Analytics can be used through statistical, econometric and machine learning methods to analyze a great amount of data about students’ and schools’ activities and results. Understanding the functioning of educational processes (and their determinants) is a key policy area that can lead to improvements in the quality of education and consequently human capital.

It is important to explore the determinants of financial literacy across countries, to understand which individuals and institutions are better prepared to live in the modern and economically complex society. OECD is an international organization which started to analyze the level of financial literacy of 15-years old students in various countries, by means of a large empirical analysis called PISA (Programme for the International Student Assessment). This big dataset, characterized by a lot of information of students and schools in several countries, can be explored by means of machine learning methods, with the aim of finding non-predictable patterns of the underlying educational process.

In the thesis, the student(s) will conduct the following activities:

• A systematic review of the academic literature which analyses the determinants of Financial literacy among students.
• A review of the potential machine learning and applied statistics methods that can be used for analyzing the available database (which is the 2015 wave of PISA, which includes data from many countries)
• An empirical analysis of the dataset, to find commonalities and differences between the determinants of financial literacy’s scores across countries.