COMPARISON OF RELIABILITY AND ACCURACY OF RESEARCH RESULTS USING RANDOM AND QUOTA SAMPLING

Abstract

The author analyses the reliability and accuracy of the results of sociological research conducted using quota sampling. The author discusses the advantages and disadvantages of using quota samples compared to random samples. The main focus is on the representativeness of the obtained data and the possibility of extrapolating them to the general population.In the problem statement, the author noted that a mechanical method of forming a random sample is most often used to study public opinion in Ukraine. The main feature of this method is to ensure an equal probability of selection of each element of the general population. The author emphasizes that although this method allows for easy calculation of possible errors and analysis of collected information, it has a disadvantage in representing specific groups of the population.To address this problem, the author proposes using an equal probability method of forming quota samples, which allows selecting respondents according to their quotas in the general population, thus providing a more representative model of the studied population. However, in the scientific community, there is no consensus on the use of indicators in forming quota samples, and all existing works remain at the level of theoretical assumptions.The author attempts to empirically substantiate the effectiveness of quota samples through experimentation. The hypothesis of the experiment assumes that the results of studies conducted according to quotas can be extrapolated to the general population with high accuracy and reliability using mathematical and statistical methods. During the experiment, both quota and random samples were used to compare the results.The general population consisted of 489 musical compositions represented in different proportions to eliminate the subjective component from the experiment and provide the opportunity to clearly assess and record the parameters of the initial general population for further comparison with the results of different types of samples. The musical compositions had two parameters: the name of the performer for identification and the duration of the track for digital and statistical evaluation. The analysis was carried out using a specially developed program for statistical data analysis. The collected data allowed for the assessment of the average track duration, median, skewness, standard deviation, and other statistical indicators.The results of the experiment showed that the quota sample provides more accurate results when the quotas are followed during the selection of elements of the general population. The sample obtained by random selection without adhering to quotas showed greater variability in indicators such as mean, median, skewness, and standard deviation relative to the general population. Thus, it was concluded that quota samples have advantages in sociological research, especially when specific groups of the population need to be considered.The author emphasizes the need for further research and improvement of quota sampling methods to enhance the accuracy and reliability of sociological studies, which are important for examining particular social issues.