What is the classification of the population when sampling, not only according to the weight of each

ProportionalStratified samplingThe differences between quota sampling and quota sampling are as follows.

At the macro level, the purpose of both is to provide better sample representativeness, and the theoretical basis of both comes from the fact that the higher the homogeneity of individuals in the population, the sampling error.

The smaller it is, the more representative the sample is. The essential difference between the two is whether they are based on probability, and proportional stratified sampling is probability sampling while the latter is non-probability sampling.

At the most macro level, the proportional stratified sampling produces a random sample that can be used to evaluate the sampling error and test the inference, and then generalize the conclusions of your sample to the population. Quota sampling itself does not have this possibility. In terms of specific operations, both need to select certain variables as the grouping basis.

And all need to be weighted (proportional) according to the number and proportion of the population in each group, stratified sampling is a kind of probability sampling, which refers to stratification first and then random sampling according to the proportion of each layer in the population. Quota sampling is a type of non-probability sampling, which refers to stratification and setting the subjective quota ratio of each layer in advance, and then sampling.

For example, if you investigate the work and rest of students in a high school, there are 1,000 students in Chakai High School, 350 students in the first year of high school, 330 students in the second year of high school, and 320 students in the third year of high school. Proportional stratified sampling:

The sample was 100, with 35 in the first year of high school, 33 in the second year of high school, and 32 in the third year of high school. Quota sampling (the main difference factor: the investigator may think that the problem of the third year of high school is more serious, so more quotas are given to the third year of high school in advance):

The sample was 100, 30 in the first year of high school, 30 in the second year of high school, and 40 in the third year of high school.

Stratified sampling has the characteristics of improving the representativeness of the sample, the accuracy of the population estimate and the efficiency of the sampling scheme, but the sampling frame is more complex, the cost is higher, and the error analysis is also more complicated. This method is suitable for situations where the overall complexity is large, and the differences between individuals are large, and the number is large.

Quota sampling. Also known as quota sampling, it is to select research units from various parts of the population according to the quota under the condition that a number of classifications are made for the population and the sample capacity is established. The specific steps are:

First, the population is divided into several types according to some attribute characteristics in the survey population, and then the various types of populations are divided into several subsets according to the classification control characteristics, and the sample amount is allocated according to the proportion of each subbody in the population, and then the sample units are selected by the sampler by arbitrary sampling or judgment sampling methods. The purpose of quota sampling is to ensure that the various subgroups of the population can adequately demonstrate the relevant attributes that the researcher needs to repent.

Quota sampling is usually classified according to certain attributes and characteristics of the overall potato unit, such as the age, gender, social class, etc. of the respondents. The advantage of this method is that it is simple and easy to ensure that all categories of the population can be included in the sample, so that the structure of the sample is similar to that of the population; At the same time, it has the advantages of high speed, low cost and convenience of data collection. However, because the sampler has a great deal of freedom to select the sample individuals in quota sampling, this sampling method can easily lose precision due to survey preference and convenience.

Quota sampling is further divided into independent control quota sampling and cross-control quota sampling.

Answer]: Quota sampling refers to the division of the population into several categories according to a certain characteristic of the overall unit of the market research object, and the amount of the sample unit is allocated next to each type of medium draft according to a certain proportion, and arbitrary or subjective sampling is carried out according to the amount of each type. Quota sampling is intended to illustrate the population, so it is important to have a good understanding of the nature of the population when classifying the population.

The rationale for the application of the quota sampling method is that it is believed that each unit in the same type of survey is roughly the same, and the differences are small, so there is no need for random sampling, as long as arbitrary or subjective sampling is sufficient.

Quota sampling has many advantages, such as simple and easy to use, time-saving and labor-saving, and can ensure that the sample units are evenly distributed in the population, and the survey results are more reliable, therefore, quota sampling is one of the most widely used methods in the non-random sampling method. Fixed-bridge sampling can be based on one criterion or more than one criterion when classifying the population. If the population of the market survey is classified according to only one standard, it is only necessary to allocate the number of sample units in each category according to the proportion of the number of units in the total number of units in the total. If the market survey population is classified by more than one criterion, the weighting on which the sample units are allocated will be different for each category.