Kalendarium

MMTC's Data Methods Initiative seminar presented by Fabienne Lind (University of Vienna)

MMTC's Data Methods Initiative seminar presented by Fabienne Lind University of Vienna

Multilingual computational text analysis 

Humans communicate in many languages, and our digital traces reflect this multilinguality. For social scientists studying multilingual societies, handling multilingual text data in valid, comparable ways is essential. This workshop introduces and operationalizes a framework for validating multilingual computational text analysis, drawing on Lind et al., (2025) https://doi.org/10.5117/CCR2025.1.13.LIND. We will translate the framework into practice, tackling real-world challenges across the research pipeline: data collection, model selection and evaluation strategies, and study design choices that affect cross-language and cross-context comparability. Drawing on examples from ongoing projects, we will discuss common pitfalls and effective strategies.

datamethodsinitiative.org