bkKey points :
1. To examine the influence of #news source bias on #affective content and #virality over time, we applied #sentiment analysis to a large #socialmedia dataset over a decade (i.e., ~ 30 M twitter posts from 2011-2020).
2. Biased news sources on both sides produced more arousing negative affective content, and this content was also most viral.
3. Biased news sources may generate " #AffectivePollution ”, engaging users at the expense of their knowledge, well-being, and harmony. (2/3)
Key points from "News source bias and sentiment on social media" :
1. To examine the influence of #news source bias on #affective content and #virality over time, we applied #sentiment analysis to a large social #media dataset over a decade (i.e., ~ 30 million twitter posts from 2011-2020).
2. Biased news sources on both sides produced more high arousal negative affective content, and this content was also most viral.
3. The virality of high arousal negative content increased over a decade, particularly for balanced news sources.
4. Biased news sources may generate " #AffectivePollution ” which engages users at the expense of their knowledge, well-being, and harmony. (2/3)
1. To examine the influence of #news source bias on #affective content and #virality over time, we applied #sentiment analysis to a large social #media dataset over a decade (i.e., ~ 30 million twitter posts from 2011-2020).
2. Biased news sources on both sides produced more high arousal negative affective content, and this content was also most viral.
3. The virality of high arousal negative content increased over a decade, particularly for balanced news sources.
4. Biased news sources may generate " #AffectivePollution ” which engages users at the expense of their knowledge, well-being, and harmony. (2/3)