The common understanding of metadata in surveys considers them as part of the documentation, which is usually prepared ex post, after the survey was conducted. Metadata standards like DDI, however, propose a new perspective on metadata, where metadata support all parts of the data lifecycle and can be re-used in subsequent iterations. Metadata-driven processes are of particular interest for panel studies, which are, by definition, designed to re-use instruments and other material across multiple waves of data collection.

The Generic Longitudinal Business Process Model (GLBPM) provides a generic perspective on processes in longitudinal studies in general and panel studies in particular. Starting with the GLBPM, we can identify digital objects in the process of conducting and managing panel studies. We take a closer look at those digital objects and how to re-use them in subsequent waves. The presentation combines a theoretical discussion of the process and its digital objects with the practical experience from the German Socio-Economic Panel Study (SOEP) and the implementation of DDI on Rails—a web-based documentation system for panel studies, utilizing the DDI standard, and currently in production on paneldata.org.

In this presentation we will take a closer look at two specific aspects of metadata in panel studies: the re-use of instruments and linking data on the variable level. Instruments might be the most complex digital objects in panel studies because we have to consider various aspects like the logical structure, layout information, different modes and control constructs. After that, we take a closer look at concepts as a generic design to link both questions and variables over time and across studies.