Analyzing undergraduate students’ performance in engineering statistics course using educational data mining: Case study in UniMAP

Abstract

The remarkable growth of electronic data in the Mathematics department provides the opportunities to extract some meaningful information from this large capacity of data. The development of the data in the mining field ensures the educational data to improve the quality of the educational processes and to classify the components that affected the performance of the students in engineering statistics course. Thus, in this study, Kanri Distance Calculator (KDC) was utilized to analyze the performance of undergraduate students in Universiti Malaysia Perlis (UniMAP) during the academic session 2015/2016 in semester 2. The performance of the students was evaluated based on two aspects. First, the academic achievement of the students from the engineering statistics course during semester 2 2015/2016 was analyzed. Second, the possible variables that affect the performance of students in this course were classified. Thus, in this study, two significant groups of students were recognized i.e. (i) the students who achieved below 60% average, and (ii) the students who achieved above 60% average that can be referred as off-target and target group respectively. Based on the analysis of this study, the two variables (i.e. variables X3 and X7) affect the performance of the students and cause the students to obtain average marks lower than 60%. The results indicate that, by focusing on the possible variables that affect the academic performance of the students, it is possible to provide an appropriate warning and support to the students below average achievement, and at the same time to provide advice and opportunities to high-performance students based on the past experience about the obtained components.