Predicting Students’ “Passing or Failing” Status with the Utilization of Motivational Factors by Machine Learning Methods

Though motivation as an educational technique has long been introduced to pedagogy in general and language teaching in particular, it seems that scant heed has been given to its importance as well as the challenges it has. In an attempt to shed more light on the status of ELT enhancement with motivational factors, the researchers as university instructors attempted to predict the “passing or Failing” of the students at the beginning of the semester by using Machine Learning (ML) which is newly entered education. In so doing, the researchers arranged with a total of ۹۹ students from General English course to participate in the survey from both genders. The required data were collected via standard questionnaire on motivational factors that was in a ۵-point Likert type scale.For analyzing data, Linear Regression or Logistic Regression are often used according to the dependent variables. But, in the present study analyzing data has been done through Rapid Minder software using Random Forest (RF) which is the innovative technique for predicting students' performance. Thus, RF is utilized to compare with logistic regression (LR). The accuracy of ۷۳.۳۳% for LR and ۹۶.۶۶% for RF indicated a much higher accuracy in RF.