Abstract

This study proposes a modular design framework for records management metadata aimed at ensuring the long-term preservation and reuse of research data, with particular attention to the distinct characteristics of different research domains. Existing metadata frameworks have primarily been developed with a focus on the science and technology fields, often failing to adequately reflect the unique requirements of research in the humanities and social sciences. To address this limitation, this study analyzed research reports and data management guidelines from four representative institutions across these domains. By comparing common elements with field-specific components, the study identified key differences in approaches to ensuring research reproducibility. The analysis employed six criteria—data type, documentation of research conditions, tools and software, research procedures, mandated requirements, and preservation and accessibility measures—drawn from existing metadata standards and prior studies. Based on this, a modular metadata element system was developed, allowing for selective expansion of field-specific elements while maintaining a common core structure. This flexible design accommodates the diversity of research data and varying reproducibility requirements. Through a pilot case analysis, the study demonstrates the versatility and scalability of the proposed framework, offering a foundation for future research across diverse disciplines and data types.