The sample is a part of the individual observed or investigated, and the whole is the whole of the research object. The total number of elements extracted from the tested population and the number of individuals in the sample are called sample size.
Some individuals actually observed or investigated in the study are called samples, and all the research objects are called groups. In order to make the sample correctly reflect the overall situation, the overall situation should be clearly defined; All observation units in the population must be homogeneous; In the process of sampling, the principle of randomization must be observed; There should be enough sample observation units.
Sample generation mode
1. Simple random sampling: This is the most random method, and every member has a chance to be selected. There is no grouping and stratification in advance, and the samples are obtained randomly. It is more suitable for the situation that the number of the overall target groups under investigation is not too large and the differences between individuals are small.
2. Systematic sampling: In this type of sampling, the first individual is randomly selected, and other individuals are selected at fixed "sampling intervals".
This method is often used in household surveys, such as randomly selecting the first household in a residential area to visit, and then visiting another household every 10 according to the right-handed principle. How big this interval is can be determined according to the total number of groups and the number of people to go.
3. Cluster sampling: In cluster sampling, subgroups of population are used as sampling units, not individuals. The population is divided into subgroups called groups, and a complete group is randomly selected as the sampling sample. Using this sampling method, it is generally required to be as homogeneous as possible between groups and as heterogeneous as possible within groups. Area sampling, a commonly used city, is an embodiment of cluster sampling.
4. Stratified sampling: In this type of sampling, the population is divided into different groups according to different characteristics, such as gender and age. Stratified sampling is characterized by being divided into several exclusive and exhaustive subgroups. Then select samples from these subgroups:, and then select samples from these subgroups.