МОДЕЛІ ФІНАНСУВАННЯ ТА ЯКІСТЬ ВИЩОЇ ОСВІТИ В КРАЇНАХ ОЕСР

Abstract

The analysis of the interrelationship of expenditures on higher education and the quality of higher education with higher education funding models in OECD countries and their cluster distribution, based on the described criteria, is done in the article. Two models of higher education funding are characterized: one is based on the source of financing and the possibility of influence of educational institutions on the system of higher education (proposed by V. Praneviciene), the other is based on the source of financing and the level of state regulation (proposed by O. Kuklin). The grouping of OECD countries was carried out based on the criteria of the amount of higher education funding per student and the quality indicator of higher education according to the U21 rating using IBM SPSS Statistics based on the Ward clustering method using the square of the Euclidean distance, the standardization of variables was carried out on the basis of Z-scores. As a result of the grouping of 23 countries, 5 clusters were formed, and the first cluster with the highest level of expenditure per student and the level of quality of higher education according to the U21 rating is represented by one country - the USA. A matrix of the relationship of higher education funding models in OECD countries and their cluster distribution based on the described criteria of the amount of expenditures and the quality of higher education was formed. Based on the analysis of the matrix, it was found that in the USA the market model is used for both classifications. In this case, it may be explained by the high level of socio-economic development of the country: due to the high level of GDP, it is possible to finance higher education mainly at the expense of private sources, and the state as a whole provides support for students who are unable to pay for education. In the countries of the second and third clusters, bureaucratic or anti-market models are mainly used. In the countries of the fourth and fifth clusters, there is a practice of using different models.