Vyborka- this set of data, combined with the appropriate procedures of the general population for the research analysis.Representativeness - the property playing the whole idea of its parts.In another, it is an opportunity to spread understanding of the parts on the whole that this part comprises.
representativeness of the sample - a measure, which consists in the fact that the sample must be fully and accurately display the attributes of the population, of which it is.It can also be defined as a property of the sample most fully represent the characteristics of the population, significant from the point of view of the purpose of the study.
assume that the general population - all students of the school (900 people from 30 classes, 30 people in each class).The object of study - the ratio of students to smoking.The sample set, which consists of 90 pupils of the senior classes only, worse will present the totality than the same sample of 90 students, which would include in each class 3 student.The main reason - the unequal distribution of the ages.Thus, in the first case a representative sample is low.In the second case - high.
In sociology, said that there is the representativeness of the sample and its representativeness.
As an example of non-representative sample can result in a classic case occurred in 1936 in the United States during the presidential election.
Magazine "Literari Digest", which to this very successfully predicted the results of the previous elections, this time was wrong in their forecasts, although sent written questions to several million subscribers, as well as respondents who are chosen from phone books and from vehicle registration lists.In the fourth ballot, which returned filled back, voting was as follows: 57% to be guided by the Republican candidate Alf Landon called, and 41% expressed a preference for the incumbent president - Democrat Franklin Roosevelt.
In fact, won the election Roosevelt, who scored almost 60% of the vote.Error "Literari Digest" was next.They wanted to increase the representativeness of the sample . And since they knew that most of their followers consider themselves to be Republicans, they decided to expand the sample due to the respondents, their choice of phone books and automobile registration lists.But they do not take into account the existing realities and actually took away even more supporters of Republican because of the Great Depression have cars and phones could afford to middle and upper class.And that was for the most part the Republicans, not Democrats.
There are different types of sampling: simple random, continuous, typical, mechanical and combined.
Simple random sampling is the selection from the entire set of units under study at random, without any system.
Mechanical sampling used when in the population is an ordered, for example, there is a certain sequence of units (the registration numbers of employees, the electoral lists, the phone numbers of respondents, the numbers of apartments and houses, etc.).
typical selection is used when the entire set can be divided into groups according to types.When working with such a population can be, for example, education, age, social group, the study of business - industry or individual organizations and others.
Serial selection is useful when the units are combined in small series or group.This series can be batch of finished products, school classes, labor groups and other groups.
Combined sampling involves the use of all previous types of samples in varying combinations.
