Aim: The authors analyze the problem of the survey sample in empirical researches. They show which methods are the most effective and how to calculate the minimum sample size to be representative in conducted researches.

Introduction: The authors illustrate this issue on the example of presidential election in the United States of America conducted in the first half of the twentieth century. The authors showed the survey of the popular American news magazine Literary Digest which several times rightly described winners of the elections in the USA. They also analyzed the case, when this magazine did not determine the winner correctly. There are given reasons for this error, including the explanation what a sampling selection is. In contrast to these events, the authors presented methods of the young researcher George Gallup, worker of the American Institute of Public Opinion, whose name later became synonymous with public opinion. In contrast to the American magazine, he correctly identified the winner of the election. He stated that when he had been doing the research, he had been trying to make the sample as representative as possible and separated from the general population by a proper draw.

Methodology: The article pointed valid, according to the authors, sampling methodology that have a significant impact on the effectiveness in empirical studies. They enumerated: purposive sampling, random selection method, snowball and quota sampling. A random sample means a part of the general population chosen in a way that guarantees each individual of the population an equal chance of being selected. Purposive selection is based on the matching test of researcher’s owns knowledge of the population as well as on the goals of the research. In snowball method, the researcher collects data about a number of members of the population, which are possible to find, and then asks these persons to provide information they have on other members of the population which is needed to locate them. In quota selection the researcher’s assumptions are that the sample is representative in terms of all the interesting features, if its structure based on a number of important features is identical to the structure of the surveyed population.

Conclusions: To accurately summarize the topic and draw right conclusions, the authors in the summary present the formula for the minimum size of the population, which you can actually utilize to determine the number of people who should be surveyed. A model is supported by examples that reflect the accuracy of the application and provide a better understanding of the problem.

Keywords: survey sample, selection of the sample, population, purposive selection, random selection, quota sampling, minimum number of members in sample

Type of article: review article