DEVELOPMENT OF A PREDICTIVE MODEL FOR SELECTING A VOCATIONAL MAJOR AMONG VOCATIONAL CERTIFICATE STUDENTS

Abstract

The research aimed to 1) develop a predictive model for vocational program selection among students at IRPC Technological College using the Decision Tree technique and 2) evaluate the performance of the developed model. The sample consisted of 559 vocational certificate students (Levels 1–3) enrolled at IRPC Technological College in the 2024 academic year. The research instruments included a vocational aptitude test for industrial education programs covering five dimensions: reasoning, perception, memory, numerical ability, and spatial relations, together with students’ academic achievement records in eight core learning areas. The study was conducted following the CRISP-DM framework, and predictive models were developed using the Decision Tree technique. Model performance was evaluated using Accuracy, Precision, Recall, and F1-score.

The findings revealed that 1) the developed Decision Tree model effectively predicted students’ vocational program selection. The most influential predictors were vocational aptitude and academic achievement in the core curriculum subjects. The model clearly demonstrated decision rules and relationships among variables. 2) Regarding model performance, Model 1 achieved the highest predictive efficiency among the nine developed models. The model demonstrated satisfactory levels of Accuracy, Precision, Recall, and F1-score, indicating its capability to accurately predict students’ program choices. The findings suggest that the developed model can be used as a decision-support tool for students in selecting appropriate vocational programs and as a guidance resource for counselors and educators. This may help students choose programs that are consistent with their aptitudes and potential, thereby reducing inappropriate program selection and improving educational outcomes in vocational education.