Predicting Academic Performance in Psychology Students from Their Social Skills Using Classification Algorithms

Abstract

The high academic performance of students is the result of several factors, one of them being the development of social skills. Social skills allow students to face different circumstances, such as academic, personal, or professional. Students who develop these skills can quickly adapt to stressful and conflictive situations, generating complete well-being that facilitates learning, reflected in their academic performance. This study is critical because it helps educators identify students at risk of poor academic performance and provide adequate academic support. The research aims to identify predictors of academic performance based on social skills through knowledge discovery in databases (KDD) to clean data using classification algorithms, specifically Random Forest. To collect the data, a sociodemographic form, and the Social Skills Scale (SSS) were applied to 93 students of General Psychology, face-to-face modality, at Indoamerica University. The research results indicate that academic performance predictors are linked to gender and social skills, such as the ability to say no and cut interactions and the expression of states or disagreement. These findings suggest structuring support programs, academic guidance, and social skills development to improve academic performance and future career success. In conclusion, the research provides a new perspective to work in student welfare departments to improve students' academic performance.