<![CDATA[Test reliability is a fundamental requirement in organizing educational evaluation instruments, especially in Islamic Religious Education (PAI), to ensure consistency and fairness in student assessment. This study aims to conceptually examine test reliability through the split-half method and parallel forms (equivalence) as well as its relevance to Islamic educational values. This study uses a qualitative approach with a literature review method. Data collection was carried out by documenting and analyzing secondary sources in the form of books, scientific articles, and related references. The literature selection was limited to publications from 2018–2025 sourced from national and international indexed journals and standard textbooks on psychometrics and educational evaluation. The results of the study show that the split-half method is effective for measuring internal consistency in a single test, where its correlation is corrected using the Spearman-Brown formula. Meanwhile, the parallel form method emphasizes cross-form consistency by arranging two equivalent test packages, which require a high level of test developer competence to minimize measurement error and prevent practice effects. In the context of PAI learning, the application of both reliability methods is not only an academic requirement but also reflects Islamic moral values such as justice, thoroughness, accountability, and professionalism as emphasized in the Qur'an and Hadith. PAI teachers are expected to improve their technical competence in statistical evaluation to produce objective and accountable learning assessments.]]>
