Paper Type
ERF
Abstract
Our moral values serve as a compass, delineating the difference between right and wrong. Interested in the moral expressions in news content, we measured and compared moral expressions in fake versus real news content in both political and celebrity news. Specifically, we utilized vector space modeling and bag-of-words methodology to quantify the expression of each moral dimension in the news content, and applied Mann-Whitney U test in the comparison of the moral expressions in two types of news content. Five morality dictionaries developed based on morality foundation theory were leveraged in our analysis, allowing us to assess the consistency and alignment of captured results across different corpora. Our findings showed that fake news tend to use less positive tone when addressing loyalty and care values. When contrasting political and entertainment fake news, entertainment articles more frequently discuss virtues, whereas political narratives are skewed towards vices, with the exception of authority, which is dominantly featured both in fake news and its political articles.
Paper Number
1682
Recommended Citation
Mitiaeva, Marina and Xiao, Lu, "Understanding Morality of Misinformation through Analysis of Fake News" (2024). AMCIS 2024 Proceedings. 2.
https://aisel.aisnet.org/amcis2024/ict_global/ict_global/2
Understanding Morality of Misinformation through Analysis of Fake News
Our moral values serve as a compass, delineating the difference between right and wrong. Interested in the moral expressions in news content, we measured and compared moral expressions in fake versus real news content in both political and celebrity news. Specifically, we utilized vector space modeling and bag-of-words methodology to quantify the expression of each moral dimension in the news content, and applied Mann-Whitney U test in the comparison of the moral expressions in two types of news content. Five morality dictionaries developed based on morality foundation theory were leveraged in our analysis, allowing us to assess the consistency and alignment of captured results across different corpora. Our findings showed that fake news tend to use less positive tone when addressing loyalty and care values. When contrasting political and entertainment fake news, entertainment articles more frequently discuss virtues, whereas political narratives are skewed towards vices, with the exception of authority, which is dominantly featured both in fake news and its political articles.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIGGLOBDEV