Simple random sampling method (1) lottery method
Generally speaking, the lottery method is to number n individuals in the population, write the numbers on the digital labels, put the digital labels into a container, and after stirring evenly, extract one digital label from it every time and extract it for n times continuously to get a sample with a capacity of n. 。
The lottery method is simple and feasible, and is suitable for a few individuals in the crowd. When there are multiple individuals in the group, it is difficult to "evenly mix the group", which is probably due to the poor representativeness of the samples produced by the lottery method. )
(2) Random number method
Another commonly used method in random sampling is random number method, that is, random number table, random number dice or computer-generated random number are used for sampling.
Edit paragraph 2. Stratified random sampling (stratified random sampling) is mainly characterized by stratified and proportional sampling, which is mainly used for individuals in people with obvious differences. * * * Similarity: the probability of each individual being drawn is equal to n/m.
1. Definition
Generally speaking, when sampling, the population is divided into disjoint layers, and then a certain number of individuals are independently extracted from each layer according to a certain proportion, and the individuals extracted from each layer are combined as samples. This sampling method is stratified sampling.
Systematic sampling When there are a large number of individuals in a group, it is troublesome to adopt simple random sampling. At this time, the population can be divided into several balanced parts, and then an individual can be extracted from each part according to predetermined rules to get the required samples. This kind of sampling is called systematic sampling.
Steps:
Generally speaking, if you want to extract a sample with a capacity of n from a population with a capacity of n, you can carry out systematic sampling according to the following steps:
(1) First number the n individuals in the population. Sometimes you can directly use the numbers brought by individuals, such as student number, admission ticket number, house number, etc.
(2) Determine the segment interval k and the number of segments. When N/n(n is the sample size) is an integer, take k = n/n;
(3) determining the first individual number L (L ≤ K) by simple random sampling in the first paragraph;
(4) Sampling according to certain rules. Usually, the interval k plus L gets the second number of individuals (l+k), then K gets the third number of individuals (l+2k), and so on until the whole sample is obtained.
Edit paragraph 4. Cluster sampling What is cluster sampling?
Cluster sampling, also known as cluster sampling, is to merge all units in a group into several sets that do not cross and repeat each other, which are called groups; A sampling method for sampling in groups.
When cluster sampling is applied, each cluster is required to have good representativeness, that is, the differences between units within the cluster are large and the differences between groups are small.
Advantages and disadvantages of cluster sampling
The advantages of cluster sampling are convenient implementation and saving money;
The disadvantage of cluster sampling is that the sampling error caused by large differences between different groups is often greater than that caused by simple random sampling.
Implementation steps of cluster sampling
Firstly, the population is divided into group I, and then several groups are selected from group I clocks to investigate all individuals or units in these groups. The sampling process can be divided into the following steps:
Firstly, the label of clustering is determined.
Second, divide the whole (n) into several non-overlapping parts, and each part is a group.
Third, according to the sample size, determine the number of groups to be extracted.
Fourthly, a certain number of groups are extracted from group I by simple random sampling or systematic sampling.
For example, investigate the situation of middle school students suffering from myopia and make statistics in the last class; Conduct product inspection; All products produced by 1h are sampled and inspected every 8 hours.
Classification of cluster sampling and stratified sampling
Cluster sampling and stratified sampling are similar in form, but quite different in fact.
Stratified sampling requires large differences between layers, small differences between individuals or units within layers, small differences between groups and large differences between individuals or units within groups;
In stratified sampling, several units or individuals are extracted from each layer, while in cluster sampling, either the whole cluster is extracted or not.