Analysis of Learning Quality Factors on Student Learning Experiences Across Study Programs at University X Using Multiple Linear Regression

Abstract

Abstract.  This study examines the influence of six learning quality indicators technology access, content quality, teaching quality, resource availability, interaction and collaboration, and learning experience on overall student satisfaction across study programs at University X. Employing a quantitative approach, data were collected from 98 students representing 18 different study programs through a structured questionnaire using a Likert scale. Multiple linear regression was selected as the analytical method due to its capacity to simultaneously model relationships between multiple independent variables and a single continuous dependent variable, offering interpretable coefficient estimates and effect size quantification through the coefficient of determination (R²). Descriptive statistics, outlier detection, and heatmap-based per-program analysis were performed prior to regression modelling. Results indicate that the model explains 87.3% of the variance in student satisfaction, with Interaction and Collaboration, Teaching Quality, and Content Quality, emerging as the three strongest predictors. Technology Access, despite receiving the highest aggregate score, demonstrated the smallest regression coefficient, suggesting that high infrastructure availability alone does not guarantee proportional satisfaction. Notable variation in scores across study programs further indicates heterogeneity in students' learning experiences. These findings provide data-driven insights for institutional policy-making aimed at improving the overall quality of education at University X.

Keywords: higher education, learning quality, multiple linear regression, Student satisfaction, technology access, teaching quality