Replication Data for “Language Shift in Polish and German Economic Journals: A Corpus-Based Comparative Analysis”

This replication package contains the datasets and analysis code used in the paper “Language Shift in Polish and German Economic Journals: A Corpus-Based Comparative Analysis.”

The package consists of two ZIP archives:

• Data.zip contains the datasets for the Polish and German journal corpora. Corpora containing full-text articles will be made available upon request. The files include the article-level data and numerical results used in the paper, including information on publication year, publication language, journal, document identifiers, and the variables used in the statistical and text analyses.
• Code.zip contains the scripts used to conduct the empirical analyses. The code includes the estimation of the Structural Topic Models (STMs), the selection of the number of topics, the analysis of topic prevalence and topic content, the comparison of topics across the two country-specific corpora, and the additional analyses and visualizations reported in the paper.

Together, the two archives provide the data and code necessary to reproduce the main analyses, results, tables, and figures presented in the paper. The package is intended for research transparency and replication purposes. Further information on the required software, R packages, file structure, and the order in which the scripts should be executed is provided in the documentation included with the replication files.