Standard error formula of repeated sampling

The formula for calculating the repeated sampling error is s = √( 2500/ sample size) ×z, and the sampling error refers to the deviation between the statistical value of the sample and the inferred overall parameters, mainly including the difference between the sample average and the overall average.

Sampling error refers to the absolute deviation between sampling index and overall index due to accidental factors of random sampling, which makes the structure of sample unit insufficient to represent the structure of overall unit. The sampling error is not caused by the investigation error, but the unique error of random sampling.

Difference between non-repeated sampling and repeated sampling

First of all, the overview of the two is different:

1. Overview of repeated sampling: repeated sampling, also known as reset sampling or return sampling, refers to the sampling method of registering the units drawn in each time and then returning them to the group to participate in the next drawing.

2. Overview of non-repeated sampling: non-repeated sampling, also known as non-reset sampling or non-return sampling, means that in statistical sampling, each company can only extract once, that is, it will not return to the population after each extraction of company records, so that each extraction will reduce the population extracted next time by one unit.

Second, they have different characteristics:

1. Features of repeated sampling: The number of units to be selected remains unchanged at each sampling, and the previously selected units may be selected in subsequent sampling, so that the probability of each sampling is equal, and n times of sampling is equivalent to n times of independent experiments.

2. The characteristics of non-repeated sampling: each unit has only one chance to be selected at most; As more and more units are drawn, the chances of the remaining units being drawn are increasing; The error of non-repeated sampling is smaller than that of repeated sampling.