Comparative Analysis of Academic Performance and Student Satisfaction: A Statistical Study of Online Versus Traditional Learning Modes
Year: 2026
Type: article
Study design: Observational Study
Abstract: The rapid expansion of online education necessitates rigorous empirical investigation into its effectiveness compared to traditional classroom instruction. This study presents a comprehensive statistical analysis examining the relationship between learning modality and academic outcomes, utilizing data from 500 student records obtained from the UCI Machine Learning Repository. Employing multiple linear regression and binary logistic regression methodologies, we evaluated how learning mode, attendance rates, study duration, and demographic variables influence academic performance and student satisfaction. Results indicated that students in traditional learning environments achieved significantly higher academic performance scores (M = 76) compared to online learners (M = 73), with attendance percentage emerging as a significant positive predictor (β = 0.10, 95% CI [0.05, 0.15], p < 0.001). Logistic regression analysis revealed that online students exhibited significantly lower satisfaction levels (OR = 0.65, 95% CI [0.45, 0.95], p < 0.05). The multiple regression model explained approximately 15% of variance in academic performance (R² = 0.15), suggesting the presence of unmeasured confounding variables. These findings carry important implications for educational policy and institutional decision-making regarding learning modality implementation. Recommendations include enhancing attendance monitoring systems for online courses, developing interactive digital learning tools, and implementing hybrid learning models that leverage the strengths of both modalities.
Authors: Chinonso Job, Onwe, Festus Chijioke
Institution: University of Greater Manchester
Language: English
Topic: Online and Blended Learning
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W7128905620
