According to the principle of randomness, that is, to ensure that every object in the population has a known non-zero probability to be selected as the research object, to ensure the representativeness of the sample. Random sampling method is that every part of the survey object has the same possibility of being drawn, and it is a sampling survey conducted in full accordance with the principle of equal opportunity, which is called an "equal probability".
Its greatest advantage is that when inferring a population from sample data, it can objectively measure the reliability of the inferred value in a probabilistic way, thus making this inference based on science. Because of this, random sampling is widely used in social investigation and social research.
Commonly used random sampling methods mainly include pure random sampling, stratified sampling, systematic sampling, cluster sampling and multi-stage sampling. The main advantage of random sampling is that because each sample unit is randomly selected, according to probability theory, not only the overall parameters can be estimated through sample statistics, but also the sampling error can be calculated.
Therefore, the reliability of inferring the overall target variable can be obtained. However, random sampling is more complicated and requires high professional skills of investigators, so sampling frames are needed in the investigation. However, the cost of constructing and maintaining a high-quality sampling framework is very high, and the sampling units may be very scattered, and the sampling units cannot be easily changed, which increases the investigation cost.
Advantages:
1. Simple random sampling method is simple and intuitive, which is the most basic organizational form and the cornerstone of random sampling theory. For example, shopping in daily life, lottery subscription when a commodity is in short supply, etc. Is a simple prototype of pure random sampling.
2. Simple random sampling is the basis of other sampling methods, that is, various organizational forms of random sampling are derived from simple random sampling. For example, cluster sampling means that a group or group composed of some general units with the same properties under a certain sign is regarded as an individual, and then simple random sampling is carried out, in which the grouping work is not random, but random sampling is only carried out on the premise of grouping.
3. Simple random sampling is a comparison standard to measure the effect of various sampling methods. Using sample indicators to estimate and infer the corresponding overall indicators, with different organizational forms, the effectiveness of the estimation results of the same survey indicators is also different.