2. A Model to Determine Factors Affecting Students Academic Performance: The Case of Amhara Region Agency of Competency, Ethiopia .
Author(s): Aklilu Mandefro Messele and Melkamu Addisu

Abstract: At this time the amount of data stored in educational institutions is increasing rapidly. These data contain hidden information for improvement of students‟ performance, guidance, teaching, planning, and so on. Identifying factors that influence students‟ academic performance help educational stakeholders to take remedial measurements to improve performance of their students. In this paper total of 7,561 students‟ data covering the period from 2008-2011 with 28 attributes is used to determine the most influential factors. The classification algorithms J48 algorithm and Naive Bayes algorithm is used to develop the model. Design science research methodology is used as a frame work while the hybrid six-step Cios model is followed to develop the model. Many experiments were done with J48 algorithm and Naive Bayes classifier by changing the default values and reducing the number of attributes. However, 8 experiments are presented for analysis which shown better accuracy than the rest. The results of this study have shown that the data mining techniques are valuable for students‟ performance model building and J48 algorithm resulting in highest accuracy (70.3468% & 83.3552%) for practical and theory exams respectively. It also reveal that Education mode of training experience, Level, Purpose of Assessment, Candidate‟s category, Age, Sector, Sex, and Employment type found to be the most influential factors for students‟ academic achievement. Hence, future research directions are pointed out to come up with an applicable system in the area.

Leave a Comment

Your email address will not be published. Required fields are marked *