(1) the purpose of the survey: the purpose of the survey should be consistent with the objective reality, is the first set of any program to clarify the issue, is the guide to action.
(2) survey object and survey unit: the survey object is the overall, the survey unit is the overall individual.
(3) survey items: that is, the content of the survey unit to be registered.
(4) questionnaire: that is, the survey items are arranged in a certain order of a form. Questionnaire generally have two forms: a single table and a list. List of many units of the project in a table, it applies to the investigation of the project is not much; a single table is in a form of registration of only one unit of content.
(5) survey methods and methods: survey methods are census, key surveys, typical surveys, sample surveys, statistical reporting system. Specific collection of statistical information on the survey methods are: visit method, observation method, report method and so on.
(6) survey location and survey time: the survey location is to determine the location of the registration information; survey time: involves the investigation of standard time and the duration of the survey.
(7) organizational plan: refers to ensure the implementation of the survey specific work plan.
Expanded Information:
. strong>
Statistical survey is based on the purpose and requirements of the survey, the use of scientific methods of investigation, planned and organized to collect data and information information on the statistical work process. According to the organization of statistical surveys, statistical surveys are divided into three categories: national statistical surveys, departmental statistical surveys, and local statistical surveys.
Types of surveys:
1. According to the different scope included in the survey object, it can be divided into comprehensive survey and sample survey.
(1) a comprehensive survey is a survey of all the units in the object of investigation in its entirety, its main purpose is to obtain the overall comprehensive, systematic and complete total information. Such as the census. Comprehensive survey to spend a lot of human, material, financial and time.
(2) sampling survey is part of the object of investigation unit survey. Such as focus surveys, typical surveys, sample surveys and non-comprehensive statistical statements.
Comprehensive surveys and sample surveys are distinguished by the different scope of units included in the survey object, not in terms of whether the final results obtained reflect the overall characteristics of the comprehensive information.
2, according to the registration time is continuous or not, can be divided into regular surveys and one-time survey.
(1) recurrent survey, is with the development of the object of investigation in time, and at any time on the changing situation of continuous registration. Its main purpose is to obtain statistical information on the whole development process of things and its results.
(2) one-time survey: is not continuous registration of the survey, it is the state of things every period of time after a certain point in time to register. Its main purpose is to obtain information about the level and condition of things at a certain point in time.
One-time surveys are divided into regular and irregular. Regular survey is every fixed period of time for a survey, irregular survey is not exactly equal to the time interval, and a long time between surveys.
3, according to the organization of the survey, can be divided into statistical reporting system and special surveys.
(1) statistical reporting system: it is in accordance with the unified national survey requirements and documents (indicators, forms, calculation methods, etc.) to provide statistical information from the bottom up a reporting system.
(2) Specialized surveys: statistical surveys specially organized for a particular purpose. Including: census, sample surveys, statistical reports, key surveys, typical surveys.
Classification
(1) Simple random sampling: also known as pure random sampling, it refers to the total does not make any treatment, not classified nor excluded, but completely in accordance with the principle of randomness, directly from the total sample units to observe. Theoretically speaking, it is the most consistent with the random principle of sampling, and is the most basic form of sampling. Specific methods include: direct sampling method, lottery method and random number table method.
(2) stratified sampling: also known as type sampling or classification sampling. It is the first overall units according to the main signs to be stratified, and then in each stratum in accordance with the principle of random sampling of a number of sample units, from the sample units of each stratum to form a sample.
(3) isometric sampling: also known as mechanical sampling or systematic sampling. It is the overall all units according to a certain sign queue, and then in a fixed order and at equal intervals in the overall number of sample units, constituting a capacity of n sample.
(3) whole group sampling: the overall units are divided into a number of groups, and then take the group as a unit, from the overall randomly selected part of the group, the sampled group of all units for a comprehensive survey. The basic requirements of whole group sampling for the overall division of groups are: first, no overlap between groups, that is, any unit in the overall can only belong to a certain group; second, all the overall units are not missing, that is, any unit in the overall must belong to a certain group.
(4) multi-stage sampling: when the total is very large, the sampling process can be divided into several transitional stages, to the end of the specific sample units.
Reference: