<![CDATA[This study aims to examine the effect of computer vision practice based on local datasets and data augmentation techniques on vocational high school (SMK) students’ data literacy. The research employed a pre-experimental design using a one-group pretest–posttest model (INTEC). Participants consisted of 34 eleventh-grade students from a Software Engineering program. Students were engaged in hands-on computer vision activities involving image classification using locally collected datasets representing contextual objects from their surrounding environment. The learning intervention also integrated data augmentation techniques, including image rotation, flipping, and brightness adjustment, to enhance dataset variability and model robustness. Data literacy was measured using a validated test instrument covering four indicators: data collection, data cleaning, data transformation, and data interpretation. Statistical analysis using paired-sample t-tests revealed a significant improvement in students’ data literacy scores after the intervention (p < 0.001), with a large effect size. The findings indicate that contextual computer vision practice combined with data augmentation strategies effectively strengthens students’ understanding of data processing and analytical thinking skills. This study contributes to the development of applied AI learning models in vocational education and supports the integration of authentic data-driven practices to enhance digital competencies in the era of artificial intelligence.]]>
