Dataset on French to Dutch novice translation with focus on compounding

This data set was used to examine how novice translators handle Dutch word formation, specifically compounding, when translating from French. The data in this study were collected through an (online) translation task. For the translation task, 30 French source texts were selected from the Multilingual Traditional, Immersion and Native Corpus (MulTINCo) (Meunier et al., 2023). The texts were produced by L1-French eighteen-year-olds, who were asked by the MulTINCo compilers to write a short email, in French, to a friend on one of two topics, either a recent vacation or a birthday party. The 30 texts collected for the current study were selected based on their having a high number of potential target compounds when translated to Dutch. In some cases, these texts were edited, namely shortened, for the translation task, so as to keep the timing of the task relatively consistent across the texts. The translation corpus for this study was collected in 2024 and consists of French-to-Dutch translations by novice translators, namely bachelor’s and master’s students from translation programmes at five Belgian universities, viz. the Universities of Antwerp (Universiteit Antwerpen), Ghent (UGent), Brussels (VUB), Liège (ULiège), and Mons (UMons). In total, there were 71 participants, of whom 40 were L1-Dutch speakers (of which four were bilingual Dutch-French speakers) and 31 were L1-French speakers. Participants performed the translation task in LimeSurvey, an online survey tool, which allocated one of the 30 French source texts randomly to the participants. They were asked not to use dictionaries or translation tools, drawing only on their own knowledge for the entirety of the task. This file contains the source sentences and their translation, as well as the annotation of these translations.