Crowdsourcing is a current trend in scholarly research. Building on digital technology and participatory methods, researchers are using crowdsourcing to collect, process, and visualize data from various domains, including historical documents, water quality, or public signage. In linguistics, crowdsourcing has seen a massive growth of interest in the last years, especially in projects that make use of mobile research applications to collect an analyze language variation and change. For example, such apps have been used to document regional variation in English and Frisian, voice quality and speech rate in Swiss German, or multilingualism in public signage. The use of crowdsourcing techniques brings about new possibilities for linguistic research in combining the participatory collection and computational processing of data. Some concerns have been raised, however, in respect of the composition, quality, and value of such data collections for academic research.

Against the backdrop of this, the talk will discuss experiences from two research projects hosted at the University of Luxembourg. One project focuses on the diversity of written language in the public sphere (Lingscape), the other project (Schnëssen) documents variation in present-day spoken Luxembourgish. Both projects are app-based and build on public participation in project work. The talk will briefly introduce the projects and their workflows, present some exemplary results, and discuss the potential and pitfalls of crowdsourcing for linguistic research, including the implementation of mobile research apps in educational settings.