Evaluation of PISA results with rough set theory

Education; it is a phenomenon that is indispensable and has no alternative for individuals, societies and states. A quality and prosperous life is only possible with good education that states provide to individuals. Education is a fundamental concept for the strength of the basic structure that sustains society and for transferring its values to the next generation. It is important to measure the quality of education that states and nations provide to their citizens. Many measurement tools have been developed to measure the quality of education provided. One of these measurement tools is the Program for International Student Assessment (PISA). Thanks to the Program for International Student Assessment (PISA), which was started to be implemented in three-year periods by the Organization for Economic Co-operation and Development (OECD), the first of which was in 2000, the participating countries; they gain information about their status in areas such as social and gender equality, school structures, education quality and standards, teacher profile, teacher education and professional development, individuals' interests and abilities, socio-economic levels within the family and geographical location. Thanks to these features, PISA determines the basic goals that countries should achieve in order to optimize their education systems. Thus, countries compete by taking these suggestions into account in order to get the best efficiency and increase the success in their education systems. In this thesis study, the PISA 2018 application, prepared by the OECD, consisting of 37 countries, including Turkey, is used to measure the literacy levels of the 15-year-old students of the participating countries in the fields of mathematics, science and reading, as well as to examine the quality of education and the social factors of the countries; by considering the scores and data in the field of mathematical literacy, the success of 37 OECD countries in the field of mathematics in this application and especially Turkey's place among these countries were examined. For this purpose, with the data obtained from the OECD official website, countries were first clustered using the k-nearest neighbor method, one of the hierarchical clustering analysis methods, and the k-means methods, one of the non-hierarchical methods. Then, decision rules for success were created using Rough Set Theory and data on variables. As a result, it has been interpreted how Turkey can follow a path to increase its success with the clusters created with the variables thought to affect success and the decision rules obtained.