Favorite and most difficult school subjects among STEM and non-STEM university students in Serbia and the USA

Attitudes toward school subjects are shaped by individual preferences, teaching practices, and broader sociocultural factors. However, cross-national research on how students interpret their favorite and most difficult subjects remains limited. This study employs a cross-sectional mixed-methods design to examine how undergraduate students from Serbia and the United States retrospectively evaluate their high school experiences. The sample consisted of 255 students (53.7% from Serbia, 46.3% from the United States; 59.2% female, 40.8% male), who provided open-ended responses regarding their favorite and most difficult subjects and the reasons underlying these choices. Responses were analyzed using an inductive coding procedure and subsequently examined through quantitative statistical analyses. The identified reasons were grouped into four dimensions: personal (interests and perceived abilities), content-related (subject characteristics), teacher-related (teacher’s role and support), and instructional (teaching methods and quality of instruction). In addition, both students and subjects were categorized as STEM or non-STEM to enable cross-disciplinary comparisons. The findings reveal systematic cross-cultural differences in explanatory patterns. Students from the United States more frequently selected STEM subjects as their favorites and tended to interpret their experiences through internal factors such as interest, self-confidence, and perceived competence. In contrast, students from Serbia more often attributed both enjoyment and difficulty to external, school-related factors, including teaching practices and curriculum structure. Mathematics emerged as a subject with a dual status, frequently identified as both favorite and difficult, highlighting the role of interpretive frameworks in shaping students’ responses to academic challenge. Overall, the results suggest that students’ perceptions of school subjects are not determined solely by subject characteristics but are constructed through context-dependent interpretations shaped by educational and cultural environments. These findings contribute to a more nuanced understanding of cross-cultural differences in students’ academic attitudes and have implications for designing educational practices that support both motivation and engagement, particularly in STEM domains.