What are the specific methods of random sampling in social investigation methods?

1, simple random sampling. The principle of simple random sampling is the same as dice or lottery. Therefore, in this method, the chances of any single unit being drawn are completely equal. Simple random sampling needs to number each sample and then use a random number table. When sampling, you can randomly determine a starting number and then read it in any direction until enough samples are selected.

2. Systematic sampling. Systematic sampling is also called equidistant sampling. When sampling, researchers can randomly select a sample as the initial sample, and then take it at regular intervals. However, it should be noted that the samples must be randomly arranged. Otherwise, once the interval conforms to the regularity of sample arrangement, the extracted samples are not random. This method is simple and labor-saving.

3. Stratified sampling. When the attributes of sample objects are quite different, the objects can be divided into several categories in advance according to certain attributes, which are called "layers". Then randomly select samples in each layer. When the sample attributes are too different, multi-layer sampling can be performed. This method can make the large-scale investigation simpler, and it is also convenient to compare different groups in the sample, and the accuracy of the investigation will be improved.

4. Multistage sampling. That is, when the scale of the survey and the number of samples are too large, the samples can be divided into several levels to extract the objects, which makes the large-scale survey easy to implement. It should be noted that the classification of multi-level surveys will generally not exceed three levels because each level will produce errors, and the more levels, the greater the errors.