Data quality is essential for reproducible and reusable research—but how can we efficiently process unstructured, erroneous, or heterogeneous datasets? This workshop introduces OpenRefine—a powerful open-source tool for cleaning and transforming data that is particularly valuable in the GLAM sector (Galleries, Libraries, Archives, Museums) and in university collections. In these contexts, where heterogeneous and historically accumulated datasets in tabular form often contain inconsistent metadata, OpenRefine offers practical solutions: from standardizing spellings to correcting person attributions, URLs, or measurements, all the way to matching data with external sources such as GND, Wikidata, or OpenStreetMap.

This workshop provides an introduction to the basic functions of OpenRefine and uses concrete examples to demonstrate how researchers can clean, structure, and enrich tabular datasets. It also covers scripted processing steps as well as the import and export of various data formats (CSV, XML, JSON). OpenRefine not only provides a quick overview of large datasets, but is particularly helpful for heterogeneous and historically accumulated datasets, which often contain inconsistent metadata.

The goal of the workshop is to teach best practices in OpenRefine for working with research data and to promote interdisciplinary discussion of challenges and solutions. Through interactive exercises and discussions of specific real-world use cases shared by participants, attendees will collaboratively develop workflows and solutions.

The workshop is aimed at researchers from all disciplines who work with metadata or heterogeneous datasets. No prior knowledge is required—the basics will be covered in the workshop.

Preparing for the Workshop

Please install the OpenRefine software on your computer in advance (https://openrefine.org/download)  

Target audience: Researchers, data managers, and anyone involved in the processing of research data. Participation is limited to members of the Berlin University Alliance.

Instructor: Louise Tharandt (Humboldt-Universität zu Berlin, Koordinierungsstelle für wissenschaftliche Universitätssammlungen; Verbundprojekt SODa – Sammlungen, Objekte, Datenkompetenzen)

Workshop language: German

Registration until 09.10.2026: Registration form

This workshop is part of the Research Data Week @ BUA 2026.