A two-stage procedure, according to which the population is divided into subpopulations (subpopulations), called strata, and then from random subpopulations, randomly sampled elements are taken. In individual strata, the draw can be simple or dependent. The purpose of dividing the studied population into subpopulations, or in other words the purpose of stratification, is to identify in a diversified population possible homogeneous groups of individuals. Thanks to this treatment, each of these groups will have the right representation in the sample. This is especially important in highly heterogeneous populations. Individual strata should be strongly differentiated between them and homogenous inside. Strat are made so that the obtained strata are disjointed and that each unit of the population belong to one and only one stratum. Stratification features, based on which the population isdivided into strata, are related to the variable under study (most often these are demographic, social, economic,etc.). One of the most important issues when applying the stratified sampling scheme is the so-called sample allocation, i.e. distribution of sample elements in individual strata. The researcher faces the problem of how many units from each stratum and how he should be drawn to the sample. We can distinguish the following solutions: proportional allocation, Neyman (optimal) allocation, uniform allocation.

• Neyman allocation
• Proportional allocation
• Sample allocation
• Uniform allocation
• Stratifying variables
• Sampling
• Minimal sample size

• Stratum
• Stratification