In studying the properties of an object is determined by the characteristics of not all units of the total population, but only some part of it.This allows performing a sample survey of the selected area adequately characterize all objects in general, any particular feature extraction or property.
able to select an object with a sufficient degree of probability to represent all the studied unit, work should be organized in a certain way.
There are several traditional methods to distribute a sample survey on the entire population, the main ones are the following: the method and direct conversion coefficients.The first option is used in cases where it is necessary to verify or validate the data obtained as a result of the continuous observation.In the second case it is determined by the product of the average value of the attribute in the whole volume of the total population.It is necessary to take into account the fact that not all factors allow to fully apply the point estimate of the propagation of the results obtained for a sample of the general population.Therefore, in practice, I found fairly widespread use interval estimation, using which it is possible to consider the size limit of error parts together and to calculate its average, and, if necessary, and to share the characteristic.
When research is carried out not on the entire population, but only a part of the inevitable errors may occur sample survey, which causes deviation of the results from the true values.This identified two main causes that can provoke distortions of the final result:
- imperfect technical conditions of the study, called sampling error;
- violation of the rules, which can be both random and systematic nature of the selection of units for research.
sample survey in the statistics is quite widely used.In this case, accurate information about the objects of study are available, significant savings and reduce costs.The guarantee of the representativeness of the research is scientifically sound selection of respondents for the study.
In practice, a sample survey carried out on most of the objects that were selected at random or regionalized way.In the former case, a same probability in sampling frames will get one or other unit.If the object of study after receiving information about it is returned to the general population, the random sample is called again.Otherwise it nonrepeated.Stratified sampling is formed as a result of the dismemberment of the population into a number of areas (groups) depending on the nature of the studied trait.Use of appropriate technology selection allows the presence of representatives from each group in the selected population.
If we talk about how it should be organized in a sample survey, in this case should necessarily meet the following conditions:
- number of units in the sample needs to be maximized to ensure the identification of the required laws;
- should adhere to the principle equiprobable hit each object in the sample;
- should use every single part of the field and the target population.
This will allow you to minimize the amount of the research, to obtain reliable information on your questions.
