There are several methods of investigation

Sampling methods: isometric sampling, type sampling, whole group sampling.

1, isometric sampling: the overall units are arranged according to a certain sign or order to become a graph or a list of forms (that is, usually referred to as queuing), and then take a sample of units at equal distances or intervals.

Features are: drawn units in the overall is evenly distributed, and the sample can be less than pure random sampling. Equally spaced sampling can be lined up either with a sign related to the item under investigation or with a sign unrelated to the item under investigation.

2, type sampling: the overall unit according to its attributes and characteristics into a number of types or strata, and then randomly selected sample units in the type or stratum.

Features are: due to the stratification through the delineation of categories, increasing the **** homogeneity between the units in each type, it is easy to draw a representative sample of the survey. This method is suitable for the overall situation is complex, the difference between the units is large, more units.

3, whole group sampling: it is from the overall group into groups of survey units, rather than one by one to draw the survey sample.

Features are: survey units are more concentrated, the organization and conduct of the survey is more convenient. But the survey units in the overall distribution is not uniform, the accuracy is worse.

Based on the method of sample selection, sampling can be divided into probability sampling and non-probability sampling two categories. Probability sampling is in accordance with the principles of probability theory and mathematical statistics from the survey and research in the overall, according to the principle of random to select samples, and from the number of certain characteristics of the overall estimation of the inference on the inference of the inference of the possible error can be controlled from the probability of the meaning.

Extended information:

General procedure

(1) Defining the population

Defining the population is the process of first clearly defining the scope and boundaries of the population from which the total sample will be drawn, prior to the specific sampling.

(2) develop a sampling frame

The task of this step is based on the clearly defined scope of the total, collect a list of all the sampling units in the total, and through the list of uniform numbering to establish a sampling frame for sampling.

(3) Deciding on the Sampling Plan

(4) Taking the Actual Sample

The actual taking of the sample is based on the above steps, and is strictly in accordance with the selected sampling plan, from the sampling frame to select a sampling unit, constituting a sample.

(5) Evaluation of sample quality

The so-called sample evaluation is the initial test and measurement of the quality of the sample, representativeness, bias and so on, and its purpose is to prevent errors due to the sample's bias is too large.

Baidu Encyclopedia - Sampling Methods