Replication Data for: Investigating positive/negative bias in Canadian newspapers through translation: A study of “confidence” in a corpus of business news

This data was used in our article Investigating positive/negative bias in Canadian newspapers through translation. To conduct our research, we used a subset of the Canadian Press Corpus in Finance (CAPCOF), composed of news items covering the 2007-2008 financial crisis and the years that led up to it (2001-2006). CAPCOF is a bilingual comparable corpus, containing texts in English and in French. The 2008 CAPCOF data subset contains 1,357,088 words in French and 1,403,907 words in English. The present Excel file was obtained using WordSmith 8.0 concordancer tool (Scott, 2020) and the 2008 CAPCOF data subset, and extracting occurrences of “confidence” and “confiance”.