Аннотация
This article explores the potential of using machine learning methods to predict students’ critical thinking skills based on academic and behavioral data. Although critical thinking is widely recognized as a critical skill in modern education, its direct assessment remains challenging. The aim of the study was to identify the relationship between critical thinking development and measurable factors such as academic achievement, participation in extracurricular activities, and course selection. A dataset of 500 anonymized student profiles was collected and processed to extract key characteristics. Three models were developed and evaluated using standard performance metrics: linear regression, decision tree, and random forest regressor. The random forest model demonstrated superior predictive performance, achieving an R² value of 0.84, compared to 0.65 for the decision tree and 0.37 for the linear regression. The study’s findings demonstrate that students’ academic achievement, course selection, and level of extracurricular activity collectively provide valuable insight into their critical thinking abilities. Overall, the results support the effectiveness of data-driven approaches for indirectly assessing cognitive skills, which has practical implications for curriculum development, early intervention initiatives, and educational policy planning. By leveraging readily available educational data, this approach facilitates the development of more scalable, objective, and personalized assessment methods in the field of learning analytics.