DEVELOPING THE MAQASID-ORIENTED DISCLOSURE INDEX (MOSDI): A PYTHON-BASED TEXT MINING APPROACH FOR EVALUATING ZISWAF INSTITUTIONS IN INDONESIA

Abstract

Research on maqasid al-shariah has primarily focused on Islamic banking, while studies on maqasid-oriented disclosure in Islamic social finance institutions, particularly Zakat, Infaq, Sadaqah, and Waqf (ZISWAF) institutions, remain limited. Existing studies also predominantly rely on manual content analysis, which is subjective, time-consuming, and difficult to replicate. This study develops the Maqasid-Oriented Disclosure Index (MOSDI) as a disclosure measurement instrument for evaluating maqasid-oriented reporting in Indonesian ZISWAF institutions using Python-based text mining. A descriptive quantitative approach was employed to analyze 24 annual reports published by four major Indonesian ZISWAF institutions–BAZNAS, Dompet Dhuafa, Rumah Zakat, and LAZISMU–between 2019 and 2024. The analysis involved text preprocessing, dictionary-based keyword matching across the five dimensions of maqasid al-shariah, and the calculation of normalized disclosure scores. The findings reveal that maqasid-oriented disclosure varies across institutions and dimensions. Hifz al-nasl is the most extensively disclosed dimension, whereas hifz al-din and hifz al-mal receive comparatively less attention. Furthermore, Rumah Zakat and Dompet Dhuafa demonstrate higher levels of maqasid-oriented disclosure than BAZNAS and LAZISMU, indicating differences in the implementation and communication of maqasid values through annual reporting. This study contributes to the Islamic social finance literature by introducing MOSDI as a maqasid-oriented disclosure index specifically designed for ZISWAF institutions and by demonstrating that Python-based text mining provides a systematic, objective, consistent, and replicable approach for measuring maqasid-oriented disclosure.