<![CDATA[This study aims to develop and evaluate an artificial intelligence-based essay assessment system integrated with the National Work Competency Standard (SKKNI) in vocational Informatics education. The system was designed using the ADDIE model, featuring two core functions: (1) automated rubric generation based on indicators, criteria, and weights provided by teachers, and (2) automated essay scoring by a multimodal large language model capable of processing both image and text inputs. Evaluation was conducted on ten student essays assessed in parallel by three experienced teachers and the AI system. Reliability analysis using the Intraclass Correlation Coefficient (ICC) revealed a score of 0.975, classified as “excellent,” indicating strong agreement between AI-generated scores and human ratings. Qualitative findings from teacher interviews confirmed that the system not only reduces administrative burden but also reinforces the teacher’s role as a pedagogical curator, ensuring assessment remains aligned with learning objectives. The key conclusion is that integrating artificial intelligence within a pedagogically centered process (rather than replacing educators) yields a reliable, valid, and sustainable approach to automated essay scoring in vocational education.]]>
Copyrights © 2026
