出版社：State Statistics Service of Ukraine, the National Academy of Statistics, Accounting and Audit (NASAA), the National Academy for Public Administration (NAPA) under the President of Ukraine
摘要：Запропоновано спосіб статистичного вивчення впливу чинників на формування результатів зовнішнього незалежного оцінювання учнів з української мови за регіонами України у 2014/15 н. р. З метою встановлення кількісних пропорцій зміни чинників та результативного показника розв’язано пряму й обернену статистичні задачі методом статистичних рівнянь залежностей. На основі множинної залежності встановлено вплив включених до розрахунків чинників на формування варіації результативної ознаки.↓Предложен способ статистического изучения влияния факторов на формирование результатов внешнего независимого оценивания учеников по украинскому языку и математике по регионам Украины в 2014/15 учебном году. С целью установления количественных пропорций изменения факторов и результативного показателя решены прямая и обратная статистические задачи методом статистических уравнений зависимостей. На основе множественной зависимости установлено влияние включенных в расчеты факторов на формирование вариации результативного признака.
其他摘要：Statistical study of correlations between socio-economic measures, their forecasting and simulation has important analytical function in statistical science, because enhanced industrial capacities is an important factor for solution of socio-economic problems, bridging regional gaps, growth of the national economy’s competitiveness, and improvement of education quality. The purpose of the article is to construct the method for factor analysis of results of external independent evaluation in Ukrainian regions in 2014/15 academic year by use of statistical regression equations. This method is used in solving direct and inverse statistical problems, to find proportions of change in factor variables and dependent variable. The direct problem is finding the dependent variable when a factor variable changes by 1 or by another standard value. Multiple linear regression is used to find the share of factors included in the calculation in the dependent variable. It is found that solution of the direct statistical problem can be interpreted by one-factor models, when one of the studied factors included in the calculation has the largest significance, that is, when its impact on the dependent variable has the major share. In cases when neither of the factors is of large significance for the dependent variable, solution of the direct statistical problem can be interpreted only by multiple regression model. The problems set according to the purpose of the study are solved by use of software “Method of statistical regression equations”. Input data are data on external independent evaluation of high school pupils in Ukrainian language and mathematics, by Ukrainian region, 2014/15 academic year; the pupils are included passing evaluation with 183.5-200.0 scores. Calculations are made by the above methodology adapted to the operation specifics of the education sector.
关键词：quantitative assessment of connections; method of statistical equations dependencies; paired association; multiple association; factor variable; dependent variable; results o f external independent evaluation.
其他关键词：quantitative assessment of connections; method of statistical equations dependencies; paired association; multiple association; factor variable; dependent variable; results o f external independent evaluation.
