Objectives The purpose of this study is to lay the foundation for establishing a customized teaching and learning support system by clustering students based on the Learning Strategy Test (MLST) for freshmen in universities. Methods For this purpose, 1,503 freshmen from A university in G city were diagnosed using the MLST to analyze their personality, motivation, emotion, and behavior characteristics. Descriptive statistics, regression analysis, cluster analysis, and cross-analysis were conducted, focusing on the sub-factors and related variables of the MLST. Results The analysis of MLST sub-factors according to learners' academic achievement revealed that for personality characteristics, the scores for self-efficacy, outcome expectation, and conscientiousness decreased as academic achievement decreased. Learning motivation and competitive motivation also decreased with academic levels, while avoidance motivation was higher in the top academic group. In emotional characteristics, depression was lowest in the lower academic group but higher in the top group compared to the middle groups. Behavior characteristics scores decreased with lower academic achievement, but memory strategy scores were lower in the top group compared to middle groups. Regression analysis showed that MLST sub-factors explained 19.5% of academic achievement, with learning time, personality, and behavior characteristics being significant variables. When all 17 variables were included, the explanatory power increased to 22.9%, with notable contributions from self-efficacy, competitive and avoidance motivation, class listening, and test strategies. Four clusters were modeled based on MLST sub-factors: Yellow Light, Red Light, Blue Light, and Green Light. Cross-analysis showed significant differences when comparing these clusters with types based on study quantity and learning strategies (latent, stagnant, diligent). Conclusions This study holds significance as a foundational study to link the MLST tests conducted for freshmen in universities to more customized learning support
Read full abstract