All about the random number method and the table of random numbers

The random number table method, also known as the "random number table method", is a method of using a random number table to draw samples.

A random number table is also known as a random number table. It is to use a special code shaker (or electronic computer) to automatically shake out (or computer generated) a certain number of numbers into a table according to the requirements of the number of encoded digits (such as a group of two, a group of three, a group of five or even a group of ten digits) by using a special code shaker (or electronic computer) for future reference. There is an equal probability of any number in this table.

When using this table to draw samples, the cumbersome process of sampling can be greatly simplified.

Directions The following table is a random number table:

The specific steps for the application of the random number table method are: the overall units of the survey are numbered one by oneArbitrarily specify the starting point of sampling and the order of sampling on a random number table;The sample unit numbers are drawn from the random number table in turn. All the numbers drawn within the numbered range are the numbers of the sample units until they are full.

Here's a specific example: For example, if a company wants to investigate consumer demand for a certain product, it needs to select 10 households from 95 households to select a sample using the household code meter method. The specific steps are as follows:

Step 1: Number the 95 households, one number for each household, i.e. 01 95. (The number of residents in each household is 2).

Step 2: In the table above, randomly determine the starting point of sampling and the order of sampling. Suppose you start with the first row, column 6, and the sampling order is from left to right. (Horizontal sequences are called "rows", and vertical sequences are called "columns.") Therefore, the sixth column here is the number 36. ）

Step 3: Draw the numbers in turn, resulting in 10 sample unit numbers for .

Households with these numbers were the subject of the sample survey.

It should be noted that the resident household number 69 appears in the sample twice. This is a duplicate sampling. The so-called repeated sampling refers to the fact that a unit in the population is selected as a sample, and then put back into the population, and may be selected as a sample for the second time.

Non-duplicate sampling means that each unit in the population can only be sampled once. That is, after a unit is selected as a sample, it cannot be put back into the population, and it is not possible to be selected as a sample for the second time.

In the above example, if it is required that there is no duplicate sampling, the method is as follows, continue to draw from 16 to the back, and the next two numbers are not within the overall number range and are excluded. Then 16, there are still duplicates, excluded. This is followed by 29, which is not repeated and can be selected.

In this way, the final 10 sample unit numbers should be .

In the upper middle and middle areas, if the survey population is changed to 800 households, the sample size is still 10 households, and the sampling starting point is row 3, column 1, and the sampling order is from top to bottom. How are the resident numbers of these 10 sample households generated?

First of all, the survey of the population of 800 households numbered from 001 800. (The household number of each household is a three-digit number). Secondly, the starting point of sampling should be "167", and the 10 sample units generated in turn are respectively .

The random number table method is used to select samples, which completely excludes the possibility of subjective sample selection, so that the sampling survey has a strong scientific nature.

Random numbers are the result of specialized randomized trials.

Random numbers are used in different statistical techniques, such as when drawing a representative sample from a statistical population, or in the process of assigning experimental animals to different experimental groups, or when performing Monte Carlo simulation calculations, etc.

1. The steps of the random number table method are to first unify the overall units of the survey and fully mix them, and then select the sample unit numbers from the random number table in turn according to the specified sampling starting point and the specified sampling order. Uniform numbering: that is, all research subjects in the population are uniformly numbered and then fully mixed, with the aim of evenly distributing the sample numbers, in line with the principle of "equal opportunity".

2. Determine the starting point of sampling: according to your needs or wishes, select a number number on the table, and the number will determine the starting point of sampling.

3. Determine the sampling order: according to the need or willingness, select a certain sequence direction, and use the sequence direction for extraction.

4. Final admission number: The number will be admitted in turn according to the sampling starting point and sampling order until the number of samples required to be drawn is full.

The random number table method is a method of generating a sequence of random numbers, and the following are detailed steps for example:

1. Prepare a table of random numbers: First, you need to prepare a ** or list containing a large number of random numbers. Each random number should be an integer or decimal in a given range. For example, we prepare a ** with 10 random numbers123。

2. Select the starting position: Select a starting position from the random number table, which can be any position. This position will be used as a starting point to generate a sequence of random numbers. For example, we choose the starting position as the 3rd random number, ie.

3. Read random numbers: Start from the selected starting position and read random numbers one by one according to a certain step size. The step size can be 1 or another integer greater than 1.

For example, let's choose a step size of 2 to start reading backwards from the 3rd random number. First Read: Second Read:

3rd Read: 4th Read: .

4. Use the random number read by the late mu: use the random number read for scenarios that require random numbers, such as simulation experiments, data analysis, etc. Taking the read random numbers as an example, in the simulation experiment, these random numbers can be applied to the corresponding model according to specific requirements to generate corresponding results.

Application of the random number table method

1. Education and learning: In the process of education and learning, the random number table method can be used to demonstrate and understand the concept of random events. For example, by tossing a coin or rolling dice, a table of random numbers can be used to simulate randomness, helping students understand the basic concepts of the Numsen rate and statistics.

2. Simulation experiment: The random number table method can be used to conduct simple simulation experiments. For example, in computer programming, tables of random numbers can be used to simulate random events during an experiment. This is useful for verifying the effectiveness of the algorithm and the randomness of the execution results.

3. Data generation and testing: The random number table method can be used to generate random data sets for testing and evaluating the performance of the algorithm. For example, in machine learning, a table of random numbers can be used to generate training samples for the Suihuai sock machine to train and validate the model.

4. Statistical inference: In some statistical analysis, the random number table method can also be used to conduct sampling and construct random samples. Although the random number table method cannot provide true randomness, in some specific statistical methods, the random number table method can be used to generate some approximately random samples for statistical inference and hypothesis testing.

1. Unified numbering.

That is, all the research subjects in the population are numbered uniformly and made into a signature. If the subject is less than 10 and only 1 digit is taken, the number is 1 9;If the study object is between 1 99 and takes 2 digits, the number is 01 99.

If the study object is between 1 999, take 3 digits and the number is 001 999;And so on. It is then thoroughly mixed in order to evenly distribute the sample numbers, in line with the principle of "equal opportunity".

2. Determine the name of the line.

According to their own needs or wishes, the researcher selects any row of numbers on the table, and the number of the row determines the number of starting rows. If you select row 25, the 25th row starts from top to bottom. If the selected number is greater than 50, subtract 50 from that number to get the remainder as the starting line.

If you select row 0 or 00, row 50 starts with row.

3. Determine the listing name.

Column names are determined in the same way as row names.

4. Admission number.

After the row name and column name are decided, the number at the intersection of the selected starting row number and the starting column number can be used as the starting point, and then the number will be admitted in a certain order direction (such as from left to right or right to left, from top to bottom or bottom to top).

If you encounter a number that is larger than the population or a duplicate number, you should discard it until the number of samples drawn is full. The numbers obtained using the above method are randomly selected.

There are several steps to sample with the random number table method:

1. Number the individuals in the population;

2. Select the starting number;

3. Get a random number table of sample numbers.

The use of this table to draw samples can greatly simplify the cumbersome process of sampling, but it is not suitable for cases where the number of individuals in the population is large.