News media in crisis: a sentiment and emotion analysis of US news articles on unemployment in the COVID-19 pandemic

This study, integrating sentiment, emotion, discourse, and timeline analyses together, conducts a corpus-based sentiment analysis of the news articles on unemployment from the<i> New York Times</i> in 2020, to capture the emotional dynamics conveyed by the newspaper as the pandemic-induced unemployment developed in the US. The results reveal that positive sentiment in the news articles on unemployment is significantly higher than negative sentiment. In emotion analysis, “trust” and “anticipation”rank the first and second among the eight emotions, while “fear”and “sadness” top the negative emotions. Complemented with a discourse analysis approach, the study reveals that the change of the sentiments and emotions over time is linked with the evolution of the pandemic and unemployment, the policy response as well as the protests against ethnic inequalities. This study highlights the important role mainstream news media play in information dissemination and solution-focused reportage at the time of severe crisis.This dataset contains 14 documents for the data of 2 sentiments and 8 emotions, generated by Python. It includes NRC lexicon categories for the sentiments and emotions in the study (data1-10), top 10 high-frequency words associated to the sentiments and emotions in each of the 12 subcorpora (data11-12), and monthly values of the sentiments and emotions in 2020 (data 13-14).