The failure to ensure spontaneous e-learning may not, in fact, be due to structural problems in the e-learning system itself, but may rather be attributable to a gap between the instructional design of materials provided by e-learning and the learning characteristics (approaches to learning) of students. Learning of 'Low completion rate' has been brought up as one of the teething problems of e-learning. When one does e-learning is drawing great attention anew as a problem yet to be solved. In this research, we propose, put forward, test...the hypothesis. There could be students who are not suitable to e-learning because the learning effect of e-learning varies depending on the approaches to learning as a cause of poor completion rate of students and conducted verification of this hypothesis. Verification revealed that the task fulfillment rate of e-learning varies depending on the approaches to learning, which is divided into seven categories. Outcomes of our study suggest that e-learning can produce almost no learning effect when the approaches to learning do not match the suitable e-learning instructional design. This issue from results also seems to be relevant for highlighting: 'high placement test scores are relevant to the respective constant learning characteristics over the entire period of e-learning'.