Abstract
This study aims to develop a model for predicting academic dropout among foreign students. In view of this, the factors impacting Vietnamese students’ dropout rates in Korea were explored, and machine learning was used to predict the likelihood of dropout. First, a survey was conducted to ascertain students' learning preferences. Five factors were extracted: behavior control, learning motivation, task difficulty, interaction, and self-efficacy. A predictive model with high accuracy was developed by considering these factors as independent variables and items related to dropout as dependent variables. After using logistic regression analysis and support vector machines, predictive models with over 70% accuracy were developed. This shows that the variables extracted in this study are correlated with the dropout of foreign students, while also providing evidence that the current issues with teaching and learning can be well avoided in advance. These research results indicate that it is essential to focus on managing students who are showing early signs of dropping out.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have