Students’ Biographical Text Writing Skills Using the Deep Learning Approach among Grade 10 Students in Phase E.5 at SMA Negeri 1 Koto XI Tarusan

This study was motivated by the low ability of students to write biographical texts. The problems included the suboptimal application of the deep learning approach, limited learning resources, and students' low understanding of the biographical text structure consisting of orientation, significant events, and reorientation. The study aimed to describe the ability to write biographical texts using a deep learning approach among Grade X Phase E.5 students of SMA Negeri 1 Koto XI Tarusan. This research employed a quantitative approach with a descriptive method. The population consisted of 100 Grade X Phase E students in the 2025/2026 academic year. The sample was 28 students of Grade X Phase E.5 selected through purposive sampling. The instrument was a performance test of biographical writing assessed through three indicators: orientation, significant events, and reorientation. Data were analyzed using percentages and mean scores. The results showed an overall mean score of 66, categorized as More than Sufficient. The orientation indicator reached 92.85% (Very Good), significant events reached 82.14% (Good), while reorientation reached 22.32% (Poor). The mean score increased from 63 to 66 after the implementation of deep learning, but the increase was not significant. Therefore, the deep learning approach had not yet proven effective in improving biographical writing ability in this research context. Students still need further guidance, particularly in developing the reorientation section.