<![CDATA[Purpose – This study tested whether the al-Abyan method improves lower-secondary students’ kitab turats (unvowelled classical Arabic) reading skills and compared learning gains against conventional instruction. It addresses persistent difficulties in decoding, grammatical parsing, and interpretation in Indonesian Islamic schooling and offers controlled evidence for a structured, interactive pedagogy. Design/methods/approach – A quasi-experimental, non-equivalent control group pretest–posttest design was implemented with two intact classes at Madrasah Dar al-‘Ilm al-Islāmiyyah, Cikulur, Serang (approximately n=27 per class). Procedures included instrument development and rater training, baseline testing, an al-Abien intervention for the experimental class, business-as-usual for the control, and posttesting with fidelity checks via structured observations and adherence checklists. Performance tasks were rubric-scored for decoding, nahwu–ṣarf parsing, and meaning rendition; analyses comprised assumption checks, independent-samples t tests, ANCOVA with pretest as covariate, effect sizes, and 95% confidence intervals. Findings – The experimental mean rose from 27.8 to 67.3 (gain +39.5), whereas the control mean shifted from 25.1 to 24.8 (−0.3). The between-group difference was highly significant, t(48)=17.93, p<.001, and corroborated by covariate-adjusted analyses. Classroom observations showed higher engagement, faster grammatical parsing, and more accurate interpretation in the al-Abyan class. Research implications – Generalisability is bounded by a single-site, nonrandom design and modest sample size, despite high implementation fidelity and equal contact time. Results support integrating al-Abien as a scalable module within Arabic curricula and motivate multi-site randomized or strong quasi-experimental replications, head-to-head comparisons with other active methods, and longitudinal tracking of retention and transfer.]]>
