Abstract
The rapid adoption of Computer-Based Testing (CBT) in higher education requires an evaluation framework to determine whether the implemented system successfully provides benefits for users. This study aims to evaluate the success of CBT implementation among cadets of a higher education institution in Indonesia using the DeLone and McLean Information Systems Success Model. A quantitative explanatory approach was employed involving 108 cadets who had experience using the CBT system. Data were collected through a structured questionnaire and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) with SmartPLS 3.2.9. The evaluation consisted of measurement model assessment and structural model testing using bootstrapping with 5,000 resamples. The results indicate that information quality significantly influences user satisfaction (β = 0.399, p = 0.001), and service quality significantly influences user satisfaction (β = 0.298, p = 0.011). However, system quality does not significantly influence user satisfaction (β = 0.225, p = 0.055). Furthermore, user satisfaction has a strong positive influence on implementation impact (β = 0.846, p < 0.001). These findings indicate that successful CBT implementation is determined not only by technical system performance but also by the quality of information and service support provided to users. This study contributes practical implications for higher education institutions in improving digital assessment systems by prioritizing information clarity, technical assistance, and user satisfaction.