European Journal of Educational Research, cilt.12, sa.1, ss.481-491, 2023 (Scopus)
In today's World, data-driven methods are behind the determination of potential action plans in every area of life. These data-driven methods help individuals or policymakers to figure out the strengths and weaknesses on the subject that are worked on and to make a comparison to the best practices. Thus, actions can be taken immediately on the specific factors that have a huge impact on the topic investigated. In the educational area, countries are using the same approach to measure, monitor, and improve the quality of education by attending international studies. In this study, for both Turkish and Singaporean students, Artificial Neural Network (ANN) model is performed to predict the students' mathematics achievement and to identify factors that have a high impact on achievement using Trends in International Mathematics and Science Study (TIMSS) in 2019 with the data of 3,586 Turkish and 4,750 Singaporean students. The reason behind comparing the results of Turkey to Singapore is that Singapore is the best-performing country in terms of mathematics achievement in the TIMSS in 2019. The model results show that the top two crucial factors in both countries are the frequency of absenteeism from school, and students’ confidence in mathematics with the accuracy of 75%. In addition, relevant policy implications are given based on the importance level of significant factors.