Stratified sampling is a probability sampling technique wherein the researcher divides the entire population into different subgroups or strata, then randomly selects the final subjects proportionally from the different strata.
It is important to note that the strata must be non-overlapping. Having overlapping subgroups will grant some individuals higher chances of being selected as subject. This completely negates the concept of stratified sampling as a type of probability sampling.
Equally important is the fact that the researcher must use simple probability sampling within the different strata.
The most common strata used in stratified random sampling are age, gender, socioeconomic status, religion, nationality and educational attainment.
Because this technique has high statistical precision, it also means that it requires a small sample size which can save a lot of time, money and effort of the researchers.
The sample size of each stratum in this technique is proportionate to the population size of the stratum when viewed against the entire population. This means that the each stratum has the same sampling fraction.
For example, you have 3 strata with 100, 200 and 300 population sizes respectively. And the researcher chose a sampling fraction of ½. Then, the researcher must randomly sample 50, 100 and 150 subjects from each stratum respectively.
|Final Sample Size||50||100||150|
The important thing to remember in this technique is to use the same sampling fraction for each stratum regardless of the differences in population size of the strata. It is much like assembling a smaller population that is specific to the relative proportions of the subgroups within the population.
The only difference between proportionate and disproportionate stratified random sampling is their sampling fractions. With disproportionate sampling, the different strata have different sampling fractions.
The precision of this design is highly dependent on the sampling fraction allocation of the researcher. If the researcher commits mistakes in allotting sampling fractions, a stratum may either be overrepresented or underrepresented which will result in skewed results.