Knowledge points for preparing for the 2020 intermediate economist's economic foundation: several basic probability sampling methods
Knowledge points: several basic probability sampling methods
Probability sampling has different sampling methods:
simplerandom sampling
1. Simple random sampling can be divided into two methods: simple random sampling with substitution and simple random sampling without substitution.
For example: draw lots, draw lots.
2. Simple random sampling is the most basic random sampling method.
(2) stratified sampling
1. stratified sampling: it refers to dividing people into different layers according to certain rules, and then randomly sampling samples at different layers, so the samples obtained are called stratified samples.
Example: 1: A university has 180 faculty members, of whom 15 were selected for sightseeing. The stratified sampling design is: 180 employees are divided into 144 faculty members, from which12 is drawn; Managers 12, of which 1 was selected; 24 logistics service personnel, 2 selected.
Example 2: Take 20 samples from 300 stores in a certain area to know the operation of each store. The stratified sampling of the design is: 300 stores are divided into 30 big stores, from which 5 stores are selected; 75 medium-sized shops, choose 5 of them; Xiaodian 195, of which 10 is selected.
2. Features:
① Stratified sampling can estimate not only the overall parameters, but also the parameters of each layer.
(2) It is convenient to organize sampling work.
(3) Each layer should extract a certain sample unit, so that the samples are evenly distributed in the population, which can reduce the sampling error.
(3) Systematic sampling
1. System sampling: it refers to arranging all the units in the population in a certain order, randomly extracting an initial unit within a specified range, and then extracting other sample units according to the pre-specified rules. The simplest systematic sampling is equidistant sampling.
For example, a school selects 50 students for physical examination, and first numbers these 5000 students, that is, 0001~ 5000;
Interval k = 5000/50 = 100, that is to say, students in each interval 100 take 1 students; From 000 1 to 0 100, randomly select 0035 students as the starting point, select 0 135, 0235, 0335 ... until 4935, and extract 50 samples.
2. Features
The advantages of (1) system sampling are:
① Simple operation.
② The requirement of sampling frame is relatively simple.
(2) The disadvantage of systematic sampling is the complexity of variance estimation.
(4) Cluster sampling
1. Cluster sampling: all basic units in a group are divided into non-overlapping groups according to certain rules. When sampling, select groups directly, investigate all basic units for the selected groups, and do not investigate the unselected groups.
For example, in nearly 1000 residential quarters, the broadband ownership rate of residents was investigated. Select the residents of 15 building, and then investigate each household in the building, and estimate the broadband ownership rate of the whole community with the survey results.
2. Features:
(1) Advantages of cluster sampling: ① It is convenient to carry out investigation, which can save cost and time and has high investigation efficiency. (2) The sampling frame is simplified, and only the sampling frame of the group is needed, instead of all the sampling frames of the basic unit.
(2) The main disadvantage of cluster sampling is that the sampling error is relatively large.
(5) Multi-stage sampling:
Sampling methods that have gone through two or more sampling stages.