Before we discuss this method of sampling, we have to define two different types of population:
- Homogeneous population: sampling units are all of the same kind and can reasonably
be dealt with in one group.
- Heterogeneous population: sampling units are different from one another and should be placed in several separate groups.
In the sampling methods already discussed we have assumed that the populations are homogeneous, so that the items chosen in the sample are typical of the whole population. However, in business and social surveys the populations concerned are very often heterogeneous.
For example, in the bank survey the bank customers may have interests in different areas of banking activities, or in a social survey the members of the population may come from different social classes and so will hold different opinions on many subjects. If this feature of the population is ignored, the sample chosen will not give a true cross section of the population.This problem is overcome by using stratified sampling, another example of probability sampling.
The population is divided into groups or strata, according to the characteristics of the different sections of the population, and a simple random sample is taken from each stratum. The sum of these samples is equal to the size of the sample required, and the individual sizes are proportional to the sizes of the strata in the population. An example of this would be the division of the population of London into various socio-economic strata.
- Advantages – the advantage of this method is that the results from such a sample willnot be distorted or biased by undue emphasis on extreme observations.
- Disadvantages – the main disadvantage is the difficulty of defining the strata. This method can also be time-consuming, expensive and complicated to analyse.
0 Comments