MODERN APPROACHES TO STATISTICAL INFORMATION PROCESSING IN SOCIAL PEDAGOGY

Abstract

Introduction. The main task of social pedagogy is to study the patterns of socio-cultural adaptation of the individual, collectives, societies with adjacent humanity and harmonization of their relations. Social pedagogy has three basic features, namely: – extensive use of statistical data, since they are the basis for identifying certain patterns; – the need to process samples of large size in order to ensure proper reliability of statistical findings; – the analysis of huge data arrays encourages each researcher in the field of social pedagogy to make wide use of statistical software products in order to automate calculations. Since social pedagogy, in Ukraine, is still quite young discipline, then this peculiarities are not given due attention. Moreover, these principles are not fully understood by specialists in the field of social pedagogy. The purpose of this study is not only the disclosure of the value of each of these features of social pedagogy, but the first modern approaches to the statistical information processing of classical and neoclassical volumes. The research methodology is based on modern achievements of mathematical statistics and probability theory in the field of statistical information processing and mathematical model. Results. It is shown that the use of statistical data requires first of all a mathematically unambiguous and accurate understanding of the essence of the mean. It is indicated that it can have a scientific application only when homogeneous data are mediated. A mathematical definition of the homogeneity of statistical data is given. The necessity of using non-classical error theory of measurement (NETM) for samples with a volume of n > 500 is justified. An overview of the main groups of software products that allow you to automate the processing of statistical data in social pedagogy is given. Scientific novelty. For the first time, a mathematical exact definition of the notion of homogeneity of statistical data is given. It is shown that these are the results of statistical observations with a normal distribution, can always be tested using favorable conditions of mathematical statistics. A completely new proposal is on the use of NETM for processing large volumes. Also for the first time, this classification of software product groups for automating the processing of statistical information by social pedagogy specialists. Conclusion. Three principal features of social pedagogy are noted, as modern science, which deals with the processing of statistical data not only classical volumes (up to 500 observations), but also data that have a much larger volume due to computerization and automation of measurements and for which the fundamental axiom of normality is not capable according to the statement of the Cambridge Professor H. Jeffreys. In this case, it is necessary to apply the methods of non-classical error theory of observations.