Building a Multiple Linear Regression Model to Predict Students’ Marks in a Blended Learning Environment

Abstract

This research attempts to build a multiple linear regression model to predict the marks of students. As a case study, the course M359 – Relational Database offered to undergraduate students of Arab Open University, Oman is taken. The model is trained using the different assessment marks of students in blended learning mode of the above course. Separate models were built based on the gender. Same datasets were used for training and testing purposes. The open source statistical software gretl was used to build and test the model. The study found that the model generated for male category shows more correlation in the process of prediction than the female category. The findings of the research suggest that it is challenging to build a prediction model for students in blended learning environment.