Significance: A census can find out a country's national conditions and strength, especially its manpower, financial resources and material resources, and provide a reliable basis for the country to formulate long-term plans and policies. Therefore, it is a comprehensive survey with the advantages of comprehensive, detailed and systematic information; He is a one-time special survey, because the census is heavy in workload, high in cost and long in time period, so it is generally not appropriate to hold it frequently.
Features:
1, a unified standard time (data time) needs to be specified.
2. Generally, it is one-off or periodic.
3. The data is highly standardized.
4. It is suitable for the collection of important data related to the national economy and people's livelihood, and its application scope is relatively narrow.
The contents and time periods of statistical surveys conducted in China's census have been standardized and institutionalized, including:
1, census, once every 10 year, once every "0" year, such as the sixth census of China in 20 10 year.
2. The general survey of the tertiary industry is conducted once every 10 year and "3" year. For example, in 2003, the second general survey of the tertiary industry was conducted in China.
3. Industrial census, once every 10 year and in "5" year, such as the fourth industrial census in China in 2005.
4. Agricultural census, once every 10 year and in "7" year, such as the second agricultural census in China in 2007.
5. The basic unit census is conducted every five years in the years of "1" and "6". For example, the fourth national basic unit census in 20 1 1 year. Definition: it refers to taking some units from the population of the research object as samples to investigate and infer the digital characteristics of the population. This is the most commonly used survey method in market research.
Features: good economy, strong effectiveness, wide adaptability and high accuracy.
Sampling survey is a statistical survey method to infer the total number of overall signs according to some actual survey results, which belongs to the category of incomplete survey. According to the scientific principle and calculation, it extracts some sample units from the population of things composed of several units for investigation and observation, and uses the obtained survey mark data to represent the population and infer the population.
classify
(1) Simple random sampling: Also known as pure random sampling, it refers to the sampling units directly extracted from the population for observation without any treatment, classification or exclusion. Theoretically, it is the most consistent with the principle of randomness of sampling survey, and it is also the most basic form of sampling survey. Specific methods are: direct selection, lottery, random number table.
(2) Stratified sampling: also called type sampling or classified sampling. Firstly, the overall unit is stratified according to the main signs, and then several sample units are randomly selected from each layer, and the sample units of each layer form a sample.
(3) equidistant sampling: also known as mechanical sampling or systematic sampling. It queues all the units in the group according to certain symbols, and then extracts several sample units at equal intervals in a fixed order to form a sample with a capacity of n.
(3) Cluster sampling: all the units in the population are divided into several groups, and then some groups are randomly selected from the population in groups to conduct a comprehensive survey of all the units in the selected groups. The basic requirements of cluster sampling for population division are: first, there can be no overlap between groups, that is, any unit in the population can only belong to a certain group; Second, all units in the group are not omitted, that is, any unit in the group must belong to a certain group.
(4) Multi-stage sampling: When the population is large, the sampling process can be divided into several transitional stages, and finally the sample unit is selected. It is a statutory reporting system, which provides basic statistical data from bottom to top according to unified forms, unified reporting procedures and reporting time.
Statistical report system is a report system which is arranged from top to bottom and provides statistical information from bottom to top.
Statistical report is a survey method based on comprehensive survey. It is a survey method in which the competent government departments arrange statistical tables and administrative means from top to bottom according to statistical laws and regulations, and then enterprises and institutions summarize and report from bottom to top to provide basic statistical data step by step.
Statistical survey method system
The collection of statistical data should not only use one survey method, but should flexibly choose different survey methods according to the characteristics of the respondents and the conditions of the survey institutions to form a survey method system. The statistical method emphasizes that statistical investigation should be based on regular census, with regular sampling survey as the main body, supplemented by necessary statistical reports, key surveys and comprehensive analysis, and collect and sort out basic statistical information.
Statistical data collection method
1, main data collection methods
(1) direct observation method.
(2) reporting methods.
(3) Access mode.
(4) Questionnaire survey.
(5) Online survey method.
(6) Satellite remote sensing method.
(7) telephone survey method.
(8) Projection method.
2. Secondary data collection method
The secondary data collection method mainly refers to the copywriting investigation method. Copywriting investigation, also known as literature investigation, desktop investigation and indoor investigation, is an investigation method to collect second-hand information.
The implementation steps of the replication survey method are as follows:
(1) Determine the data type required for the survey according to the purpose of the survey.
(2) Determine the sources and channels of survey data.
(3) Search and collect data.
(4) Collating the secondary data searched and collected.
Design of statistical investigation scheme
1, determine the purpose of the investigation.
2, determine the investigation object and investigation unit.
3. Ensure that the investigation items (1) are few and precise; (2) Only include the items for which a clear answer can be obtained based on the principles of necessity and possibility; (3) Try to keep in touch with each other so as to check and balance each other; And (4) options for selecting some items.
4. Determine the investigation time and period.
5. Formulate the investigation organization and implementation plan.
6. Choose the survey method (1), direct observation method (2) report method (3) interview method (4) network survey method (optimal).
The meaning and types of statistical investigation errors
Statistical survey error is the difference between the statistical data obtained from the survey results and the actual number of the survey population.
There are two kinds of statistical investigation errors, one is registration error, and the other is representative error.
The key survey is an incomplete survey in which some key units are selected from the survey objects. Although the number of these key units is small, their total number of symbols accounts for the vast majority of the total. Therefore, when the investigation task only requires grasping the basic situation and basic development trend of things, but does not require comprehensive and accurate information, and there are indeed key units on the whole, it is more appropriate to carry out key investigations.
There are two organizational forms of key surveys: one is a one-time survey organized by specialized agencies; The other is to use regular statistical reports to conduct regular surveys on some key units. Typical survey is a kind of non-comprehensive special survey, which is based on the purpose and requirements of the survey, and on the basis of comprehensive analysis of the respondents, some typical or representative units are consciously selected for investigation. Its main functions are: first, to supplement the deficiency of comprehensive investigation; Second, verify the authenticity of the comprehensive survey data under certain conditions. Compared with other survey methods, typical survey has the advantage of flexibility, and in-depth and detailed statistical data can be obtained through a small number of typical surveys. However, this kind of survey is greatly influenced by people's subjective understanding, because it is limited by consciously choosing some representative surveys. Therefore, it must be combined with other surveys to avoid subjective one-sidedness.