set of homogeneous objects are often examined with respect to any sign that characterizes them, measured quantitatively or qualitatively.
For example, if there is a batch of parts, the quantitative traits can be the size of the details of the guests, and the quality - standard parts.
If necessary, check them for compliance with sometimes resorted to a census, but in practice it is rarely used.For example, if the general population contains a large number of objects being studied, it is virtually impossible to carry out a census.In this case, selected from the totality of a certain number of objects (items) and examined.Thus, there is a general and sample.
General is the set of all objects that are subject to inspection or study.The population typically contains a finite number of elements, but if it is too large, in order to simplify mathematical calculations it is assumed that the entire set consists of an infinite number of objects.
sample or sample set is a part of selected elements from the totality.The sample can be repeated any repetitions.In the first case it is returned to the general population, in the second - no.In practice, often use repetition-free random selection.
population and sample must be linked representativeness.Speaking differently, so that the characteristics of the total sample could confidently identify the totality of symptoms, it is necessary that the elements of the sample as accurately as they were.In other words, the sample should be representative (representative).
sample will be more or less representative, if it is made randomly from a very large number of the totality.It could be argued on the basis of the so-called law of large numbers.Moreover, all members have an equal probability of being selected.
There are different options for selection.All these methods are, in principle, can be divided into two options:
- Option 1. sampled items when general population is not divided into parts.This embodiment may include simple random sampling without replacement re-selection.
- Option 2: The population is divided into parts and components are selected.These include a typical mechanical and serial selection.
Simple random - selection, in which the elements are extracted one by one from the totality of random.
typical - this selection, in which the elements are selected not from the totality, and of all its "typical" parts.Mechanical
- is selected such that the entire set is divided into the number of groups equal to the number of elements that must be in the sample, and hence each group is selected one item.For example, if you need to take away 25% of the parts manufactured machine, then choose every fourth item, and if you want to take away 4% of the items, choosing every detail of the twenty-fifth, and so on.It is necessary to say that sometimes the mechanical selection may not provide a sufficient representative sample.
Serial - is a selection in which elements taken from the totality of "series" is subjected to the continuous research, rather than one by one.For example, when a large number of parts are manufactured automatic machines, it is a continuous survey is carried out only for products of several machines.Serial selection is used, if the attribute in question is a minor variation in different series.
In order to reduce errors apply mathematical and statistical methods for estimating the population using a sample.Moreover, sampling can be both single-stage and multi-stage, which increases the reliability of the survey.
