Analysis of Students Performance in Mathematics before and during Covid-19 Pandemic using PageRank: A Preliminary Study

Abstract

In a normal situation, a university is constantly utilizing standard statistical analysis tools to provide a student's ranking through their academic achievement. The tools provides analysis of pairwise comparison among students on their academic performance but unfortunately visualization of the results is limited to the use of line graph, histogram, tables and boxplot which are not easily explained. In this study, another way of analyzing the pairwise comparison of academic performance on Mathematics among students is introduced that is by using interaction graph which is based on a weighted directed graph approach. The ranking of student’s performance in Mathematics is calculated using Page Rank (PR) algorithm. A sample of final examination result for twenty-one students whom enrolled in Mathematics courses in Fakulti Sains Komputer dan Matematik (FSKM), UiTM Shah Alam are investigated. Their performance by marks in two Mathematics courses taken before the pandemic and three Mathematics courses taken during the pandemic are analyzed. The graph with twenty-one nodes represent the students, while the directed links between two students represent the Mathematics relative performance is established. The rank of the students’ Mathematics performance is obtained from PR value of the graph. The largest PR value indicates the highest performance of the respective student. The result revealed that 62 percent of the students have shown better Mathematics performance even though the learning platforms before and during Covid-19 pandemic were drastically changed. Although this result does not indicate the whole picture of FSKM students' Mathematics performance, it gives a good insight to the academic administrator in making better decision. Besides, the interaction graph provides an easy way to explain the result only by looking into the graph.