What are the commonly used probability sampling methods? What do they mean?

Simple random sampling with substitution randomly selects a sample unit from the population, records the observation results, puts them back into the population, then extracts the second one, and so on until it is full of n units.

Units may be drawn repeatedly, which may easily lead to overlapping information and affect the estimation efficiency, and are rarely used.

2. Don't put it back into simple random sampling.

Randomly extract units from a population containing n units, one by one, and extract the next unit with equal probability from all non-sampled units until n units are extracted.

Each unit can only be drawn once at most, which will not provide overlapping information because of repeated drawing of sample units, and has lower sampling error than putting back sampling.

3. Stratified sampling

Firstly, the population is divided into different layers according to certain rules, and then samples are randomly selected at different layers independently, so the samples obtained are called stratified samples. If the sampling in each layer is simple random sampling, it is called stratified random sampling.

4. Systematic sampling

Systematic sampling refers to arranging all units in a group in a certain order, randomly extracting an initial unit within a specified range, and then extracting other sample units according to pre-specified rules. The simplest systematic sampling is equidistant sampling.

5. Nested sampling method

Cluster sampling is to divide all the basic units in a group into non-overlapping groups according to certain rules. When sampling, select groups directly, investigate all basic units of the selected groups, and do not investigate the unselected groups.

Extended data

Probabilistic sampling has the following advantages: investigators can obtain information of people of different ages and levels; The sampling error can be estimated; The survey results can be used to infer the population. ?

For example, in a survey using the probability sampling method, if 5% of the respondents give specific answers, then the investigators can combine this percentage with the sampling error to summarize the overall situation.

On the other hand, probability sampling also has some shortcomings: in most cases, the cost of probability sampling of the same scale is higher than that of non-probability sampling; Probabilistic sampling takes more time to plan and implement than non-probabilistic sampling; The sampling plan implementation procedures that must be followed will greatly increase the time for data collection.

Baidu encyclopedia-probability sampling