<![CDATA[This study examines the integration of Evalbee OMR and CSV-based analytics to enhance curriculum quality assurance by enabling efficient assessment data collection in vocational higher education. Conducted within an English Vocabulary course involving 39 students and a 30-item multiple-choice summative examination, the study evaluates how automated data capture and structured item diagnostics can inform curriculum-level decision-making. Assessment responses were collected using OMR scanning technology and exported in comma-separated values (.csv) format, enabling systematic cohort profiling and item-level psychometric analysis. Internal consistency was estimated using the Kuder–Richardson Formula 20 (KR-20), yielding a reliability coefficient of 0.78, indicating acceptable consistency for classroom-based assessment. Descriptive statistics showed a mean score of 17.51 (58.3% proficiency) with symmetrical distribution. Item-level forensics identified two problematic items characterized by extreme difficulty and negative discrimination, as well as distractor ambiguity affecting response validity. Findings demonstrate that efficient digital assessment workflows not only reduce administrative processing time but also generate actionable evidence for curriculum refinement, item revision, and instructional recalibration. The integration of OMR-generated CSV analytics establishes a structured micro-feedback loop linking assessment practice to quality assurance processes. This study positions automated assessment data collection as a strategic instrument for strengthening evidence-based curriculum management in vocational higher education contexts.]]>
