Abstract—It is widely recognized that mid-term examinations serve as a fundamental assessment method for evaluating students’ learning progress at the midpoint of a semester. A plethora of previous studies have underscored the significance of mid-term exams as a determinant of final grades, exhibiting a positive correlation with final exam outcomes. Nevertheless, these investigations frequently analyze the entire student population without accounting for disparities in students’ mental states and learning objectives, particularly in the aftermath of receiving their mid-term exam results. In the present study, we scrutinize the mid-term and final grades of 171 students participating in a statistics course. Diverging from prior research, which typically employs the entire student data as a singular group, we partition students into two distinct categories: those who attain higher mid-term exam scores and those who secure lower mid-term exam scores. Our analysis reveals that students achieving higher mid-term exam scores are more likely to obtain lower final exam scores, while students with lower mid-term scores tend to attain superior final exam results. This phenomenon is attributed to the influence of learning from failure, which motivates students who initially underperform to adopt new strategies and strengthen their learning objectives. Furthermore, we utilize a non-linear Support Vector Machine (SVM) model to forecast students’ final performance, recognizing that learning is a non-linear process replete with uncertainties. The model’s interpretation discloses that the mid-term exam and assignments administered during the mid-term exam period constitute the most influential factors impacting students’ final performance. Consequently, the meticulous monitoring of students’ mid-term grades and the implementation of strategic incentives to bolster their learning outcomes are paramount for ensuring their success in academic courses. 
 

Keywords—adjusting teaching strategies, student performance prediction, mid-term exam, machine learning in education 

Cite: Liyuan Liu and Meng Han, "Deciphering the Influence of Mid-term  Examinations on Student Learning Outcomes: A  Comprehensive Investigation Employing  Statistical and Machine Learning Approaches," International Journal of Learning and Teaching, Vol. 10, No. 1, pp. 32-39, 2024.

Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.