ Size/geographical distribution ⚫ Heterogeneity with respect to variable Availability of list of sampling units Level of precision required Resources available