The selection of a proper sample is an obvious prerequisite to a sample survey. A sample is defined to be a collection of units which is some part of a larger population and which is specifically selected to represent the whole population. Four aspects of this definition are of particular importance: first, what are the units which comprise the sample; second, what is the population which the sample seeks to represent; third, how large should the sample be; and fourth, how is the sample to be selected?
While many books have been devoted purely to survey sampling, it is the view of the authors that sampling is only one part of the overall survey design process. A perfect sample is irrelevant, if the questionnaire used to obtain the information is the source of considerable measurement bias. A balanced approach needs to be taken between all apsects of the survey design to obtain a valid result from the survey.
This chapter on Sampling covers the following topics:
4.1 TARGET POPULATION DEFINITION
4.2 SAMPLING UNITS
4.3 SAMPLING FRAME
4.4 SAMPLING METHODS
4.4.1 Simple Random Sampling
4.4.2 Stratified Random Sampling
4.4.3 Variable Fraction Stratified Random Sampling
4.4.4 Multi-Stage Sampling
4.4.5 Cluster Sampling
4.4.6 Systematic Sampling
4.4.7 Non-Random Sampling Methods
4.5 SAMPLING ERROR AND SAMPLING BIAS
4.6 SAMPLE SIZE CALCULATIONS
4.6.1 Sample Sizes for Population Parameter Estimates
4.6.2 Sample Sizes for Hypothesis Testing
4.7 VARIANCE ESTIMATION TECHNIQUES
4.7.1 Variability in Simple Random Samples
4.7.2 Design Effects
4.7.3 Replicate Sampling
4.8 DRAWING THE SAMPLE