In contemporary educational systems, there is increasing emphasis on individualizing the teaching process to meet the specific needs of each student. This necessity arises from the recognition that students possess different learning styles, interests, and abilities, which require the adaptation of teaching methods. In this context, hierarchical cluster analysis serves as a valuable tool for identifying groups of students with similar characteristics, enabling educators to tailor their approaches to these groups' specific needs.

This study aims to examine the formation of student groups in a STEM course and a social science course. The goal is to determine whether groups of students with similar characteristics will form in both courses or if the groupings will differ, and if so, by which characteristics these differences will be evident. The data used for this analysis were collected from the e-learning system and corresponding learning analytics data, which provide detailed insights into student behavior and performance.

The research methodology involves using hierarchical cluster analysis to form groups of students based on their characteristics in both courses. After forming the groups, ANOVA analysis was applied to compare the characteristics and performance of students within and between the formed groups. This methodological combination allows for a thorough analysis and comparison of learning patterns and student success in different types of courses.

The key findings of this study should provide insight into whether groups of students with similar characteristics form similarly in both courses or if the groupings differ. If differences are identified, the study will further analyze which characteristics these differences manifest. These findings are crucial for developing effective strategies for individualizing the teaching process. If it is shown that groups with similar characteristics form similarly in both courses, it may be possible to apply the same methods of individualization. However, if significant differences are identified, this will indicate the need for different approaches to individualization in different types of courses.

The limitations of this study include the fact that the analysis is based on only two courses from the same university. Therefore, the results may not be generalizable to all courses or universities. Broader and more comprehensive studies are needed, involving a larger number of courses from different universities, to confirm these findings and provide a broader understanding of the dynamics of student group formation. Despite these limitations, this study represents an important step toward understanding and improving the individualization of the teaching process.

Keywords: Hierarchical clustering, STEM courses, social science courses, individualized teaching, learning analytics.