SE-Authentic Assessment: Implementing Intelligent Systems for Adaptive and Practical Learning Evaluation in Higher Education

Abstract

Conventional assessment approaches in higher education often fall short of accurately evaluating students' applied competencies due to their lack of adaptability and contextual sensitivity. This study introduces the SEAuthentic Assessment Model, integrating intelligent systems to enhance student evaluations' precision, relevance, and objectivity. Validated through expert assessments with a mean Aiken's V of 0.82, the model aligns closely with learning objectives, curriculum standards, and competency measurement criteria. Practicality evaluations among faculty and students yielded mean scores of 81.07% and 83%, respectively, underscoring its accessibility and effectiveness in academic environments. The model's effectiveness was further demonstrated in an experimental study, with the experimental group exhibiting a 79.01% improvement in post-assessment scores and achieving a moderate N-Gain of 0.62, significantly outperforming the control group. Hypothesis testing confirmed these results, with the SE-Authentic Assessment Model significantly enhancing learning outcomes (p = 0.002). Faculty and students noted increased engagement and efficiency, with a 30% reduction in grading inconsistencies and a 40% decrease in assessment processing time. These findings position the SE-Authentic Assessment Model as a viable and scalable solution for elevating assessment accuracy and engagement in higher education contexts.