Managing Research Data with DDI-L: Supporting Interoperability Between Multiple Systems

Research data rarely consists of a single physical file containing all the needed metadata and data in a compact unit. Normally there are multiple data files, documentation, code lists, and related items. In short, each study is a mini-collection of material. Traditionally, library, archive, and data systems have managed their information in slightly different ways. Libraries create one record for one physical object. Archives have a more collection centric approach expressed as fonds, series, files, and items. DDI like many metadata specifications incorporates the object’s ““discovery metadata”“ record within its overall structure creating essentially an object that serves as an extended record. But is this the most effective way to manage diverse and highly interconnected collections?The Minnesota Population Center is in the process of organizing our metadata in a way that will allow us to effectively interact with a number of discovery and repository systems around the world. The goal is to use DDI-L as a conduit for linking the contents of our data delivery systems, archive metadata, and related holdings to external systems (e.g. DataONE, da|ra, World Bank) that use a wide variety of standards and formats. This case study reflects our work to date to achieve this goal.