A student’s engagement level when learning is critically related to his learning performance. Current methods of measuring student engagement levels digitally in the e-learning context include electroencephalography. However, this method lacks comprehensiveness and accuracy. This paper aims to devise a more holistic and accurate measurement of student engagement by proposing an original Engagement Index which assesses student engagement using a multimodal brain-computer interface. In our experiment, 10 student participants undertook a 2-phase experiment. Phase 1 involved collecting training data for the classifier, while phase 2 required participants to complete two reading comprehension tests with passages they liked and disliked, simulating the e-Learning experience. During the experimental session, Participants’ electroencephalography signals, eye gaze coordinates, heart rate variability and galvanic skin response data were recorded. An average classification accuracy of 91.9% was achieved. The proposed Engagement Index, through corroboration of data from multiple modalities, achieved a more accurate, reliable and comprehensive measurement of student engagement. Future work can look towards proposing a more accurate weighted engagement level score by modality, or achieving real-time engagement level feedback for educators in order to improve its function as a strong indicator of learning performance.