<![CDATA[Abstrak. Integrasi Kecerdasan Buatan (AI) dalam penilaian pendidikan menawarkan peluang untuk meningkatkan kualitas evaluasi dan hasil belajar, namun penerapannya pada pendidikan literasi Al-Qur'an masih terbatas. Penelitian ini bertujuan menguji implementasi platform penilaian berbasis AI, Q-Rate (ngaji.ai), dalam mendukung kualitas penilaian dan kemampuan literasi Al-Qur'an siswa. Penelitian menggunakan desain mixed methods kuasi-eksperimental dengan pendekatan one-group pretest-posttest yang melibatkan 30 siswa SMK dan tiga guru Pendidikan Agama Islam di Yogyakarta. Data dianalisis menggunakan statistik deskriptif, uji paired-samples t-test, ukuran efek, dan analisis tematik. Hasil menunjukkan skor literasi Al-Qur'an meningkat dari 28,40 menjadi 41,75 (47,01%), dengan perbedaan yang signifikan (t(29) = −8,75; p < 0,001) dan ukuran efek yang besar (Cohen's d = 1,20). Guru juga memberikan penilaian positif terhadap objektivitas, efisiensi, kualitas umpan balik, dan dokumentasi penilaian. Temuan ini menunjukkan bahwa penerapan penilaian berbasis AI berpotensi mendukung peningkatan kualitas evaluasi dan hasil belajar literasi Al-Qur'an. Namun, mengingat penggunaan desain one-group pretest-posttest tanpa kelompok kontrol, temuan ini perlu ditafsirkan secara hati-hati dan memerlukan konfirmasi melalui penelitian dengan desain eksperimental yang lebih kuat.Abstrak: The integration of Artificial Intelligence (AI) in educational assessment offers opportunities to improve the quality of evaluation and learning outcomes, but its application to Qur'anic literacy education is still limited. This study aims to test the implementation of an AI-based assessment platform, Q-Rate (ngaji.ai), in supporting the quality of assessment and literacy skills of students. The study used a quasi-experimental mixed methods design with a one-group pretest-posttest approach involving 30 vocational school students and three Islamic Religious Education teachers in Yogyakarta. Data were analyzed using descriptive statistics, paired-samples t-test, effect size, and thematic analysis. The results showed that the Qur'an literacy score increased from 28.40 to 41.75 (47.01%), with a significant difference (t(29) = −8.75; p < 0.001) and a large effect size (Cohen's d = 1.20). Teachers also provide positive assessments of objectivity, efficiency, quality of feedback, and assessment documentation. These findings show that the application of AI-based assessments has the potential to support improving the quality of Qur'anic literacy evaluation and learning outcomes. However, given the use of a one-group pretest-posttest design without a control group, these findings need to be interpreted with caution and require confirmation through research with a more robust experimental design.]]>
Copyrights © 2026
