Refining Student Marks based on Enrolled Modules’ Assessment Methods using Data Mining Techniques

Abstract

Choosing the right and effective way to assess students is one of the most important tasks of higher education. Many studies have shown that students tend to receive higher scores during their studies when assessed by different study methods - which include units that are fully assessed by varying the duration of study or a combination of courses and exams - than by exams alone. Many Educational Data Mining (EDM) studies process data in advance through traditional data extraction, including the data preparation process. In this paper, we propose a different data preparation process by investigating more than 230,000 student records for the preparation of scores. The data have been processed through diverse stages in order to extract a categorical factor through which students’ module marks are refined during the data preparation stage. The results of this work show that students’ final marks should not be isolated from the nature of the enrolled module’s assessment methods. They must rather be investigated thoroughly and considered during EDM’s data pre-processing stage. More generally, educational data should not be prepared in the same way normal data are due to the differences in data sources, applications, and error types. The effect of Module Assessment Index (MAI) on the prediction process using Random Forest and Naive Bayes classification techniques were investigated. It was shown that considering MAI as attribute increases the accuracy of predicting students’ second year averages based on their first-year averages.