Sistem Evaluasi Dosen oleh Mahasiswa Berbasis Web dengan Pendekatan MODEL PROTOTYPE

Studi Kasus di Fakultas Teknik Universitas Dr. Soetomo

Abstract

The Student Evaluation of Lecturers (EDOM) at the Faculty of Engineering, Universitas Dr. Soetomo has historically been conducted via Google Forms and processed manually, resulting in slow response times, repetitive recap procedures, and insufficiently documented reporting validity. This study aims to design, build, and evaluate a web-based EDOM system using the Prototype Model (MODEL PROTOTYPE) to improve data collection efficiency, recap accuracy, and the effectiveness of feedback for teaching quality assurance. The development method comprises requirements elicitation with the Quality Assurance Unit through a review of SPMI documents and brief interviews, architectural and interface design (Laravel, MVC pattern, and MySQL), staged prototyping, and rapid iterations based on stakeholder feedback until functional conformity is achieved. The resulting system provides authentication and role-based authorization (Admin, GPM, Lecturer, Student), odd/even period management, a question bank employing a Likert scale, anonymous questionnaire submission, score recap per lecturer and per category (pedagogical, professional, personal, social), dashboard visualizations, and report export to support governance needs. Functional evaluation was conducted through black-box testing on core scenarios (login, period scheduling, submission, recap, and export) and indicated conformance to specifications. User acceptance evaluation employed the Technology Acceptance Model (TAM) to examine perceived usefulness, perceived ease of use, attitude toward use, and behavioral intention; the results indicate positive acceptance and potential for operational adoption. Practically, the prototype approach accelerates requirement alignment, reduces the risk of mis-specification, and facilitates change control, while the Laravel-based implementation supports maintainability, role-based access security, and further feature development. The study is limited to a single faculty; future work may include SSO integration, audit trails, and longitudinal, cross-semester analytics for more comprehensive monitoring of lecturer performance.