The sampling process involves a number of steps in which decisions must be made with respect to what samples will be obtained and how:

1. Define the population. The target population to be sampled must be defined in terms of the purpose of the study, units, elements, geographic area and time period. Note that it is the population containing the phenomena under study being defined here, not the sample...as yet.
2. Define the sampling frame. How are the elements of the population to be described for sampling? Ideally it would be the entire target population, but most likely it will only be the elements of the population that are accessible (ie. a subset).
3. Specify the sampling unit. Identify the units by which the sample will be obtained and measured. For example, this may be streets, households, paddocks, local government areas, etc.
4. Determine the sampling method. This involves specifying a specific sampling method to identify the observations from which the sample data may be obtained.
5. Determine the size of the sample. What number of sampling units/observations will be selected? It should be not too small where the population is not adequately represented, nor too large which may result in an inexpensive and inappropriate exercise.
6. Specify the sampling plan and method of sample data collection. This involves the operational considerations in obtaining the samples.

Where to from here?

In order to appreciate the issues involved in making the sampling decisions, we will need to be aware of the various sampling methods and how the data can be obtained. We will examine the types of data sources next...

No easy formula

There is no easy formula to say how it should be done for each step. The decisions to be made are entirely dependent on the application needs.

The sampling plan must ultimately be: suitable, applicable, feasible, economical, reasonable, "do-able", etc.