The progressive economic development of countries in the modern world is based on the corresponding development of scientific and innovative activities. The paper analyzes the main indexes characterizing the state of scientific potential and indexes reflecting the effectiveness of scientific activity in 43 countries of the world, including Russia and China. Interrelations of indexes are revealed and regression equations describing the existing dependencies are constructed. Based on the obtained models, the results of scientific activity are estimated for two countries: Russia and China. The estimates are compared with the actual levels of indexes and conclusions are drawn about the effectiveness of the use of available scientific resources.Purpose of the study. The aim of the study was to identify homogeneous groups of regions that are similar in their economic and innovative indexes, statistical analysis of these groups based on non-parametric methods and methods of correlation and regression analysis, and the formation of conclusions and recommendations regarding their innovative activities.Materials and methods. The following statistical methods were used in the study: non-parametric, correlation-regression, multivariate classifications (cluster analysis), discriminant analysis, descriptive statistics (averages, structural averages, variation indexes, etc.). The work used the statistical data of the World Bank, OECD, Rosstat. The calculations were carried out using the STATISTICA 12.0 software package.Results. The paper classifies countries according to the level of scientific potential and scientific performance. The cluster affiliation of Russia is determined. The search for a circle of countries that have similar conditions of scientific potential with Russia for further use of the experience of these countries is one of the goals of the paper. As a result of the analysis, it can be noted that the inventive activity of the Russian population is quite high, at the same time, the scientific potential in relation to scientific publications is used extremely poorly. In China, high levels of inventive activity and average citation of scientific publications can be noted. The number of patents granted, taken as a whole, has a strong linear relationship with GDP per capita and a strong non-linear relationship with domestic spending on research and development as a percentage of GDP and the number of people employed in research and development per 10,000 employed in the economy. By cluster groups, the listed dependencies were not found in the developed countries included in the first cluster, but were confirmed for the other of the countries.Conclusion. An analysis of the inventive activity of the Russian population showed that, with the existing scientific potential, the country managed to achieve much better results in this area than it could be based on the values of per capita GDP, research and development costs, the number of personnel, etc. (the excess according to different models is approximately 1.4 - 2.7 times). In general, for the totality of countries, the citation rate has a sufficient relationship with GDP per capita and the number of staff involved in research and development, but the volume of internal costs per researcher and the share of internal costs as a percentage of GDP do not significantly affect it. The experience of China confirms this conclusion: with a low cost per researcher, the country managed to achieve high results in the innovation field, but this phenomenon can be explained by the existence of a certain lag between the development of indexes. Separately, for cluster groups that divide the entire set of countries into highly developed, underdeveloped and countries occupying an intermediate position, no relationship was found between the citation index and other indexes.This paper is focused, first, on specialists dealing with the problems of the development of science, in particular Russian. The established relationships between indexes characterizing the level of scientific potential and scientific performance, described using linear and non-linear models, will help practitioners who decide on the organization and financing of science to find the best ways to solve emerging problems.